Emacs inferior to XEmacs?
Mike Humski
I am an emacs user. But I am going to try xemacs and read NEWS file at
the xemacs www site. It lists a lot of points in which xemacs is superior
to emacs.
Can I take them straight?
Is emacs inferior to xemacs?
If so, why emacs is still around. Why doesn't every one switch to xemacs?
It seems there must be something in which emacs is better at than xemacs.
Otherwise, emacs would have perished.
If you know both, please let me know in which area emacs is superior to
xemacs.
Mike
Bryan O'Sullivan
m> Can I take them straight? Is emacs inferior to xemacs?
It depends on what you want to do. GNU Emacs is much less bloated,
and faster, than XEmacs, but it has a much less "finished" look to it.
On the other hand, it doesn't handle graphics, nor is it integrated
with Sun's development environment nearly as nicely as XEmacs. On the
third hand, XEmacs's support for character terminals totally bites, so
GNU Emacs is still the best choice for distance editing.
m> It seems there must be something in which emacs is better at than
m> xemacs. Otherwise, emacs would have perished.
Although the differences between GNU Emacs and XEmacs are many, the
actual differences in functionality are sufficiently small that the
choice is more a matter of preference and religion than clear
technical superiority.
For what it's worth, I use both GNU Emacs and XEmacs all the time.
<b
--
Let us pray:
What a Great System.
Please Do Not Crash.
^G^IP@P6
Hrvoje Niksic
> Otherwise, emacs would have perished.
GNU Emacs (maintained by the FSF) is inferior to XEmacs (maintained by
the XEmacs Advocacy Group) in *some* aspects. In other, GNU Emacs is
superior.
XEmacs has a much better integration with the X Window system, is in
some aspects nicer to work with, especially for beginners. It has
toolbars and horizontal scrollbars, it can display proportional fonts,
comes with more bundled packages, and has a host of other features.
Furthermore, it can display arbitrary graphics within its buffers and
play various sounds, which makes it quite ahead of GNU Emacs in many
fields of usage.
However, GNU Emacs is at the moment faster than XEmacs. It has a
larger base of users, and more packages work on it unchanged than on
XEmacs. Furthermore, Emacs can be used on a larger base of systems,
like MS-DOS or Windows, where the current XEmacs has not yet been
ported. It is sponsored and developed by the FSF, which can be a
reason for itself to be used by many.
There are philosofical differences between approaches to Lisp
implementations. Whereas XEmacs developers advocate data abstraction
(like specialized types for keymaps etc.), GNU Emacs developers are
against it.
As always, I tend to think that the most important thing is getting
used to something. I began with XEmacs, got used to it, and do not
intend to change. Yes, I like speed of GNU Emacs, but then again GNU
Emacs users like the graphics they see in my buffers. All of it works
both ways, and people have strong feelings in these matters.
> If you know both, please let me know in which area emacs is superior to
> xemacs.
I hope this helped. And yes, I did try to be objective and I *do*
hope that this won't bring about another flame war. But then again,
it's been quite some time from the last one.... ;-)
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
`VI' - An editor used by those heretics that don't subscribe to
the Emacs religion.
Erik Naggum
* Mike Humski <m...@rahul.net>
| I am an emacs user. But I am going to try xemacs and read NEWS file at
| the xemacs www site. It lists a lot of points in which xemacs is
| superior to emacs.
of course it does. it's called "marketing".
| Can I take them straight?
as much as you can take any marketing straight. (i.e., a resounding "NO".)
| Is emacs inferior to xemacs?
no.
| It seems there must be something in which emacs is better at than xemacs.
| Otherwise, emacs would have perished.
precisely, and since Emacs has not perished, and is even more unlikely to
perish than XEmacs, something must be wrong. you have just learned lesson
#1 in detecting fraudulent marketing: assume that the premises and
arguments are true, then see if the conclusions that follow from them fall
flat on their face. if they do, you have fraudulent marketing.
XEmacs has one forte: it is better suited to those who have no inclination
to adapt their working environments to their own needs, and who instead go
for the prepackaged stuff they get from the vendor. that's it.
| If you know both, please let me know in which area emacs is superior to
| xemacs.
well, since you ask. Emacs is faster, is supported on more platforms, has
_far_ fewer bugs, uses _much_ less memory, and requires about half the X
protocol data of XEmacs. Emacs is more conservative in its adoption of new
packages, since not all packages are fit for distribution as soon as they
are written and it is also obvious that if a package is distributed while
immature, it will hang around for a very long time in that immature state,
even if it is upgraded centrally. this is a major problem for "first to
market at all costs"-type marketing strategies. Emacs tries above all to
be stable.
the primary purpose of Emacs packages is of course to be used, but a strong
secondary purpose is to be read and understood. the whole idea behind free
software is to enable people to understand, improve, modify the code. to
do so requires a code base with a certain consistency in style and which
does not reimplement the same ideas all over the place. all good writers
read before they write, and read much more than they write, but programmers
are expected to write before they read, and reading other people's code is
not exactly cherished. obviously, this is because code written by people
who had to invent every issue of style on their own _sucks_. this is often
the case for C, Perl, and the like, but not usually the case for Lisp,
which has rich traditions to draw on. XEmacs is more drawn to the C style
of programming, where the Lisp is merely an extension language, and all the
real work is done at the C level. Emacs is more drawn to the Lisp style of
programming, where Lisp is the real language, and C is just necessary to
implement the Lisp system and to optimize some often-used processes, such
as the redisplay engine. this difference permeates many aspects of each of
the Emacsen.
Emacs' Lisp is also simpler. where Emacs has 10 functions and use existing
types and access functions, XEmacs has 50 functions and use an entirely new
datatype. where Emacs tries not to do too many things that break working
code, XEmacs doesn't care at all, claiming instead that the way it was done
in the past was "braindamaged" and other favorite terms of respect, clearly
implying that should somebody improve on _their_ design, they will happily
accept similar treatment of their now "superior" solutions.
a lot of packages are developed either for XEmacs or for Emacs, because it
is often too hard to do both. new, immature, fancy packages are usually
developed for XEmacs. robust packages are usually developed for Emacs.
the question you should ask if you want to use XEmacs is "does it have the
latest fanciest stuff" and the question you should ask if you want to use
Emacs is "can I trust it never to lose my work".
in almost all respects, XEmacs gives more to the impatient, hands-on, "new"
generation which doesn't have time to read manuals, while Emacs gives more
to the experienced and dedicated users who read manuals to learn and
understand. there are still at least two orders of magnitude more of the
latter type, but they don't make the news for the same reason that it is
not news if a 747 takes off from JFK and flies to LAX, but it would be if
an elephant did. very seldom, "new" means "improved". "new" usually means
"has to do all the old mistakes over again".
that said, there are real improvements in XEmacs, too. however, the only
way they think they can communicate this to the world is through the use of
state-of-the-art marketing practices. one must accept a certain amount of
guilt by association merely by choosing to employ techniques chosen by used
car and computer salesmen, presidential campaigns, etc, and one must also
expect to have to overstate everything one says by at least a factor of two
because all listeners have been cynicized by the other uses of those sales
techniques to divide by at least two automatically. so, if XEmacs says
they are superior to Emacs, you must expect them to compare the latest
development version to Emacs 19.28, that they found and fixed a bug in both
Emacsen, but neglected to report it to the Emacs maintainers until after
they could gloat about how braindamaged it is, etc. when somebody starts
up the "I'm better than you" machine, you can be pretty certain they are
dishonest creeps. they need to tell they are superior because you would
never realize it if it weren't for their "helpful suggestions" about what
you should compare. naturally, the real improvements in XEmacs are not in
the excessively hyped parts.
#\Erik
--
Please address private replies to "erik". Mail to "nobody" is discarded.
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> protocol data of XEmacs. Emacs is more conservative in its adoption of new
> packages, since not all packages are fit for distribution as soon as they
> are written and it is also obvious that if a package is distributed while
> immature, it will hang around for a very long time in that immature state,
> even if it is upgraded centrally. this is a major problem for "first to
> market at all costs"-type marketing strategies. Emacs tries above all to
> be stable.
You follow the very same type of marketing strategies you have
attacked several lines before. Give us several examples of "immature"
lisp packages in XEmacs.
Give us examples of all those XEmacs bugs. Real examples, not "every
version of XEmacs has crashed on me". How come you are the only one
to whom XEmacs is only crashing? Why doesn't it happen to Barry
Warsaw or Bill Perry? Bah!
[...]
> that said, there are real improvements in XEmacs, too.
Are we seeing a miracle happen? It is fun noticing how much time Erik
spends flaming XEmacs, and how little he devotes to those
improvements.
Erik Naggum
* Hrvoje Niksic <hni...@srce.hr>
| Are we seeing a miracle happen? It is fun noticing how much time Erik
| spends flaming XEmacs, and how little he devotes to those improvements.
could you provide me with a summary of how much time I spend flaming
XEmacs, please? I don't keep track of such things, but I have spent less
than two hours today (and none in several months) on countering the
marketing from the XEmacs camp that makes an innocent bystander think that
Emacs would "perish" were all of it true. obviously, the XEmacs marketing
is a bunch of vile lies, and that needs to be said. equally obviously,
XEmacs folks are very touchy in this area. people's whose bluffs are
exposed usually are quite touchy about it.
incidentally, I don't keep track of how much time XEmacs proponents spend
flaming Emacs, either. do you have the latest data, Hrvoje? you might
find it instructive if you don't already keep records. you might also find
my summary of keybindings in the XEmacs FAQ a tad bit less inflammatory
than the previous version, which, if I recall correctly, referred to the
Emacs way as "braindamaged" several times in the course of a few lines.
I'm sure you were happier with the previous version, but please, let it
stay the way I wrote it. I guess you also think it's a miracle that I
would improve something related to XEmacs in such a was as to make XEmacs
less hostile to Emacs, considering the bile that XEmacs proponents have
emptied over Emacs over the years.
oh, speaking of which, XEmacs folks talk about Emacs all the time, they
compare themselves to Emacs, they hate Emacs, they flame Emacs, they tell
Emacs users how wonderful things are in XEmacs. I find this a bit tiring,
especially since _every_damn_time_ some Emacs user tries to level the
balance by countering their exaggerated lies, all sorts of whining losers
creep out of the woodwork to complain how unfair somebody is to XEmacs. is
XEmacs really so bad that it needs this kind of defense?
I sincerely wish that XEmacs folks just grew up and start to do their own
job well, instead of comparing themselves to others. anybody can be better
than somebody else if they only portray those others as sufficiently bad.
(miracles happen only to people who don't understand what is going on.)
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> could you provide me with a summary of how much time I spend flaming
> XEmacs, please? I don't keep track of such things, but I have spent less
Neither do I. But it seems a lot. Your way is "I don't flame XEmacs,
it's just that I have this terrible urge to intervene in every
discussion about XEmacs about how crappy it really is".
> incidentally, I don't keep track of how much time XEmacs proponents spend
> flaming Emacs, either. do you have the latest data, Hrvoje? you might
I read comp.emacs and comp.emacs.xemacs. I have never seen people
flame GNU Emacs the way you flame XEmacs. The small number of
examples (Ben's heated words several months ago) have mostly been in
answer to your, equally heated, postings.
> my summary of keybindings in the XEmacs FAQ a tad bit less inflammatory
> than the previous version, which, if I recall correctly, referred to the
> Emacs way as "braindamaged" several times in the course of a few lines.
> I'm sure you were happier with the previous version, but please, let it
Nope. I have suggested a correct approach to the FAQ maintainer
myself, and I am glad that XEmacs FAQ now (AFAIK) contains no "bad"
material against GNU Emacs. You have obviously misjudged my opinion
upon the matter.
> stay the way I wrote it. I guess you also think it's a miracle that I
> would improve something related to XEmacs in such a was as to make XEmacs
> less hostile to Emacs, considering the bile that XEmacs proponents have
> emptied over Emacs over the years.
Of all the hostilities on the net, I your articles seem to be the most
hostile. Of course, you say that you need to say the truth. OK, but
then Ben has the same right to call some things in GNU Emacs
braindamaged -- why not, it's just the truth, after all?
I think each side should be more careful of the tone it assumes, if it
does not want to provoke similar reactions from the other one.
> oh, speaking of which, XEmacs folks talk about Emacs all the time, they
> compare themselves to Emacs, they hate Emacs, they flame Emacs, they tell
> Emacs users how wonderful things are in XEmacs. I find this a bit tiring,
It is perfectly normal to compare the two major branches of Emacs 19
that are being actively developed.
> especially since _every_damn_time_ some Emacs user tries to level the
> balance by countering their exaggerated lies, all sorts of whining losers
> creep out of the woodwork to complain how unfair somebody is to XEmacs. is
> XEmacs really so bad that it needs this kind of defense?
I don't think so. You are not unfair to XEmacs -- you just follow the
same marketing principles you argue against -- attacking the "enemy"'s
weaknesses, and hyping your own advantages.
> I sincerely wish that XEmacs folks just grew up and start to do their own
> job well, instead of comparing themselves to others. anybody can be better
> than somebody else if they only portray those others as sufficiently
> bad.
> (miracles happen only to people who don't understand what is going on.)
I think that what you sincerely wish has already happened. But maybe
you are the one who doesn't understand what's going on...
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
"What is the sound of Perl? Is it not the sound of a wall that
people have stopped banging their heads against?" -- Larry Wall
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> * Mike Humski <m...@rahul.net>
> | I am an emacs user. But I am going to try xemacs and read NEWS file at
> | the xemacs www site. It lists a lot of points in which xemacs is
> | superior to emacs.
>
> of course it does. it's called "marketing".
>
> | Can I take them straight?
>
> as much as you can take any marketing straight. (i.e., a resounding "NO".)
Not quite, I'm afraid. All of the listed points are valid.
> | Is emacs inferior to xemacs?
>
> no.
Can't compare, really. For everyday use, I find GNU Emacs for
convenient (stability, speed). If you have other needs than merely
programming, more in the line of text processing, or WEB browsing,
there is no alternative to XEmacs which supports proportional fonts
and inline graphics. Apart from these essentials (*if* you are into
that area), XEmacs offers quite a few other niceties.
> | It seems there must be something in which emacs is better at than xemacs.
> | Otherwise, emacs would have perished.
At quite a lot of which it supports.
> precisely, and since Emacs has not perished, and is even more unlikely to
> perish than XEmacs, something must be wrong. you have just learned lesson
> #1 in detecting fraudulent marketing: assume that the premises and
> arguments are true, then see if the conclusions that follow from them fall
> flat on their face. if they do, you have fraudulent marketing.
With XEmacs, I don't find their marketing fraudulent. Point out the
lies in it.
> Emacs' Lisp is also simpler. where Emacs has 10 functions and use existing
> types and access functions, XEmacs has 50 functions and use an entirely new
> datatype. where Emacs tries not to do too many things that break working
> code, XEmacs doesn't care at all, claiming instead that the way it was done
> in the past was "braindamaged" and other favorite terms of respect, clearly
> implying that should somebody improve on _their_ design, they will happily
> accept similar treatment of their now "superior" solutions.
Care for examples? In fact XEmacs maintainers have a hard time
keeping compatibility with GNU Emacs development where quite a few
things change they way they were done. After some "new" functionality
(which was present in XEmacs before) gets reinvented in GNU Emacs, it
typically takes one or two releases to get an interface wrapper in
upcoming XEmacses. A nuisance, of course.
> a lot of packages are developed either for XEmacs or for Emacs, because it
> is often too hard to do both. new, immature, fancy packages are usually
> developed for XEmacs. robust packages are usually developed for
> Emacs.
Don't be silly. The quality of the packages has nothing to do with
for what you develop them. The reason "fancy" packages are more often
developed for XEmacs is simply that you have the display stuff to be
"fancy".
> in almost all respects, XEmacs gives more to the impatient, hands-on, "new"
> generation which doesn't have time to read manuals, while Emacs gives more
> to the experienced and dedicated users who read manuals to learn and
> understand.
Well, the hypertext apropos mechanisms of XEmacs might make it easier
to get to the information you want, but that does not make them
inferior to GNU Emacs, IMO. Having easily accesible features is no
mistake in itself. You can still use traditional key sequences if you
want.
> that said, there are real improvements in XEmacs, too. however, the only
> way they think they can communicate this to the world is through the use of
> state-of-the-art marketing practices. one must accept a certain amount of
> guilt by association merely by choosing to employ techniques chosen by used
> car and computer salesmen, presidential campaigns, etc, and one must also
> expect to have to overstate everything one says by at least a factor of two
> because all listeners have been cynicized by the other uses of those sales
> techniques to divide by at least two automatically.
I have compared both and found that the claims of XEmacs people are
correct. I use GNU Emacs for everyday work because of speed reasons,
mainly.
--
David Kastrup Phone: +49-234-700-5570
Email: d...@neuroinformatik.ruhr-uni-bochum.de Fax: +49-234-709-4209
Institut fuer Neuroinformatik, Universitaetsstr. 150, 44780 Bochum, Germany
Erik Naggum
* Hrvoje Niksic <hni...@srce.hr>
| Neither do I. But it seems a lot.
please try to be objective, if you can. out of _many_ thousand news
articles a year, less than a dozen even _mentions_ XEmacs. I also spend
more than four thousand hours a year working with Emacs (it's the whole of
my working environment), upwards of a thousand of which are spent writing
new code or fixing bugs, in response to my own use or bug reports. I have
spent a few hours this year to respond to the typical XEmacs idiocy about
Emacs, and Hrvoje Niksic thinks "it seems a lot". are you utterly and
completely _insane_?
| Your way is "I don't flame XEmacs, it's just that I have this terrible
| urge to intervene in every discussion about XEmacs about how crappy it
| really is".
excellent! this will score you several points in the XEmacs crowd. never
mind that it utterly and completely false. never mind that this is the
kind of foul lie and idiotic exaggeration that virtually _defines_ how a
good XEmacs user should behave towards the Enemy. never mind that you are
trying to portray me as something I am not, have never been, and never will
be. never mind that "intervene in every discussion about XEmacs" is such a
silly exaggeration that even XEmacs users must be embarrassed.
I'm also saddened to see that it is so hard for an XEmacs user to
distinguish between the lies and exaggerations and fraudulent marketing of
XEmacs on the one hand and XEmacs itself on the other. XEmacs is as good
as any commercially hyped product, probably even better. it is the
despicable idiots who defend it, who lie about it, who do anything they can
to hurt Emacs and the Free Software Foundation that is crappy. people
who's motives are destruction, whose character is of the same kind as Bill
Gates', who also seeks to destroy his competition instead of being better
than them on his own terms, the kind of scum who fights to win at all
costs, no matter how large, not to improve or create anything.
| Of all the hostilities on the net, I your articles seem to be the most
| hostile.
wow! this must be a ten-point exaggeration with the XEmacs crowd. what do
you win if you become the most exaggerating XEmacs user? do you get a
nice, black T-shirt? does the "X" in "XEmacs" stand for "eXaggerated".
you know, if you could try to calm down a little, you would see how
completely idiotic that statement of yours is. it's so mindbogglingly
stupid I seriously wonder how anybody could manage to write it. but then
again, I'm not an XEmacs user, and I don't use the word "braindamaged" and
"truth" in the same sentence:
| OK, but then Ben has the same right to call some things in GNU Emacs
| braindamaged -- why not, it's just the truth, after all?
you don't understand the word "braindamaged". look it up.
you might find it instructive to read what I _actually_ write, instead of
inventing the crap you wish I had written. I have never said any of the
utterly stupid things you attribute to me.
just quit lying, and there will no longer be a need to correct your lies.
it's as simple as that.
Thomas A. Horsley
Just to toss in my random 2 cents, I use emacs rather than Xemacs primarily
because I already had it before Xemacs showed up, and because even emacs has
too many features. Most of my .emacs file is dedicated to turning off and
utterly squashing all the thousands (sometimes it seems *billions*) of
"helpful" features that are configured by default. If I used Xemacs, I'd
just have a lot more stuff to disable :-).
Now if someone would produce an emacs that does realtime garbage collection
and supports multi-threaded lisp so things like GNUS could be doing hairy
header processing in one buffer while I can still edit a file in another
buffer, now *that's* a feature that would get me to switch emacsen!
--
See <URL:http://ourworld.compuserve.com/homepages/TomHorsley> for
information on Government by Performance
Richard Cognot
>>>>> "Hrv" == Hrvoje Niksic <hni...@srce.hr> writes:
Hrv> Erik Naggum (nob...@naggum.no) wrote:
>> could you provide me with a summary of how much time I spend
>> flaming XEmacs, please? I don't keep track of such things, but
>> I have spent less
Hrv> Neither do I. But it seems a lot. Your way is "I don't
Hrv> flame XEmacs, it's just that I have this terrible urge to
Hrv> intervene in every discussion about XEmacs about how crappy
Hrv> it really is".
Erik just tend to have a finger on the "flame" key ready for any
thread where comparison between Emacs and XEmacs is involved.
Since the last one, he's been living happilly at the bottom of my kill
file (first and last one to ever get in there, hope you won't feel too
lonely Erik ;-) and mail from his site is also filtered out. I'm
feeling much better now that I don't have to read his insanities.
Anyway, more to the point of the initial poster: I think that when you
begin with Emacs, then XEmacs is *much* better, as it really takes you
by the hand and helps you out with what you want to do by providing
easy access to a lot of functionnalities. It basically provides easy
means of configuring your Emacs environment into something you feel
comfortable with, without *requiring* you to spend most of your time
digging out the features you want (now, if you *really* want to dig,
you still can ;-).
If you already are a "power user", I don't think there is that
much of a difference as for the text editing part. Most things can be
done in both Emacsen, albeit differently sometimes. As for other
functionnalities, such as embedded graphics/windows, integration with
X, etc..., XEmacs is far superior, IMO. The only tradeoff:
speed... But I'm quite confident: it will someday be history, either
because XEmacs will have been more optimized, or because the machine on
which I work will be fast enough I won't notice anymore ;-).
Richard
--
|-------------------------------------------------------------|
| Richard Cognot | Proceed, with fingers crossed... |
| | /\^^/\ |
| <cog...@ensg.u-nancy.fr> | `' |
|-------------------------------------------------------------|
| http://www.ensg.u-nancy.fr/~cognot |
|-------------------------------------------------------------|
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> * Hrvoje Niksic <hni...@srce.hr>
> | Neither do I. But it seems a lot.
> articles a year, less than a dozen even _mentions_ XEmacs. I also spend
Then how come I find that your participation in otherwise useful
discussions about XEmacs results in a flame war? Is it you, or is it
everyone else? But of course, *I* am the one who is insane.
> | Your way is "I don't flame XEmacs, it's just that I have this terrible
> | urge to intervene in every discussion about XEmacs about how crappy it
> | really is".
> excellent! this will score you several points in the XEmacs crowd. never
> mind that it utterly and completely false. never mind that this is the
> kind of foul lie and idiotic exaggeration that virtually _defines_ how a
> good XEmacs user should behave towards the Enemy. never mind that you are
> trying to portray me as something I am not, have never been, and never will
> be. never mind that "intervene in every discussion about XEmacs" is such a
> silly exaggeration that even XEmacs users must be embarrassed.
Why don't you just refute my judgments, instead of speaking of "XEmacs
users behaving towards the Enemy"? Your tone doesn't help the
discussion... When I asked you to list some of those terrible XEmacs
bugs and asked why they don't happen to other people, you never
answered.
> I'm also saddened to see that it is so hard for an XEmacs user to
> distinguish between the lies and exaggerations and fraudulent marketing of
> XEmacs on the one hand and XEmacs itself on the other. XEmacs is as good
> as any commercially hyped product, probably even better. it is the
> despicable idiots who defend it, who lie about it, who do anything they can
> to hurt Emacs and the Free Software Foundation that is crappy. people
> who's motives are destruction, whose character is of the same kind as Bill
> Gates', who also seeks to destroy his competition instead of being better
> than them on his own terms, the kind of scum who fights to win at all
> costs, no matter how large, not to improve or create anything.
Later in your article you request that I cool down. Judging by the
quoted paragraph, you should be the one to cool down. Defending
XEmacs is *not* the same as lying about it, hurting GNU Emacs and the
FSF, destroy, act like B.G., etc. Seeing your other articles about
Emacs, I consider you a highly intelligent and competent person (in
spite of people from the "crowd" warning me otherwise). I cannot
fathom how you can produce such rubbish as the paragraph above.
> | Of all the hostilities on the net, I your articles seem to be the most
> | hostile.
> wow! this must be a ten-point exaggeration with the XEmacs crowd.
I may have not been clear enough -- hostilities against XEmacs.
> you know, if you could try to calm down a little, you would see how
Ha -- calm down! This one is good. :)
> again, I'm not an XEmacs user, and I don't use the word "braindamaged" and
> "truth" in the same sentence:
> | OK, but then Ben has the same right to call some things in GNU Emacs
> | braindamaged -- why not, it's just the truth, after all?
> you don't understand the word "braindamaged". look it up.
Are you sure you understood the sentence (in its context)? I'll try
to explain, just in case you really did not understand. If your
opinion is that XEmacs is bad and evil, and you feel free to say all
you think about it (which you consider the truth) without the least
bit of tactfulness, then other people have the right to behave the
same way to GNU Emacs -- just say whatever they think about it, using
the ugly terms like "braindamaged". Get it?
> you might find it instructive to read what I _actually_ write, instead of
> inventing the crap you wish I had written. I have never said any of the
> utterly stupid things you attribute to me.
> just quit lying, and there will no longer be a need to correct your lies.
> it's as simple as that.
I have tried, *very* hard, to read what you actually write. I have
tried to find somewhere within it a person who only wishes to discuss
two pieces of software. But I haven't. All your posts concerning
XEmacs are full of general and personal insults to your
correspondents, and are not in the least helpful -- but you know it
quite well. I am sorry about that, but I cannot change it.
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Erik Naggum
* Hrvoje Niksic <hni...@srce.hr>
| Then how come I find that your participation in otherwise useful
| discussions about XEmacs results in a flame war?
because you (1) only remember it when you see a flame, and (2) only read
articles in comp.emacs.xemacs from me when they are crossposted from other
Emacs groups. you draw your conclusions on known insufficient grounds.
| I have tried, *very* hard, to read what you actually write. I have tried
| to find somewhere within it a person who only wishes to discuss two
| pieces of software. But I haven't. All your posts concerning XEmacs are
| full of general and personal insults to your correspondents, and are not
| in the least helpful -- but you know it quite well. I am sorry about
| that, but I cannot change it.
once upon a time, it must have been a little over two years ago, I tried to
find out why the hell the XEmacs crowd was so hostile to the Emacs crowd.
I came into this debate as a complete novice, I tried to understand, I
asked around, I irritated a lot of people by poking in the wrong places and
found a lot of sore spots, but almost all of them in the XEmacs crowd. I
was told that I was defending RMS, that I would get sick of it, just like
they had been, that RMS is impossible to work with (obviously due to a bad
filksong). I tried to learn, but what did I learn? I learned that the
XEmacs crowd is obsessed with denouncing Emacs, RMS, and _anybody_ who
worked or works on Emacs and Emacs design and who hasn't seen "the light".
I learned that the Emacs crowd behaves as if XEmacs does not exist, mostly
because the XEmacs crowd behaves like a bunch of idiots when it comes to
discussing Emacs. well, I didn't like what I saw.
I have looked at XEmacs seriously for about three weeks (full time), and I
have read the XEmacs Lisp Reference Manual. I have very serious doubts
about hundreds of design decisions in XEmacs, in particular the misguided
data abstraction they engage in. (if it were such a good idea, why aren't
there any `defstruct' or `defclass' in XEmacs Lisp, only at the C level?
obviously, it is not a good idea at all -- the only point in hiding stuff
in C is that Ben Wing and others didn't want bonehead programmers to fuck
with internals.) if encapsulation is such a grand objective, why haven't
XEmacs Lisp implemented Common Lisp's package-concept or something like it?
all they have done is reorganize the directory structure and call the flat
structure in Emacs "braindamaged" or worse. XEmacs is a C program with a
very complex Lisp extension language, whereas Emacs is a Lisp program with
a C stratum. XEmacs Lisp has hundreds more functions implemented in C than
Emacs Lisp has, and most of those are there to access data structures.
again, if data abstraction is such a good idea, why doesn't XEmacs Lisp
uniformly use `setf' forms instead of creating _tons_ of setter functions?
if the design in Emacs was so bad, why the hell have they only improved on
minor things in it? there are _lots_ of ways to _really_ improve Emacs
Lisp, which any good Lisp programmer can enumerate standing on one foot! I
found _none_ of them in XEmacs. all I found was bitching and moaning about
how keymaps were lists and vectors in Emacs and a "real" abstract data type
in XEmacs. now, what difference does _that_ make? who _cares_?
unlike most of the vocal XEmacs folks, I respect the fact that GNU Emacs is
RMS's brainchild. XEmacs was created because of differences in attitude to
a lot of things. the discussions (which I read in full when I was looking
for an explanation for the hostilities), showed me two sides who were
equally stubborn and at times childish, but I found that RMS was able to
spell out _why_ he thought the way he did, and the renegades were not. I
found an idea about a grand design (larger than Emacs alone) at work in
RMS' messages where I found discontent, rebelliousness and impatience in
the renegades'. I _also_ found a desire to modernize the user interface in
the renegade camp, but the only real difference was the speed at which it
would be done, one side seeking to minimize the ramifications involved, the
other careless about such consequences. I have had discussions with a few
people in the XEmacs camp who are exceptionally bright, I have talked with
a few good XEmacs people on the phone, but every time I see a discussion of
Emacs in the XEmacs camp, it's always that gloating "we're so superior,
you're so inferior" that armies and politicians engage in when they know
they are about to lose. I wondered, what is it that the XEmacs crowd is
afraid of? what is it that they fear will happen to them? then I realized
it. there is room for two cooperating Emacsen, but not for two competing
ones, and they have chosen to compete. I don't think they'll survive, and
they fear they won't. they have taken free software, grafted on a lot of
improvements, some solid, some experimental, and through either ill will or
incompetence made it harder to integrate the improvements back into the
host, which continues to develop at a fast pace. they haven't cared about
copyrights as much as they should have, but they have in consequence been
more welcome to people who only wanted to make some small improvements that
FSF's copyright policies made just that one bit too hard or inconvenient.
quite contrary to the goal of protecting code from destructive influences
from people who didn't know what they were doing, from looking at XEmacs
code, I found a wide variety of styles and much evidence of spliced-in
code. in Emacs, such is restricted to individual packages. in tracking
the source code for Emacs (I mirror it to my own system, and build new
Emacs versions to test and verify changes daily), I see many small
aesthetic improvements in the vicinity of bugs that were fixed. despite
what the XEmacs renegades say, there is a strong sense of _elegance_ in
Emacs, an elegance that I find totally lacking in XEmacs, which bears all
the evidence of a system whose primary goal is to make it _work_.
to me, free software is the idea that programmers should be able to read
source code like authors read other books before they write their own, like
playwrights go to the theater, like all other areas of creative exercises
in our society are based on learning from the masters. this means that the
masters must focus on one thing: never willingly prodArticle Unavailable
Martin Boyer
Mike Humski <m...@rahul.net> writes:
> If you know both, please let me know in which area emacs is superior
> to xemacs.
1. Emacs is faster, specially on character-based terminals.
2. Emacs starts much faster.
3. Emacs is smaller, but that isn't noticeable for most users.
4. I don't remember in which *year* Emacs last crashed on me. XEmacs
has crashed on me or a user here every week for versions 19.12,
19.13, and 19.14 (the latest version). I have been told this is
due to garbage-collection bugs, which are very difficult to debug.
5. The Emacs "core" is more stable. I have found, however, that since
the XEmacs maintainers include more packages in the "distributed"
realease, the actual XEmacs system as a whole is more stable. This
really depends on the usage YOU make of Emacs/XEmacs.
6. Emacs has a shorter "debugging cycle" (i.e. patches are issued more
often). Some will argue this is because Emacs is released too
soon, and requires more patches. In practice, I have had very few
answers to my bug reports in XEmacs (and no solution), while many
of the bugs I reported for Emacs have resulted in quick patches.
Yet, I prefer XEmacs because I can do more with it, it is easier to
use, and because the users here prefer XEmacs (i.e. they *really*
wanted the nicer GUI). Moreover, I don't have to install and maintain
as many packages "on the side", since XEmacs comes configured with
more packages.
I have to keep Emacs 19.28 around for the days when XEmacs crashes.
I have used 4 different Emacsen (Gosling, UniPress, Emacs, and
XEmacs), almost every day since 1985.
--
Martin Boyer http://www.robot.ireq.ca/Martin.Boyer/
Division Robotique mbo...@robot.ireq.ca
Institut de recherche d'Hydro-Quebec Telephone: +1 514 652-8412
Varennes, QC, Canada Fax: +1 514 652-1316
John Turner
Erik Naggum <nob...@naggum.no> writes:
> * Mike Humski <m...@rahul.net>
> | I am an emacs user. But I am going to try xemacs and read NEWS file at
> | the xemacs www site. It lists a lot of points in which xemacs is
> | superior to emacs.
Uh-oh. Here we go again...
[big snip]
> when somebody starts
> up the "I'm better than you" machine, you can be pretty certain they are
> dishonest creeps.
[little snip]
Hmmm. Interesting. I'll keep that in mind.
Thanks.
--
John A. Turner |"Music is the cup which holds the wine of silence;
Los Alamos Natl. Lab. | sound is that cup, but empty;
e-mail: tur...@lanl.gov| noise is that cup, but broken."
| - Robert Fripp
Erik Naggum
* Richard Cognot <cog...@fronsac.ensg.u-nancy.fr>
| Erik just tend to have a finger on the "flame" key ready for any
| thread where comparison between Emacs and XEmacs is involved.
XEmacs should be able to sell on its own merits, without constantly
portraying Emacs as inferior in order to get an edge on it, and without
ridiculing that from which they came, that without which they would not
exist. you will find no hostilities from the official Emacs sites, towards
XEmacs, nothing in the FAQ, there is no match for _any_ of the hostilities
from the _official_ XEmacs sources and FAQs and folks towards Emacs. this
is patently unfair. I object to that unfairness.
| Since the last one, he's been living happilly at the bottom of my kill
| file (first and last one to ever get in there, hope you won't feel too
| lonely Erik ;-) and mail from his site is also filtered out. I'm
| feeling much better now that I don't have to read his insanities.
over the past ~12 years, I have seen strong disagreement among people on
the Net and in electronic discussion groups, I have seen groups of people
gang up on one individual or another, I have seen a few individuals go
after another with an intent to kill. I have found myself seriously *hurt*
by _three_ messages in all these years. Richard Cognot has the dubious
distinction of having authored the fourth.
it's one thing to get sick and tired of any ranting and raving on XEmacs,
if that is what one thinks it is, few and far between as they are. that
can easily be avoided by not reading (or starting!) threads that compare
Emacs and XEmacs, in comp.emacs.xemacs, which I don't read and to which I
never post, unless by accident when the thread is crossposted. it's quite
another to brand everything I do for the Emacs community as "insanities".
I get _sick_ to my stomach to think that some foul bastard like Richard
Cognot can benefit from my work. I hate the fact that what I do can be of
any benefit to creep like Richard Cognot and whoever else gloats over what
he does.
_users_ of Emacs and XEmacs don't see this and may be very surprised to
find that there is so much personal hatred between developers of tools that
share such a strong common heritage. however, if you put yourself in my
place and others, and see what Richard Cognot and that crowed can do,
_SMILING_ in the middle of it, and you read some of the purely destructive
material emanating from the XEmacs crowd about Emacs and RMS and others who
have helped develop Emacs, you will understand that this is not a technical
issue, it's a question of hurt feelings and intentionally destructive
people. XEmacs has been a welcoming place for those who hate some aspect
of Emacs or another, and that is a stronger part of its legacy than
anything else could be.
I don't know who began, I don't know what each has done in the past outside
of the written record and what I have heard, but what I do know is that no
Emacs developer (past or present) has a web page explaining why it is
impossible to cooperate with any XEmacs developer, there Article Unavailable
Barry A. Warsaw
>>>>> "EN" == Erik Naggum <nob...@naggum.no> writes:
EN> Emacs' Lisp is also simpler. where Emacs has 10 functions and
EN> use existing types and access functions, XEmacs has 50
EN> functions and use an entirely new datatype. where Emacs tries
EN> not to do too many things that break working code, XEmacs
EN> doesn't care at all, claiming instead that the way it was done
EN> in the past was "braindamaged" and other favorite terms of
EN> respect, clearly implying that should somebody improve on
EN> _their_ design, they will happily accept similar treatment of
EN> their now "superior" solutions.
Of course the flip side is that for a particular given feature, where
Emacs exposes and requires you to deal with the low-level
representation, XEmacs gives you a higher level abstraction to
manipulate. This means that if the low-level implementation changes,
your code breaks. I've been bitten by this in Emacs a few times
(e.g. keymap parents and mode menus).
EN> a lot of packages are developed either for XEmacs or for
EN> Emacs, because it is often too hard to do both.
This is true. But the fault is mutual and indicates that the highest
priority for both camps *should* be to converge on a common API, and
to spend considerable effort to implement to an existing API when
possible.
EN> new, immature, fancy packages are usually developed for
EN> XEmacs. robust packages are usually developed for Emacs.
This is simply not true.
EN> in almost all respects, XEmacs gives more to the impatient,
EN> hands-on, "new" generation which doesn't have time to read
EN> manuals, while Emacs gives more to the experienced and
EN> dedicated users who read manuals to learn and understand.
Again, not true. Many very experienced programmers use and develop
primarily on XEmacs. Most try hard to make sure that the lisp they
develop runs on both Emacsen. Sometimes that's very hard to do.
-Barry
Barry A. Warsaw
I'm not disputing that Martin (or others) has experienced these
things, but here's another perspective from someone who practically
lives in XEmacs.
For me, XEmacs is usually fast enough for what I do, which includes
programming in various languages, reading mail and news, web browsing,
etc. XEmacs 19.14 does exhibit an occasional glitch where it takes a
long time on saving a file, but the problem is very intermittent in my
experience. A nit, but not an app killer. XEmacs 19.14 seems to me
to be very stable -- it almost never crashes, and certainly for me
does not crash any more, or less, than Emacs. XEmacs has some bugs,
but then, so does Emacs -- just different ones, so that's a wash. I
really dislike the display defaults that Emacs uses (e.g. font sizes,
colors, etc) but then if I really cared, I could spend some time
configuring it to my tastes.
For me, the deciding factor is the XEmacs ability to open many
different frames on many different devices, including combinations of
X and tty frames. For someone else it might be that Emacs has been
ported to Windows, where XEmacs has not.
Pick one or the other, and I'm sure you'll be happy with it. Or at
least you won't be more unhappy than if you'd chosen the other one.
:-) Most of the folks at my office use XEmacs because that's what I
maintain, because that's what I started using and have been tracking
since, I dunno, the early 90's when there wasn't even an Emacs 19 to
be had. Since then, I've never had a reason to switch.
-Barry
Peter Berger
Erik Naggum writes:
> [lots of incredibly hateful stuff.]
Dude, you have some seriously unresolved issues. Good luck
getting over it.
--
Pete Berger, Esq.
Coordinator, Regional Information Infrastructure
Pittsburgh Supercomputing Center
pet...@psc.edu http://www.psc.edu/~peterb
Rajappa Iyer
bwa...@cnri.reston.va.us (Barry A. Warsaw) writes:
> For me, XEmacs is usually fast enough for what I do, which includes
> programming in various languages, reading mail and news, web browsing,
> etc. XEmacs 19.14 does exhibit an occasional glitch where it takes a
> long time on saving a file, but the problem is very intermittent in my
> experience. A nit, but not an app killer. XEmacs 19.14 seems to me
> to be very stable -- it almost never crashes, and certainly for me
> does not crash any more, or less, than Emacs.
I'd have to more or less agree with this with the caveat that ugly
things happen when Emacsen run out of memory and it is easier to reach
that situation with XEmacs than with Emacs.
In any case, I have stopped using XEmacs altogether because I find it
unusably slow on my main platform: SunOS 4.1.3. And I'm not the only
one who has this problem... I tried everything from recompiling with
gcc -O4 to disabling lazy font lock... but it is still too slow. So,
if one is using SunOS 4.1.x, this whole discussion is really
moot... XEmacs is unusably slow on it.
--
Rajappa Iyer <r...@lucent.com> #include <std_disclaimer.h>
Unsolicited commercial mail will be proofread for a charge of $100.00
per message. Sending me unsolicited commercial mail constitutes agreement
to pay these charges.
Darin Johnson
>This is true. But the fault is mutual and indicates that the highest
>priority for both camps *should* be to converge on a common API, and
>to spend considerable effort to implement to an existing API when
>possible.
The priority should be to merge the two packages. However, I see too
much bickering. RMS strongly appears to resist anything he doesn't
like, effectively making him the final arbiter in what gets in Emacs
and what doesn't. Yes, having a better keymap data type is *good*,
yet RMS is opposed. RMS was also opposed to integrating MULE features
into Emacs as well (he wanted to wait until they had unicode support,
even though people were making use of Nemacs/MULE with local national
standards before they had heard of unicode). The history as recorded
by XEmacs, if accurate, gives good reason for its existance. Its hard
to deny that there are a lot of thing Emacs that need fixing or
redesigning (I'm a big believer that a complete redesign and overhaul
is good for a product). And if you want to fix most of them with a
differeing view from RMS, currently you have to split off from GNU
Emacs.
On the other hand, XEmacs people go out of their way to antagonize
Emacs developers. I'd love to see cooperation here, but saying
something like "we don't use a brain damaged way of doing..." doesn't
help matters. And continuing to widen the differences between the two
versions instead of working to closely track GNU code exacerbates things
as well. And the API is much more complex; sometimes this is necessary
because of lapses in Emacs, but too often it seems needless as well.
I prefer XEmacs though; even though I don't make use of most of the
X11 support. But I'd prefer a single GNU/Xemacs/MULE.
--
Darin Johnson
da...@connectnet.com
David Masterson
>>>>> "Erik" == Erik Naggum <nob...@naggum.no> writes:
> the best solution _is_ probably to regard XEmacs as non-existent. but that
> is wrong, too, according to complaints from the XEmacs folks, who are
> constantly complaining that Emacs web sites and documents don't mention
> XEmacs. why should they?
Why should there be Emacs web sites and documents that mention XEmacs? To
help answer the simple question that all new users ask -- "which version of
Emacs should I use?"
> the XEmacs folks were childish rebels on a mission of destruction and
> competition, and so got disinherited. is that really surprising to them?
> and why do they want to be listed on the web page for a "braindamaged"
> competitor, anyway? the fact is, XEmacs is _inferior_ to Emacs in all
> respects that count. so to hell with it.
Don't think that way...
There seems to be a lot of resentment on both sides of the issue. If you
carry that resentment forward in this way, then people find it hard to believe
a statement like "XEmacs is _inferior_ to Emacs in all respects that count"
(or the reverse statement for that matter).
--
==================================================================
David Masterson KLA Instruments
408-456-6836 P.O. Box 49055 M/S F1-9440
dav...@prism.kla.com San Jose, CA 95161-9055
==================================================================
I only speak for myself -- it keeps me out of trouble
zinc
Erik Naggum <nob...@naggum.no> writes:
>
> * Hrvoje Niksic <hni...@srce.hr>
> | Neither do I. But it seems a lot.
>
> please try to be objective, if you can. out of _many_ thousand news
> articles a year, less than a dozen even _mentions_ XEmacs. I also spend
> more than four thousand hours a year working with Emacs (it's the whole of
you spend 12 hours/day, 365 days a year in front of a terminal?
perhaps there's a bigger problem here...
-pjf
--
"Those that give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety." -- Benjamin Franklin (1773)
finger for PGP key
zifi runs LINUX 2.1.7 -=-=-=WEB=-=-=-> http://zifi.genetics.utah.edu
Per Abrahamsen
Hrvoje Niksic <hni...@srce.hr> writes:
> Whereas XEmacs developers advocate data abstraction
> (like specialized types for keymaps etc.), GNU Emacs developers are
> against it.
I think the term `information hiding' is more accurate than `data
abstraction'. GNU Emacs supports many of the same abstract
interfaces, but without any attempt to hide the underlying
implementation.
Alain Ketterlin
Rajappa Iyer <r...@lucent.com> writes:
> bwa...@cnri.reston.va.us (Barry A. Warsaw) writes:
>
> > For me, XEmacs is usually fast enough for what I do, which includes
> > programming in various languages, reading mail and news, web browsing,
> > etc.
[...]
> In any case, I have stopped using XEmacs altogether because I find it
> unusably slow on my main platform: SunOS 4.1.3. And I'm not the only
> one who has this problem...
Sorry I have to disagree.
I don't have any problem on SunOS 4.1.3 (on a Sparc2 --really-- with
24Mb of RAM, from an x-terminal via xdm). However I must admit that I
had to turn off func-menu's auto-rescan and display-in-modeline. But
that's all. I'm spending my time on programming and typing text:
XEmacs (+cc-mode+auctex) is simply the best environment I've ever
met. I've very happy that I can *also* read mail and news easily with
it.
I've seen several people complaining about XEmacs' startup time: this
is simply irrelevant for me, since the instance of XEmacs from which I
post was started 17 days ago.
I tried several times to get into Emacs, but found it really too hard
to use. Maybe now (after one or two years spent on XEmacs) I could be
able to do something with it. Not sure. I won't try: I simply have no
time to spend on the tools, I just need to use them.
My 0.02 euros.
---Alain.
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> * Richard Cognot <cog...@fronsac.ensg.u-nancy.fr>
> | Erik just tend to have a finger on the "flame" key ready for any
> | thread where comparison between Emacs and XEmacs is involved.
>
> XEmacs should be able to sell on its own merits, without constantly
> portraying Emacs as inferior in order to get an edge on it,
For many people the choice is not "should I try using XEmacs instead
of cat" but "should I try using XEmacs instead of Emacs". To
enumerate basic differences is thus understandable.
> and without
> ridiculing that from which they came, that without which they would not
> exist. you will find no hostilities from the official Emacs sites, towards
> XEmacs, nothing in the FAQ, there is no match for _any_ of the hostilities
> from the _official_ XEmacs sources and FAQs and folks towards Emacs. this
> is patently unfair. I object to that unfairness.
Would you care to quote some bad verbage from the current FAQ and NEWS
files? I am unable to see anything akin to that which you describe in
the current versions.
I agree that there *were* times when the FAQ was pretty outspoken
(some XEmacs maintainer must have been real mad at RMS, for whatever
reason, at that time), but this has been cleaned up.
The most I see of hostility and hate messages towards the "other camp"
nowadays is from you.
BTW, in case you missed it: XEmacs is freeware under the GPL (being a
derived work from GNU Emacs, it never could be anything different) and
freely available in source, so I find your rants about the "commercial
traits" of XEmacs and its marketing and whatever slightly obtuse.
Holger Schauer
>>"MB" == Martin Boyer schrieb am Wed, 6 Nov 1996 12:27:13 GMT:
In article <x74tj3w...@amadeus.ireq.ca> Martin Boyer <mbo...@robot.ireq.ca> writes:
MB> Mike Humski <m...@rahul.net> writes:
>> If you know both, please let me know in which area emacs is
>> superior to xemacs.
MB> 1. Emacs is faster, specially on character-based terminals.
MB> 2. Emacs starts much faster. 3. Emacs is smaller, but that isn't
MB> noticeable for most users. 4. I don't remember in which *year*
MB> Emacs last crashed on me. XEmacs has crashed on me or a user
MB> here every week for versions 19.12, 19.13, and 19.14 (the latest
MB> version). I have been told this is due to garbage-collection
MB> bugs, which are very difficult to debug. 5. The Emacs "core" is
This is very interesting. In fact I'm using XEmacs (19.11 through to
19.14) since almost three years now on Sun, Solaris and Linux. I
encountered some problems some times but mostly they had to do with
packages which had some bugs in them. I *never* had any real crash
with XEmacs up to this week when the new version of Ilisp 5.8 came out
which crashed XEmacs 19.14 twice (which was the reason to reinstall 5.7
again).
So, I am just curious, when happend the crashes, what did the users do
to kill it ? (Note, I'm just a private not somebody devoted to
developing XEmacs.) Perhaps this has to do with the type of machines
XEmacs runs on ?
Holger
--
holger_schauer :-
mail_address("Holger....@gmd.de"),
project("BGP-MS/AVANTI, GMD Sankt Augustin, FIT.MMK"),
www_home_page("http://www.uni-koblenz.de/~schauer/index.html").
(^:= A donkey came to my office. It had a theory about people anaphora...
Hrvoje Niksic
William Paul Vrotney (vro...@netcom.com) wrote:
> I read every word. Well spoken. I've always avoided XEmacs, not knowing
> why, now I do. Frankly, I could never see a reason for it's existence with
> GNU Emacs being there for us. It might serve as a healthy competition for
> GNU Emacs (and vice versa) if the XEmacs crowd were not so hostile as Erik
> says.
"XEmacs crows" is not as hostile to GNU Emacs as Erik would like them
to be. Of course, you can believe Erik's word, or you can try reading
the thread and finding who uses the word like "liar", "idiot",
"insane", etc. Feel free to read the XEmacs FAQ (it's posted on
comp.emacs.xemacs) and find all those foul wordings.
Or, you can trust Erik. It's up to you.
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
I'm a Lisp variable -- bind me!
Hrvoje Niksic
William Paul Vrotney (vro...@netcom.com) wrote:
> I read every word. Well spoken. I've always avoided XEmacs, not knowing
> why, now I do. Frankly, I could never see a reason for it's existence with
> GNU Emacs being there for us. It might serve as a healthy competition for
> GNU Emacs (and vice versa) if the XEmacs crowd were not so hostile as Erik
> says.
"XEmacs crowd" is not as hostile to GNU Emacs as Erik would like them
Barry A. Warsaw
>>>>> "WPV" == William Paul Vrotney <vro...@netcom.com> writes:
WPV> In fact, in general, why is there so much hostility between
WPV> computer tool user groups? Is this human nature?
Yes, it is. It's because the environments we use on a daily basis is
our religion, and humans will fight to the death to defend their
religions.
Why does vi still exist when we have Emacs? Why does Emacs still
exist when there are perfectly good commercial IDE's that will do it
all for you? For that matter, why does Unix still exist since we now
have Windows? Or FORTRAN now that we have Java? etc., etc...
-Barry
Patrick Giagnocavo
Erik Naggum wrote:
>
> * Kevin Fenzi <ke...@scrye.com>
<a whole bunch of other people wrote too!>
GOOD GRIEF!
No wonder Unix is slow to get into the business market, but thrives in
academia (and I say this as someone working for a commercial software
publisher): the reason everyone's so petty is because the stakes are so
small.
The reason I use Xemacs is because it was preconfigured with VM, etc.
And I wanted to try out GNUS, etc. For someone who previously used vi,
all this configuration stuff for Emacs was too hard to figure out.
Maybe later, I will switch to Emacs on some other systems since I now
understand how everything is configured, and I will have a lower end box
to run it on.
The reason people use an OS (or a text editor) is because it solves a
problem or fills a need. Windows sure as heck doesn't win on technical
merits, but it's out there in volume because it is seen as providing a
solution (the unstated problem is apparently putting money in BG's
pockets).
Why don't you folks grow up? What does having two Emacsen mean, except
that the 'market' or 'number of people wanting to use this' is high
enough to support it? That is, both programs fill a need.
I am not going to continue dignifying this thread with any more
responses. I urge everyone else to do the same. Use what you like.
Cordially
--
Patrick Giagnocavo - pat...@xinside.com
X Inside Incorporated - Accelerated X Servers
Technical Support Department
William Paul Vrotney
In article <30562684...@naggum.no> Erik Naggum <nob...@naggum.no> writes:
>
> once upon a time, it must have been a little over two years ago, I tried to
> find out why the hell the XEmacs crowd was so hostile to the Emacs crowd.
> ...
>
> thank you for listening, whoever you are. I fully expect the folks who
> could have had some use for this message to have seen red and become all
> enraged just by seeing the headers fly by. if you think it's worth it,
> alert some of them to this message. thanks.
>
I read every word. Well spoken. I've always avoided XEmacs, not knowing
why, now I do. Frankly, I could never see a reason for it's existence with
GNU Emacs being there for us. It might serve as a healthy competition for
GNU Emacs (and vice versa) if the XEmacs crowd were not so hostile as Erik
says.
In fact, in general, why is there so much hostility between computer tool
user groups? Is this human nature? Is there too much sting to the emailed
word? Being this close you would think that we would stick together. Would
someone write a book and explain this to poor us.
--
William P. Vrotney - vro...@netcom.com
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> XEmacs has chosen a path of competition. Emacs still maintains a path of
> cooperation, but not the "democratic" style that some people want.
Pretty wrong. Lucid was in the need of an Emacs with X functionality,
and the official Emacs had no date-line to come up with something like
that. So they made their own variant and developed quite a bit of it
further. Later, Emacs 19 was concocted by GNU, with quite different
inner workings and some similar functionality, some different. The
Lucid line of work was not discontinued, however, since quite a bit of
work depended on it already. Later this became XEmacs.
The XEmacs team have not decided a "path of competition", but have
just not let the work of Lucid Emacs go to the dumps. If any has
chosen a "path of competition", it is the FSF with their completely
own interface to X instead of trying some compatibility to existing
copylefted code. Much of the work of XEmacs maintainers goes into
imitating any new interfaces Emacs happens to introduce for features
XEmacs already has, so that XEmacs users may use packages written for
Emacs as well. This accounts for some of the "bloat" of XEmacs.
Although some people would have the XEmacs maintainers drop dead
rather now than later in order not to be nagged about missing features
in Emacs, I don't think this the best path of choice. After Emacs has
possibilities to embed graphics in buffers (indispensible for
integrated WEB browsing, and useful for, say, TeX WYSIWYG display of
formulas, X-Face inclusion in buffers and quite a few other things),
possibilities to use proprotional fonts with variable line widths,
possibilities to use colours on ttys, possibilities to have frames on
different devices (ttys *and* X displays) at once, and quite a few
other packages adapted (hyper-apropos is pretty hot, for example),
XEmacs still has a place, and if just to remind the Emacs people what
would be nice to have. Once the basic display engine of Emacs has the
power of that of XEmacs (but hopefully better speed...) one might
think of fading XEmacs out by offering similar feature sets.
Bruce Stephens
>>>>> "BAW" == Barry A Warsaw <bwa...@cnri.reston.va.us> writes:
> Pick one or the other, and I'm sure you'll be happy with it. Or at
> least you won't be more unhappy than if you'd chosen the other one.
I think this is good advice. Although I'm using the other GNU Emacs
now, a couple of years ago I used XEmacs all the time, and liked it:
the interface is pretty, and it's nice having everything set up
reasonably so it's easy to access lots of different packages (many
packages in the other GNU Emacs require installing and hand editing of
.emacs files). The other GNU Emacs should certainly steal XEmacs's
"Options" menu.
From the philosophy point of view, I can see arguments on both sides.
I really don't understand rms's objections to opaque data types. I
can see objections in specific instances, but I think keymaps, for
example, make an obvious candidate for a data type. On the other
hand, I think overlays and text properties make sense, and I agree
with rms's doubts about how extents work in a theoretical sense. But
hey, if you're a user like me, so long as things work you'll not
notice how dubious things are underneath.
--
Bruce Stephens | email: B.Ste...@math.ruu.nl
Utrecht University | telephone: +31 30 2534630
Department of Mathematics | telefax: +31 30 2518394
P.O. Box 80010, 3508 TA Utrecht, The Netherlands
Erik Naggum
* David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de>
| Lucid was in the need of an Emacs with X functionality, and the official
| Emacs had no date-line to come up with something like that. So they made
| their own variant and developed quite a bit of it further. Later, Emacs
| 19 was concocted by GNU, with quite different inner workings and some
| similar functionality, some different. The Lucid line of work was not
| discontinued, however, since quite a bit of work depended on it already.
| Later this became XEmacs.
this is the kind of marketing I have been talking about. somehow, somebody
has managed to dupe David Kastrup into believing this line, and it is this
line that continues to hurt Emacs, and make fools who believe it exhibit
the "we're so superior" attitude.
"later, Emacs 19 was concocted by GNU". in _fact_, FSF was already working
on Emacs 19 when Lucid came in and supported the work with a lot of money.
they withdrew the money and hired away the programmer from the FSF when the
pace was not to their liking. the first is OK, the latter is not. they
took over the decision process, and lots of people who were unable to work
with RMS (that's the kindest way I can put it) found a forum to vent their
frustration and spleen in the process.
| This accounts for some of the "bloat" of XEmacs.
geez. talk about marketing lies. "now look what you made us do!"
| Although some people would have the XEmacs maintainers drop dead rather
| now than later in order not to be nagged about missing features in Emacs,
| I don't think this the best path of choice.
you really have a knack for dirty campaigning. did you work for Dole?
what _I_ would like to see is that XEmacs' maintainers would suggest common
Lisp functions that deal with the various data types and actually implement
data _abstraction_. (Per Abrahamsen said it very well when he pointed out
that it is more "information hiding" than "data abstraction" in the XEmacs
code.) real data abstraction would not be visible to the user in terms of
"we use abstract data types, and they don't". data abstraction is about
_not_ forcing users to see the implementation. in XEmacs, users are in
fact _forced_ to see the (different) implementation. I can find nothing to
fully explain this other than ill will: the desire to break things on
purpose, or just to spite RMS for using a more open design.
| After Emacs has possibilities to embed graphics in buffers (indispensible
| for integrated WEB browsing, and useful for, say, TeX WYSIWYG display of
| formulas, X-Face inclusion in buffers and quite a few other things),
| possibilities to use proprotional fonts with variable line widths,
| possibilities to use colours on ttys, possibilities to have frames on
| different devices (ttys *and* X displays) at once, and quite a few other
| packages adapted (hyper-apropos is pretty hot, for example), XEmacs still
| has a place, and if just to remind the Emacs people what would be nice to
| have.
yeah, right. XEmacs continues to draw away programmers from the Emacs
community who _could_ have implemented this in Emacs. "we have XEmacs, so
why bother" is an oft-heard argument against working on X stuff in Emacs.
this is the most hurtful part. had the XEmacs folks been a little less
spiteful and hateful, had they written their code with _abstraction_
instead of gratuitously changing the Emacs Lisp language, and had they
cooperated, we could have had two display engines for a single Emacs,
instead of two Emacsen with each their own display engine.
another very useful item in judging Emacs vs XEmacs is to look very
carefully at how they approach the Lisp code: in Emacs, it is a real
programming language, and the C is just support code for it, whereas in
XEmacs, you find people talking about the "API" all the time, and they are
right: in XEmacs, Lisp is just an API to a much more complex C layer. I
spent a fair amount of time studying XEmacs Lisp, and I got the same icky
feeling I get from most of the many incredibly obsolescent solutions in
Unix: I had to "think implementation" to understand how something was
supposed to be used and what it did. the most curious aspect to this is
that the XEmacs folks complain (bitterly!) about how much they had to work
at the implementation level in Emacs Lisp, such as the now tired story of
the keymaps. the question I kept asking myself through my XEmacs excursion
was: why did they have to change the low-level implementation to gain what
they say they gained? why do you, David, try to blame Emacs for the bloat
in XEmacs and say that so much effort goes into interface wrappers, when
that is solely the (historic) fault of the XEmacs designers?
I'm reminded of the situation between the two competing mobile phone
companies in Norway. company A reduces its prices by "up to 25%" soon, but
not on calls to their competitor, company B. company B will follow suit
and they also has far lower prices for calling company A to begin with.
now, I'm the inquisitive kind of guy, so I called around and got the _real_
reason: company A has to pay 80% of the price it charges its customers per
minute for calls into company B's mobile network, and is squeezed for
profits because they charge so much. company B has to pay only 40% of the
price it charges its customers per minute for calls into company A's
network. this is because company A is an arm of the Norwegian (land-line
monopoly) phone company and can't charge arbitrarily much from their
competitors. their competitors can charge arbitrarily much from them, of
course. and so it appears, to the people who want to be disgruntled and
don't want to understand, that company A _overcharges_ for calls to B,
while B is your regular nice guy. the _fact_ is that this false appearance
is created _on_purpose_ by company B in their marketing campaigns to gain
market share. so if it is hard to move code from XEmacs to Emacs, every
XEmacs user and every developer who complains about it is likely to blame
Emacs, not their own environment and the childish little bastards who laid
the ground for the incompatibility.
if you don't like the fact that there is a strong sense of incompatibility
between Emacs and XEmacs in their keybindings, ask the doofus who broke it.
if one instead wanted to maintain compatibility, find another way to talk
about key bindings. I think using the functionality in `read-kbd-macro'
would be a good way, if we can agree on the external appearance, which is
that from `key-description', and I think we can. _that_ would be _real_
abstraction. the function should probably be called something else, such
as `read-key-sequence-from-string' and have a reader macro defined to call
it more conveniently, such as one would do in Common Lisp, like #K"C-M-;",
to mean the key printed as "C-M-;" by `single-key-description', whether it
would be represented as the list (control meta ?;) or the integer -67108805
(which is the value of ?\C-\M-;). clearly, having to expose the user to
the implementation is wrong in _both_ cases if there is more than one way
to do it. but _again_, why did the XEmacs folks have to change the
_implementation_ in order to improve it? ill will, incompetence, or both!
| Once the basic display engine of Emacs has the power of that of XEmacs
| (but hopefully better speed...) one might think of fading XEmacs out by
| offering similar feature sets.
you're proving my point: XEmacs wants competition, not cooperation. Emacs
has wanted to employ the experience and the code from XEmacs for years, but
why can't we do that? because the XEmacs folks chose a way to do their
work that made it much, much harder than it should have been, and they did
it _on_purpose_. progress on the unified display engine is _very_ slow. I
don't know precisely why, but what I do know is that the reason Emacs
doesn't have a better X interface is that the XEmacs people hasn't _wanted_
it to. if they do now, it does not show in action.
incidentally, I don't want proportional fonts, I don't want colors, I don't
want inline graphics, I still prefer the 1200 dpi quality of printed text
(and high-end laser printers) and _real_ pictures, I think low resolution
is the worst aspect of computers, and colors don't help. I want the screen
to repaint itself _instantly_, I think the "slow motion" of most Motif and
Windows applications and constant repainting (as in Netscape 2.02) is
evidence of bad programmers suffering from technology fixation. if I can't
get at _least_ as good quality as I got without computers, they would have
to offer a _lot_ more than I had, and usually they don't. XEmacs also
follows the "modern" trend where appearance rules over contents. this is
an additional pet peeve of mine, and it doesn't help XEmacs' cause, but it
would have been irrelevant if they could point to _real_ improvements in
areas that matter to me. they can't. Emacs is in fact superior to XEmacs
in areas that matter to me. but I recognize (and so should the XEmacs
folks) that this is due to my personal preferences. just like my ideas
about what free software should be like define my attitudes to a lot of
what goes on in our industry, but may not be shared by others, I want to
XEmacs folks to stop portraying their personal preferences as inherently or
intrinsically better than any others'. there is already enough of the
bullshit in the marketing from Microsoft and others about how the "normal"
users are those who prefer their _sick_ version of "user friendly", while
this is _wrong_ for at least 40% of the population, quite possibly more.
e.g., the "mouse" as in input device is creating new health problems that
may well have offset any perceived gains already in terms of total costs,
not to mention the personal disasters for those who get ill from it. I
have been a computer user since about 1978, I have _no_ problems with any
of the things that others are attributing to their working environment
after only a few years. could it be because I have resisted ways of
working that "modern" designers are hailing as "improvements" when they are
in fact no improvement at all, except to their sales figures. the short
range by which we judge things today is going to hurt all of us later.
people behave as if they don't even expect to become 40 years old, with
loss of eyesight, etc. well, enjoy your youth. I plan on enjoying my
_life_. I find the constant rush to do things that are harmful in the long
term, but probably "sexy" or "cool" in the short term, to be significantly
bothersome. more importantly, short-term euphoria is not "superior" to
long-term quality, except to people who have no desire for quality. call
it "conservative" if you like. I call it factoring out the psychological
requirements for stability into areas where they cost very little, so I can
handle more changes elsewhere, such as in spending my braintime on hard
(programming) problems instead of tweaking display features.
sorry for the length, but at least it should be more constructive, now.
BTW, I noticed that Jamie Zawinski saw it fit to have a little fun. that's
good for him, but I'd rather spend my time writing the above paragraphs
than run a "complaint" script off the web.
#\Erik
--
Please address private replies to "erik". Mail to "nobody" is discarded.
Steve Jones - JON
Erik Naggum wrote:
>
[.. snip repeat of old arguments ..]
>
> incidentally, I don't want proportional fonts, I don't want colors, I don't
> want inline graphics, I still prefer the 1200 dpi quality of printed text
> (and high-end laser printers) and _real_ pictures, I think low resolution
> is the worst aspect of computers, and colors don't help. I want the screen
> to repaint itself _instantly_, I think the "slow motion" of most Motif and
> Windows applications and constant repainting (as in Netscape 2.02) is
> evidence of bad programmers suffering from technology fixation.
Now I have to take issue, colours don't help ? Try presenting anything
other than simple text in B/W and see the issues. Why do I use colours
in Emacs and XEmacs ? Because these can quickly highlight the code and
give it a quicker perception than without colour. There is of course a
limit at which point colours become useless (ie don't have more than 4
or 5 colours when using text), but colour can add alot to a document.
If colour doesn't help why do humans have colour vision ? Why are
traffic lights Red/Amber/Green what information is inherent in these
colours ?
> if I can't
> get at _least_ as good quality as I got without computers, they would have
> to offer a _lot_ more than I had, and usually they don't. XEmacs also
> follows the "modern" trend where appearance rules over contents. this is
> an additional pet peeve of mine, and it doesn't help XEmacs' cause, but it
> would have been irrelevant if they could point to _real_ improvements in
> areas that matter to me. they can't.
Appearance v. Content ? For most modern systems the appearance is the
content, working on ATC systems yuo can have the worlds greatest
Conflict algorithm but if the presentation is poor it means nothing.
The Human Computer Interface is the key part of modern systems, XEmacs
has its problems and Emacs is not a bad interface but I would say that
the XEmacs trend is more in line with what will produce better
productivity. Sorry to bring in the MS world but what makes Word for
Windows easier to use than Word for Dos ? The answer is interaction.
The Mouse is an excellent tool, but looking at some of the modern touch
screens I think that they may soon be taking over (well in the next few
years). Also there is evidence that the keyboard has more effect on
health than the mouse (as long as it is a well designed mouse). The
natural urge of people is to point, the mouse is the interaction tool
that allows the user to do this.
As someone who specialises in GUIs I would say that the problems
encountered more than match those of the internal working of computers.
Interface meet a thing called a "human" these input devices are badly
designed and constantly fail to obey protocol, if they were made by a
company they'd never be bought. I feel the objective of modern
computing should be to make the computer a tool that is accesible to
all. GUIs do cause problems for speed, they are resource hungry, and
indeed there is a body of evidence that says that it does not always
speed a process up, it does however improve quality.
A secretary, a physist, a chemist and an English student have no need to
know about the complexities of using a computer system to be able to use
what is an editor. Just as a programmer is not required to know about
General and Special Relativity to programme that computer. XEmacs does
have teething problems, but then it is attempting to do something new,
and being amoung the first is always harder, but if it creates a simpler
editor with more features this is for the better.
The Interface should be the most important thing for and Editor, this is
where people interact with the system. If a tool is hard to use then it
is a poor tool, no matter how good the functionality is behind it.
Emacs and XEmacs are both excellent tools, and both have over the past
few years improved their interfaces to the user. Long may this
continue.
Steve Jones
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> "later, Emacs 19 was concocted by GNU". in _fact_, FSF was already working
> on Emacs 19 when Lucid came in and supported the work with a lot of money.
> they withdrew the money and hired away the programmer from the FSF when the
> pace was not to their liking. the first is OK, the latter is not. they
> took over the decision process, and lots of people who were unable to work
> with RMS (that's the kindest way I can put it) found a forum to vent their
> frustration and spleen in the process.
So presumably that was how the animosities came about. If that was
how the things went...
Sort of, say, somebody supporting Joyce's initial work on "Ulysses" with
quite a bit of money, and giving him a coworker supposed to help crank
out a snazzy story pretty soon. The great plan and the quick goals
are not compatible, a splitup occurs, and pretty fast a "snazzy story"
gets cranked out playing in Dublin. It has quite a nice effect on the
masses who'd not be able to appreciate the long-term plan, anyway, and
get what they were waiting for (although a bit unshapely) before.
But it is an error to assume that this would have stolen much of an
audience for "Ulysses"... For most people the choice would have been
reading the snazzy thing, or nothing.
Now one point is that Lucid has closed the shop, and the commercialism
has quite gone out of XEmacs anyhow (although possibly not the rapid
development). Both still serve different people. In the XEmacs
developers' camp by now care has been taken to remove most traces of
the once more valid animosities between the programmers, from docs,
FAQ and mostly from the code. Some users, however, try to fight on
for a war most of their leaders have abandoned.
Talk about a merger being obstructed by one of the parties is pretty
pointless, in my opinion: the methods of development are quite a bit
too different to make cooperation without quite a lot control and
discipline on both sides seem fun. And creating free software is like
writing books: it has to be fun for the writer, and suit his style.
Using RMS' careful style, it would take several years to get those
things people like about XEmacs integrated properly as he likes it
into Emacs, and this is probably not even a bad thing, because one can
still rely on Emacs being fast and reliable for everyday's use.
There have been signs in the last time of developers at least trying
to coordinate how functionality is accessed in both Emacsens, and this
I find encouraging.
Brian Masinick
Mike Humski <m...@rahul.net> writes:
> I am an emacs user. But I am going to try xemacs and read NEWS file at
> the xemacs www site. It lists a lot of points in which xemacs is superior
> to emacs.
Xemacs is much more graphically oriented, and has nice integration
of a number of applications, including News (I'm writing this using
the Xemacs version of GNUS), Web browsing, Mail, and many others. The
interface has a number of icon buttons that can be clicked to call up
these applications, which makes it nice and easy to use many things.
> Can I take them straight? You have to consider what is important to
you, then make up your mind which interface to use, and for what purposes.
> Is emacs inferior to xemacs?
No, it is just different. Both are extensible, but they have
diverged somewhat due to philisophical differences in what the base
interface should provide.
> If so, why emacs is still around. Why doesn't every one switch to
> xemacs?
For me, there are a couple of reasons to still use Emacs. One, my
employer ships a moderately recent release of GNU Emacs with the base
OS, and makes the most recent updates available on a public server.
Secondly, I find that GNU Emacs in base form is more compact than
Xemacs and runs faster. Third, GNU Emacs is not quite as complex, and
it has been around longer, so there is more stuff available (at least
to me) that I need and can use, so it is my primary work environment.
Fourth, Xemacs can not always keep up with my typing, even though I
type at best just over 30 words per minute. In this reply, for
instance, it dropped spaces several times, which I had to go back to
insert.
> It seems there must be something in which emacs is better at than xemacs.
> Otherwise, emacs would have perished.
Speed, availability, number of users, and number of supported
platforms. Xemacs is quite tied to the X Window System, although the
base functionality is nothing more than GNU Emacs. I've not seen it
used anywhere except UNIX systems, although it could be conceivably
ported elsewhere. But I know for a fact that GNU Emacs is being used
on UNIX, OpenVMS, Windows NT, and Windows 95, and people port it, or
at least part of it to many other platforms as well.
> If you know both, please let me know in which area emacs is superior to
> xemacs.
Both GNU Emacs and Xemacs are based on GNU Emacs source code. I
use Xemacs whenever I want a highly graphical user interface, multiple
active frames, and several applications running. GNU Emacs can do
these things, too, but the Xemacs package ships with them turned on.
I use GNU Emacs for most of my code development, and general text
editing. I also use vi, ed, ex, edit, vim, jed, jove, nedit, xedit,
pico, sedt, TPU/EVE, edt, CDE's dtpad (notepad), Microsoft Word,
Microsoft Write, and Microsoft Notepad on occasions, so I know how to
use many different editors and word processors. GNU Emacs is my
favorite of them all, but I like all of the Emacs variants because of
their extensibility. I use the others for speed, convenience, or just
to keep my fingers familiar with other tools in case I run into a
situation where I need them.
Brian Fri Nov 8 13:48:55 1996
--
Brian W. Masinick WWW: http://www.zk3.dec.com/~masinick
Mailstop: ZKO3-3Y25 Internet mailto:masi...@zk3.dec.com
110 Spit Brook Road FAX: (603) 881-2257
Nashua, NH 03062-2698 DTN: 381-0013; Phone: (603) 881-0013
Lower your monthly phone bill: http://www.freedomstarr.com/MA9668183.rep
Brian Masinick
da...@connectnet1.connectnet.com (Darin Johnson) writes:
> I prefer XEmacs though; even though I don't make use of most of the
> X11 support. But I'd prefer a single GNU/Xemacs/MULE.
I'd second that motion, although I would want to keep the base
functionality as close to GNU Emacs as possible for speed and
efficiency, but make the interfaces with Xemacs and MULE as common and
extensible as possible, so that I could use a fast implementation by
default, but add font, graphics, and I18N support easily without
having to write or rewrite it every time I want to do something new.
Cooperation and compromise would go a long way here. Criticism and
slander do little to help. I think each of the approaches has some
merit, and it would be nice to have a common programming interface and
let the user and programmer decide how much of it they want to
activate. I use GNU Emacs as my editing and development editor, and
Xemacs as my applications platform. I haven't used MULE a lot yet,
even though I do a lot of I18N work. The more I work with Asian code
sets, the more likely I am to use MULE in the future, however, unless
we can merge this stuff together in a cordial manner.
Erik Naggum
* Steve Jones <Steve...@eurocontrol.fr>
| Now I have to take issue, colours don't help?
not me. that's the point I'm trying to make here. color is not a
universal good, the way some people believe it is. I don't to stuff that
needs color. I do stuff that is _hampered_ by color.
| If colour doesn't help why do humans have colour vision? Why are traffic
| lights Red/Amber/Green what information is inherent in these colours?
if you wish to universalize your argument, you need more than that. a
simple counterargument: I have over 1800 books, of which 50 or so have
color illustrations. am I unable to learn from the rest, or enjoy them?
how come I find the imagery in good books better than the imagery in
movies? clearly, some people are more verbal than visual. you, like most
of the people who are unable to deal with natural variation in human
preferences, want to stuff colors down my throat, and I have to pay for it
in reduced resolution, more expensive displays, not to mention _larger_
displays to handle 19" or 21" screen real estate. have you asked me? no,
you _argue_ against _my_ preferences. your arrogance is unbelievable!
| For most modern systems the appearance is the content, ...
exactly! is this a good thing? no. this caters to a class of people who
obviously are unable or unwilling to handle words. with all the lack of
contents in published material these days, _all_ they have to speak for
them is appearance. this has been a recognized problem on the Internet
(WWW) for at least two years, now. do you wish to tell me that I _must_
prefer your colorful, contentless, "modern" pages over an information-rich
"dull" page? if so, go to hell.
| The Human Computer Interface is the key part of modern systems, XEmacs
| has its problems and Emacs is not a bad interface but I would say that
| the XEmacs trend is more in line with what will produce better
| productivity.
"better productivity"? (whose marketing department are you paid by?)
that's nothing short of a load of bullshit. better productivity depends on
the task and on the person. you ignore the _invidual_ completely, to be
replaced by your obnoxious categorization of what is best for humans as
such. humans don't exist as such, they exist as individuals. like any
good dictator, however, some people would like to take away the working
environments that make "deviants" more productive, in order to make the
vast masses more productive.
| The natural urge of people is to point, the mouse is the interaction tool
| that allows the user to do this.
do you actually _believe_ this "natural urge" stuff? my "natural urge" (if
I have any that exceed sex and hunger in complexity) is to speak or write.
| A secretary, a physist, a chemist and an English student have no need to
| know about the complexities of using a computer system to be able to use
| what is an editor.
your saying so does not make this true, no matter how hard you try to make
it so. the fact is that this _stupid_ line of argumentation has kept
computing away from those who could have benefited from it decades ago,
because what you are _really_ saying is that "you poor idiots, computers
are still too hard to understand for your puny little brains to be useful
to you". the _belief_ that computers are any more complex than, say, cars
and driving in heavy city traffic is so incredibly condescending to those
users that I find myself speechless when faced with the incredible stance
"we know what's good for you, just you sit still until we fix it for you".
you live in your graphical user interface world with colors and idiots for
users. I prefer people who can communicate and think in words. I have
been to France once (Aix-en-Provence, a lovely little town, and Marseilles,
a not so lovely city). I had to point at things because my French wasn't
up to speed. I made an intense effort to learn to _speak_ French, and
after two weeks, I could speak it well enough to discuss the best packaging
to send home a bunch of books I had bought in one of the excellent
bookstores in Aix, after having spent a day in delighted (albeit patient)
discussion with the bookstore owner. the thrill of being able to speak a
new language was just exhilarating. unfortunately, I can't speak French, I
can't even write French, anymore, but I can still read French, and I read
German, as well. reduced to pointing, I feel like an illiterate moron.
that feeling carries over into computing. pointing is for people who have
yet to discover _thought_. in my view. you obviously disagree, but you
won't see me agree to your "natural urge" bullshit any time sooner than you
stop universalizing that bullshit to include me and millions of other
people who feel _disenfranchised_ by the now point-and-click "interaction"
you want to make the universal mode of communication with a computer.
this has nothing to do with any Emacsen, anymore, but I just want you (and
others) to know your "natural urge" nonsense is disputed by _individuals_
who don't actually _like_ your "interaction for dummies". and for those
who are inclined to point and not think: this is an example of how one
person (namely, I) think and have preferences, not a universalizable
argument about what Emacs users prefer over what XEmacs users prefer.
however, I'm still inclined to believe that XEmacs users (not the least
after reviewing the comments I have received), are less verbal and more
visual than Emacs users, are less interested in studying manuals and
learning _languages_ (case in point: XEmacs calls its Lisp an "API", Emacs
calls its Lisp a programming _language_). David Hughes felt _insulted_ by
this, for God knows what reason. observation it is, and it has been
confirmed by this debate, and in no way disputed.
a word says more than a thousand images.
exercises for the visually inclined: illustrate "appreciation", "humor",
"software", "education", "inalienable rights", "elegance", "fact".
Rick Baartman
Naggum sez:
>a word says more than a thousand images.
Indeed. This contradiction of a what most people take as a truism nicely
sums up your philosophical position. All kinds of things follow from this.
DOS is better than Windows, emacs is better than Xemacs, ... Fine. I can
except that you have this point of view and won't endeavour to change
it. But keep in mind that most people _have_ gone over to using
GUIs. Therefore, your position that GUI designers are foisting this on a
public that doesn't need it (if only everyone would learn how to
communicate with their computers in text) is the arrogant and intolerant
position. Not the other way around.
Just a casual observation: Those people using the strongest language are
the intolerant ones. Check back on this thread and see which side is using
the strongest language.
--
rick
David Kastrup
vro...@netcom.com (William Paul Vrotney) writes:
> In article <30562684...@naggum.no> Erik Naggum <nob...@naggum.no> writes:
> > once upon a time, it must have been a little over two years ago, I tried to
> > find out why the hell the XEmacs crowd was so hostile to the Emacs crowd.
>
> I read every word. Well spoken. I've always avoided XEmacs, not knowing
> why, now I do. Frankly, I could never see a reason for it's existence with
> GNU Emacs being there for us. It might serve as a healthy competition for
> GNU Emacs (and vice versa) if the XEmacs crowd were not so hostile as Erik
> says.
Well, Erik Naggum has warned you about getting sucked in by immoral
marketing strategies, too (and I think he's quite a bit from being
accurate, here). So I just recommend that you read the XEmacs FAQ
checking for any amount of "advertising lies", hate and spitefulness,
which should be there in abundance, according to Erik. You need not
believe anything in the FAQ, if you don't want to, but you should at
least be able to point out the bile Erik claims. If you don't, think
again. But perhaps this is a marketing ploy again: they don't want
you to think they are the bile and hateful crowd they really are...
In short, just like you should not believe the XEmacs developers (or
disbelieve) without checking up on them, you should do the same with
Erik. Looking through the FAQ is not much work, and will tell you how
much you can rely on Erik's opinion here.
hac...@user1.channel1.com
Erik Naggum <nob...@naggum.no> writes:
> not me. that's the point I'm trying to make here. color is not a
> universal good, the way some people believe it is. I don't to stuff that
> needs color. I do stuff that is _hampered_ by color.
Fine, so turn it off. Choice is the universal good. Many programmers I
have talked to say that their productivity went up once they were able
to use an editor with syntax coloring. It helped in debugging, and in
preventing the bugs in the first place.
> prefer your colorful, contentless, "modern" pages over an information-rich
> "dull" page? if so, go to hell.
You are going to unreasonable extremes to support your point. WWW
pages are content-free because they have color, their content-free
because the people that made them content-free. But there are plenty
of content-free pages with nothing more fancy than a horizontal rule.
> | The Human Computer Interface is the key part of modern systems, XEmacs
> | has its problems and Emacs is not a bad interface but I would say that
> | the XEmacs trend is more in line with what will produce better
> | productivity.
I'd be hard pressed to agree with that one. No one interface is good
for everyone. For instance, an iconic interface totally sucks for the
blind, who need text that can be read to them by the computer.
> | The natural urge of people is to point, the mouse is the interaction tool
> | that allows the user to do this.
>
> do you actually _believe_ this "natural urge" stuff? my "natural urge" (if
> I have any that exceed sex and hunger in complexity) is to speak or write.
Oh please, tell me you've never pointed at something behind a glass
case and said "That one, please." Tell me you don't gesture while
talking.
A mouse is kind of stupid for a lot of things, to be sure. Menu bars
add questionable benefit to productivity. Never the less, the
combination is incredibly useful for learning about the capabilities
of the program. Ideally one day we will be able to *say* any of:
Move this paragraph here
Move the second paragraph to after the third
Stick line 2 before line 12
Bag the first second and fourth paragraphs, as well as this line
> | A secretary, a physist, a chemist and an English student have no need to
> | know about the complexities of using a computer system to be able to use
> | what is an editor.
> your saying so does not make this true, no matter how hard you try to make
> it so. the fact is that this _stupid_ line of argumentation has kept
> users that I find myself speechless when faced with the incredible stance
> "we know what's good for you, just you sit still until we fix it for you".
Did someone say this? You didn't quote anyone saying this.
Most of these folks don't need to know about the complexities of the
computer system. They want to get a paper out. This should involve
starting the program, typing, occasionally changing the format, and
then printing.
Unfortunately, most of the time what they really do is something like
this:
Start the program. Discover that for whatever reason, the program
won't start. Call tech support. Tech support works with them for
several hours. Finally it starts after 5 reboots. They type for a
while. Try to make a header bold and larger. The computer prints
really ugly characters. Call tech support, Oh, I need ATM, or a new
font.
Curse, curse, curse.
The fact is, despite great advances, computers a still a pain to use
for most folks. Hell, they are a pain to use for experienced folks,
just ask any admin who has had to try and switch from SunOS to
Solaris.
One "innovation" I could do without: Icon palates. These generally are
not very clear. Case in point: On the MetroWerks IDE for Macintosh, no
one could identify the icon for Save As. For whatever reason it was a
lifesaver. Why? Who knows.
> you live in your graphical user interface world with colors and idiots for
> users. I prefer people who can communicate and think in words. I have
> that feeling carries over into computing. pointing is for people who have
> yet to discover _thought_. in my view. you obviously disagree, but you
> people who feel _disenfranchised_ by the now point-and-click "interaction"
> you want to make the universal mode of communication with a computer.
Oh yes, that is a great demonstration of your open mindedness.
> this has nothing to do with any Emacsen, anymore, but I just want you (and
> others) to know your "natural urge" nonsense is disputed by _individuals_
> who don't actually _like_ your "interaction for dummies". and for those
> who are inclined to point and not think: this is an example of how one
Wow, what a nice prejudice against the visually oriented. It may come
as a shock to you, but some of the greatest minds in history and in
computing, who are fantastic writers and thinkers, have been known to
prefer to point at things instead of describing them. That way, they
don't have to waste time formulating the words of how to describe a
paragraph, and can instead concentrate of the contents of the
paragraph as a set of concepts.
So, nyah. :-)
> a word says more than a thousand images.
Which one?
And the funny thing is, I like Gnu emacs better, mainly because I have
been using it for 10 years.
--
-####------------> Nipple!, Is qui iacit in hamas marsupiales. | Melior
#### Rev. Irreverend Hacksaw, Omnibenevalent Polyparrot (ULC) | amans
#### http://www.channel1.com/users/hacksaw/ | per
#### <-- Tartan of the ScotchBrite Masons (Are you two of us?) | chemia
Erik Naggum
* <hac...@user1.channel1.com>
| Many programmers I have talked to say that their productivity went up
| once they were able to use an editor with syntax coloring.
you missed the point entirely. various people say that "color improves
programmer productivity", "color and graphics help people communicate with
computers", and other _universal_ statements about _all_ users. I object
to this because it is _not_ universalizable, and the programmers and other
users who do _not_ improve their productivity with _lots_ of irrelevant
sensory input, new and application-specific symbolic languages (icons),
etc, should not be sacrificed to the "let's pretend all people are the
same" that greatly improves the profit margin of companies who can sell
billions of copies of the same program all over the globe, instead of
having the machine adapt to the user. whether we are ruled by a program
that a majority or a minority of users object to is irrelevant -- the fact
remains tha programs should not rule people. a pretty dictator is no
better than an ugly one as far as freedom is concerned.
I'm reminded of the old philosophical discussion about swans. "all swans
are white", was the proposition after someone had only seen white swans.
"no, that's not true", would be a legitimate objection by one who had seen
a black swan. if, at this point, some _idiot_ should object with "are you
saying that NO swan is white!?", all hope of resolution would evaporate.
and that is _exactly_ what it is like to tell people that "no, not all
people find color and graphics and lots of icons and mouse clicking to be
beneficial".
the point in this whole exercise to make people aware that people are not
all exactly the same. individual preferences that may account for less
than 10% of the population should not be ruled out with "sorry, you can't
use a modern computer as efficient as the next guy, so you'll be unemployed
in our modern future". computers should not demand a particular kind of
people no more than any other _universal_ tool should.
to those who have asked: no, I'm not blind, colorblind, or hard of seeing.
if, however, I were, I would object even _more_ strongly to the claims
(effectively) made by soi disant user interface designers that people with
visual challenges have nothing to do in the modern world. this is the
first time in history when a blind person's intellect can come to full use
without regard to the disability. however, GUI designers, WWW designers,
etc, are actively at work to reduce their chances to succeed and to utilize
their talents so much that they are left behind _completely_, and their
situation is much worsened by those _modern_ "user interfaces".
the user interface should be an issue between the individual user and his
computer (i.e., _total_ customizability), _not_ an issue between a vendor
and the most profitable _class_ of users, at the expense of all other
users. optimize too much for X, and you lose the necessary flexibility.
due to the Emacs/XEmacs split on the user interface, and the _stupid_ need
for the splitters to break everything else in the process, we have yet a
long way to go to abstract out the user interface so mouse-clicking users
can benefit as much as typing users can.
five years ago or so, I wrote an article to comp.lang.c about the _massive_
difference between "user input" and "program input". I subsequently lost
the article, but it seems even more important today: I would like to see
all applications talk in a language reasonable for program input, and then
build user interface who talk that language to the application on the one
hand and does its thing towards the user on the other.
I already said in my first message in this thread that XEmacs is more
suited to people who like graphics and such, and David Hughes found that
insulting, but my gripe with XEmacs is not about the user interface (about
which I could hardly care less, as long as I can choose something else),
but about the careless breaking of the rest of Emacs. _that_ is the
extremely annoying aspect to XEmacs. if they had done it right, we could
have had one Emacs with two different user interfaces (and then more would
be easy to add). however, they did it wrong, and we now have two Emacsen
where the rest of Emacs is also different. how _incredibly_ stupid!
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> learning _languages_ (case in point: XEmacs calls its Lisp an "API", Emacs
> calls its Lisp a programming _language_). David Hughes felt _insulted_ by
Nonsense, Erik, nonsense. This is from Lispref introduction:
Most of the XEmacs text editor is written in the programming
language called XEmacs Lisp. You can write new code in XEmacs Lisp and
install it as an extension to the editor. However, XEmacs Lisp is more
than a mere "extension language"; it is a full computer programming
language in its own right. You can use it as you would any other
programming language.
To quote your own words, Erik:
"just quit lying, and there will no longer be a need to correct your lies.
it's as simple as that." (message-id <30562088...@naggum.no>)
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
The end of the world is coming... SAVE YOUR BUFFERS!
Erik Naggum
* Hrvoje Niksic
| Nonsense, Erik, nonsense.
is it safe to assume that this is the most important point you could bring
up at this time? is it safe to assume that you have found nothing else to
argue about? if so, that's pretty good, don't you think?
please watch Emacs and XEmacs developers talk about Lisp. I have. if they
talk about an "API", they have consistently been XEmacs developers. fact
is, XEmacs cares about internal implementation details (abstract/opaque
data types) and interfaces to them -- that's why they had to break with the
way Emacs did it to implement their "improved" version. fact is, Emacs is
perfectly content to provide a C stratum for Emacs Lisp to run on, and to
_think_ in Lisp. in XEmacs, types are i C. in Emacs, types are in Lisp.
(this is a coarse generalization, of course.) in consequence, XEmacs does
sport an "API" to its opaque types, while Emacs does not need to.
_again_, if opaque data types were as hot as they are supposed to be, how
come XEmacs Lisp has no `defstruct' or `defclass'? if modularity and such
is so important, how come XEmacs has not implemented a package system to
distinguish between internal and external symbols in a package, and still
use the prefix-every-name-with-its-package-name deal? the only conclusion
that can be drawn is that encapsulation, abstraction, etc, are _not_ good
ideas in Lisp, but instead are used as a means to make the code harder to
understand and change for programmers that the XEmacs developers don't
trust. (this latter point was made _explicitly_ by Ben Wing earlier.) of
course, if "trust" is an issue, how much can you trust people who willingly
and on purpose breaks existing code because it is in some irrelvant way
"wrong"? I can understand that they don't want this to happen _again_, but
it's ludicrous to argue against what renegade programmers can break when
the dangers they try to fend off are mainly their own creation.
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> * Hrvoje Niksic
> | Nonsense, Erik, nonsense.
> is it safe to assume that this is the most important point you could bring
> up at this time? is it safe to assume that you have found nothing else to
> argue about? if so, that's pretty good, don't you think?
No, Erik, of course not. I have only corrected a most blatant lie.
It is not nitpicking, as you would have it.
<<
another very useful item in judging Emacs vs XEmacs is to look very
carefully at how they approach the Lisp code: in Emacs, it is a real
programming language, and the C is just support code for it, whereas in
XEmacs, you find people talking about the "API" all the time, and they are
>>
As you can see, your mentioning of "API" was within a context, and it
helped to prove a point. The fact of unveiling the lie does not
necessarily make the whole point invalid, but it does degrade the
credibility of the article writer.
What you are doing, Erik, is finding the motes in other people's eyes,
while you don't see the beams in your own eyes. That was the point I
made, and you know it very well.
> please watch Emacs and XEmacs developers talk about Lisp. I have. if they
> talk about an "API", they have consistently been XEmacs developers. fact
I have grepped the XEmacs reference manual for the word "API", and
found it only within the context of Tooltalk support.
> course, if "trust" is an issue, how much can you trust people who willingly
> and on purpose breaks existing code because it is in some irrelvant way
> "wrong"? I can understand that they don't want this to happen _again_, but
The same thing was done by GNU Emacs developers, many times. Look at
relevant posts by Barry Warsav in this thread.
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
main(){printf(&unix["\021%six\012\0"],(unix)["have"]+"fun"-0x60);}
Erik Naggum
* Hrvoje Niksic
| What you are doing, Erik, is finding the motes in other people's eyes,
| while you don't see the beams in your own eyes. That was the point I
| made, and you know it very well.
this is getting to be _really_ stupid. if you wish to disprove my point
that XEmacs developers and users refer to, think about, and interact with
what they themselves call the "API" to C-level internals, you have to show
that this is not the case, not that they do not _always_ do it, which I
have never claimed. since you mentioned Barry Warsaw and "grep", grep his
messages for "API" while you're grep'ing.
you seem to be excessively interested in individual words, instead of
trying to understand what is being said. (just as you obviously find it
more important to count negative words than detect negative meaning, you
use _grep_, of all irrelevant tools, to argue against a distinction in
attitudes towards a programming language.) the very _nature_ of "abstract/
opaque data types" requires "API"s. the _concept_ of an "API" was invented
by the same guys who invented them. now, for primitive types, this is only
ridiculous. for complex constructions, it's _wrong_. e.g., `let' is not
an "API" to the stack-unwinding facility of the Emacs Lisp machine, it's a
binding form. if you think C-level implementation, you will have to think
about `let' as a complicated `unwind-protect' form that resets a variable
to its old value upon exit from the form, regardless of how that exit
should take place. however, that is not a useful way to think about `let'
forms, even if you were implementing them.
likewise, programmers should not think in terms of manipulating _types_,
they should think in terms of getting things done. the type-centric view
of programming is at odds with dynamic types in the first place. in Emacs
Lisp (and other Lisps), you _don't_ know which types your arguments have
until you check them, and you don't check unless you need to. in XEmacs,
there are hundreds of functions that access data structures according to
their type, and _not_ according to the role of the data or the function it
has for that type. this is _exactly_ how one would expect a C++-programmer
to implement types, and very foreign to Lisp programmers. (for a wider
example of the distinction: C/C++ programmers think about object
orientation in terms of objects that "have" functions. in the Common Lisp
Object System, object-orientation is in terms of generic functions whose
behavior is specialized on objects of differing kinds.) I take it that the
XEmacs folks don't appreciate Lisp's dynamic typing, and thus want to
impose the silly notion of static typing on XEmacs, too, mostly because
it's so disgustingly "trendy". again, who _cares_ whether the code inside
is trendy or not, except people with a _marketing_ agenda? it's as if some
marketing "genius" had invented a sticker that said "abstract data types
inside" that would make all the hype-pushers all excited about _nothing_.
| I have grepped the XEmacs reference manual for the word "API", and
| found it only within the context of Tooltalk support.
you missed "annotation API functions" within the context of extents. as
long as you wish to fuss about minutiae, at least get them right.
| > course, if "trust" is an issue, how much can you trust people who willingly
| > and on purpose breaks existing code because it is in some irrelvant way
| > "wrong"? I can understand that they don't want this to happen _again_, but
|
| The same thing was done by GNU Emacs developers, many times. Look at
| relevant posts by Barry Warsav in this thread.
you seem to be utterly unable to _understand_. most of the other people
who continue this thread at least understand what's going on. happily,
those who only count negative words or see only flaming must have utilized
their kill or score files, but _you_ misunderstand and argue completely
irrelevant points, constantly. this leads me to conclude that you have no
substantial disagreement, and only want petty arguments.
Emacs breaks existing code _very_ seldom, and only when there is very good
reason to do so, such as because of a reported bug. whoever reported bugs
in the design of keymaps or the syntax of keybindings? these were changes
made for some misguided "puristic" idea, and broke code BECAUSE IT WAS IN
SOME _IRRELEVANT_ WAY "WRONG". get it? I will argue that every time Emacs
has made a change that was not backwardly compatible, there was a very
specific reason to do it, and most of the time it broke working code, it
was because that working code had taken liberties about internals. (Ben
Wing argued vociferously that people taking advantage of knowledge of
internals is a good reason to make such knowledge hard to obtain. I
maintain that Emacs programmers form a community of people who should let
eachother learn and understand and thus avoid such errors through the good
will of cooperation. if, however, your default mode of thinking is
competition, people will work in isolation and not work together.) lots of
changes were made to XEmacs because it would be easier to program the thing
if things were different, but that's implementation details that should not
affect users, unless, of course, your focus is on APIs instead of
programming languages.
I find it incomprehensible that somebody should at all want to make a whole
lot of fuss about this. the XEmacs marketing includes a lot of hype about
abstract/opaque data types. XEmacs have hundreds of functions that deal
with access to datatypes unique to XEmacs. this _is_ the API concept!
Tom Kludy
Mike Humski (m...@rahul.net) wrote:
: Is emacs inferior to xemacs?
Though I won't even attempt to answer this question, I do wish that both
sides would create fewer incompatibilities with each other. IMHO, _both_
sides are incredibly arrogant. GNU Emacs should recognize that XEmacs
exists; if XEmacs adds a new feature, then if GNU Emacs adds the same
feature, they should do it in a compatible way. OTOH, XEmacs should realize
that their roots are in GNU Emacs, and quit bitching about how RMS doesn't
see things the same way that they do.
What it comes down to is that I, as a user, don't care. I want something
that works, is (at least mostly) bug-free, and runs quickly. (I still find
myself using VI for over 50% of my text-editing needs, even though I _hate_
it and would much rather use Emacs, simply because both Emacsen are too
slow! This on a PPro 200 w/64 Megs!!!)
My vision for the future of both Emacs' is for me to have all lisp files,
and all other support files, identical for both programs, and simply have
the binaries for both programs on my system. I can certainly spare 10
megs extra to have both binaries, but I have a hard time justifying ~50
megs apiece for Emacs and XEmacs.
I say, let both emacs' live. Just make them live in the same house :)
- Tom Kludy (tkl...@engin.umich.edu) | #include <std_dislaimer.h>
If God wanted us to have a President, He would have sent us a candidate.
-- Jerry Dreshfield
Christopher Davis
EN> == Erik Naggum <nob...@naggum.no>
EN> Emacs breaks existing code _very_ seldom, and only when there is very
EN> good reason to do so, such as because of a reported bug.
Why, then, does Emacs reimplement XEmacs's "teach-extended-commands-p" as
"suggest-key-bindings"?
Why did ange-ftp's support for non-UNIX ftp servers disappear?
Why was code that tested system-type potentially broken (temporarily,
since that problem has been fixed) for a couple versions? (The original
change was not "because of a reported bug".)
--
Christopher Davis <c...@kei.com> <URL: http://www.kei.com/homepages/ckd/ >
"I conclude that the CDA is unconstitutional and that the First Amendment
denies Congress the power to regulate protected speech on the Internet."
-- Judge Stewart Dalzell in _ACLU v. Reno_
Erik Naggum
* Christopher Davis
| Why, then, does Emacs reimplement XEmacs's "teach-extended-commands-p" as
| "suggest-key-bindings"?
since this and similar questions are raised quite often by disgruntled
XEmacs workers, I'd like to know why you don't ask the question more
important to one who would like to _understand_ instead of _bitch_: "did
whoever who implemented `suggest-key-bindings' know that the feature
existed in XEmacs under another name?"
`suggest-key-bindings' was implemented 1995-10-02, as per the ChangeLog.
when was `teach-extended-commands-p' implemented? I can't find any
ChangeLogs in the 19.14 distribution. (and whoever decided to abuse the
-p convention for a _variable_?)
in case you wish to have me answer all sorts of weird questions, I have
never made any attempt to speak for or on behalf of the Emacs maintainers
or RMS or the FSF or anything like that -- I clearly don't know everything
you might want to know and my not knowing is not evidence in either
direction.
e.g., I understand that ange-ftp was handled by RMS in a way that made Andy
Norman sorely pissed at him. consequently, efs will not make it into
Emacs. this fact alone tells me that I should be careful not to draw
conclusions before I have heard both sides. in this case, I have not even
heard _one_ of the sides properly. I also don't know anything about the
timing of the supposed disappearance of these features. it seems to me,
however, that this particular argument has been used for at _least_ three
years. its relevance may have expired.
however, is it in fact true that such support has disappeared? where can I
find the source that Andy Norman gave to RMS, and where is the first
version where the support has supposedly disappeared? and what is the
state today?
Hrvoje Niksic
Erik Naggum (nob...@naggum.no) wrote:
> * Christopher Davis
> | Why, then, does Emacs reimplement XEmacs's "teach-extended-commands-p" as
> | "suggest-key-bindings"?
> existed in XEmacs under another name?"
It was/is quite trivial to check (there are NEWS files, and similar).
XEmacs developers regularly check for new features in GNU Emacs, and
do their best to implement them compatibly.
> `suggest-key-bindings' was implemented 1995-10-02, as per the ChangeLog.
> when was `teach-extended-commands-p' implemented? I can't find any
> ChangeLogs in the 19.14 distribution. (and whoever decided to abuse the
> -p convention for a _variable_?)
teach-extended-commands-p was in 19.13, which was released before that
date. It was probably in 19.12 too.
> e.g., I understand that ange-ftp was handled by RMS in a way that made Andy
> Norman sorely pissed at him. consequently, efs will not make it into
> Emacs. this fact alone tells me that I should be careful not to draw
> conclusions before I have heard both sides. in this case, I have not even
Fair enough. I'd like some background on ange-ftp issues too. I
think there was a file telling about this somewhere...
--
Hrvoje Niksic <hni...@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
`VI' - An editor used by those heretics that don't subscribe to
the Emacs religion.
Erik Naggum
this could have been an interesting discussion had it been possible for
Steven Jones to understand what I'm trying to tell him.
* Steve Jones
| Now your being silly, I have not said that _you_ need this or _you_ need
| that, I am saying that for the _majority_ of users a GUI is a preferable
| mode of interaction.
when did the majority receive the right to speak for all? please try to
understand that all I'm trying to do is to communicate to you that _not_
_every_ user agrees with the "majority" view. incidentally, there never
was a vote on this, and if there were, you would find that users buy what
they think is necessary to get the job done, at the lowest cost, and with
the least relevant blemishes and flaws. _millions_ of users have accepted
that they lose work with Microsoft tools. should we then claim that the
"majority" is willing to lose their work, so there is no need to improve
the quality of the products?
however, when you make universal statements like the very next statement:
| Colours shouldn't be stuffed down peoples throats, but colours do help.
then I must again object. the question is "_whom_ do colors help?" for
what _reason_ do they help? _colors_, per se, do not help. colors per se
are just photons in a different hurry.
you don't even seem to _register_ that I don't have a color display to
begin with, so, no, "the links will be in a different color than the
surrounding text" is a false assumption on your part. I seriously wonder
how, in your world of pointing and colors, the idea of "false assumption"
could ever be expressed, but you _assume_ that I have colors even though
I'm actually telling you that in _fact_ I don't. why?
the rest of your message is equally out of touch with what I tried to tell
you: that _not_every_ user will be a new user who does not understand
computers (if you're looking for 70's cliche's, you're a master at using
them -- today's users are far more educated in the workings of a computer
than they were when this "oh, we must protect the ignorant users" argument
was at its peak), that _not_every_ computer will be used by people who are
expected to be unskilled for their job, that _not_every_ programmer has to
work _only_ with the user interface. I'm quite certain that your world
consists only of user interface programmers who deal with stuff where color
helps, but I'm asking you only one thing: to remember, if not understand,
that there are uses for computers where colors do _not_ help, where colors
add to the cognitive load, where lots of gimmickry on the screen _reduce_
productivity, etc. if users matter to you, _I_ will matter to you, and _I_
say, as a long-time computer user, that what you say about all users is
simply _not_true_ of _all_ users. if you ignore this fact, as you have
shown an amazing willingness to do, you are not interested in users, you
are interested in finding users who agree with you, or can be forced to
agree with you, and then sell them things. well, I don't, and I won't,
you mention CLI's and think DOS or Unix (!) is the best CLI to compare
against. it is obviously the best when you wish to let colors and GUIs
win, which is just what they did in the DOS world: because the DOS CLI is
so crummy. Unix came later, not because vendors were slower or it was
harder to do or anything like that, but because users saw no need for it.
if you are honest and actually want a useful comparison, look at TOPS-20's
command interface. yes, TOPS-20 is defunct, but you can read about it, and
the interface lives on in Cisco's routers, in Kermit, in lots of places.
should a command line interface be able to help the user? DOS doesn't
think so. Unix' shells don't think so. I think so. and COMND does.
should a program and the command line interpreter work together? DOS
doesn't think so. Unix' shells don't think so. I think so. and COMND
does. etc.
whether you get a menu by typing `?' or clicking on a mouse is irrelevant.
whether you speak an actual _language_ to the computer _is_ relevant. I
want to use a _language_ to talk to the computer, and I want it to be the
best language that each of us can deal with (i.e., I'm willing to learn
something new in order to be better at my job, and I don't accept the
argument that professionals in any discipline should demand _not_ to learn
to use their tools), such that we achieve maximum information with the
least amount of effort. GUIs lose _big time_ in this regard, except for
some specialized environments, where the information has already been
exchanged, such as in training air traffic controllers. colors may help in
reminding people or stirring them into action. if they were to handle a
new situation, or get new regulations or whatever, I hope they can still
get information in a _language_.
take syntax highlighting -- it wouldn't help if the user doesn't know the
language and the meaning of the colors, it confuses users greatly when
there are mistakes in it, it's very CPU intensive, and it is very wasteful
for people who don't make syntax mistakes in the first place. in other
words, it's a beginner's support tool, and that beginner will probably
never acquire the proficiency in the language that another beginner would
that did not have such "crutches" to lean on. but more beginners who don't
know what they're doing means more "jobs" and those of us who earn more on
a two-month project than they do slaving all year round, can reap even
higher rates when there are fewer "competitors" at our level.
| This is just an editor, the war you seem to be wageing against GUIs and
| useability is over. The majority of people like GUIs, they don't like
| CLIs.
I want the _choice_ of a non-mouse-infested environment, I want the
_choice_ of a command line, I want the _choice_ of a high-contrast screen
like a black-and-white screen. _YOU_ want to deprive me of those choices.
_YOU_ want to take away what makes _me_ productive, and you think you have
the right to do it, don't you? who do you think is waging a war? YOU are,
against the productivity of people who don't agree with you.
wars have _casualties_. democracy was invented to stop the majority from
completely destroying the minority. wars takes the shape of _ethnic
cleansing_, and you, Steven Jones, want to rid the world of all the users
who either cannot or will not use your colorful GUIs. I want the _freedom_
to choose the interface between me and the program. you want to deprive me
of that freedom. who do you think is waging a war here?
I respect the fact that _some_ people like colors and GUIs and a mouse as
an input device, for _some_ tasks. I represent the fact that _some_ people
do _not_ like colors or GUIs or a mouse as an input device for _their_
tasks. I want choice. you want a single way all users should interact
with the computer, just like that William H. Gates III fellow. you can
keep your colors and GUIs and whatever you want, but you can be dead
certain that if you wish to exclude people based on handicaps or wish to
dictate working conditions for your users, you will face very serious
lawsuits. if you want to listen to your users, you listen to users who
have the guts to stand up to "trends" and hyped-up nonsense, not just the
users who have been beaten to death by your bludgeoning arguments and can
only wimper about _which_ color they prefer.
but you don't even understand what I'm objecting to, do you, Steven Jones?
you think I'm objecting to the use of colors in Air Traffic Control, don't
you? you don't understand that I'm not working as an Air Traffic
Controller, but as an author, a programmer, a lecturer, and I find myself
significantly reduced by so-called "audio-visual aids", not because it is
not well made when I make them, not because it is not easier to look better
and more "professional" with it, not because it might well help a student
or reader or two get the point quicker, but because I spend _five times_ as
much time preparing material for those "multimedia shows" as I do with a
blackboard and chalk, and it costs a lot of _real_ money to make them. on
top of this, I get through less material in an hour, because people don't
_understand_ from figures, they have to be guided to understand _both_ the
figures _and_ what they represent. _after_ they have understood, a figure
can indeed be benefitial, but _before_ they know what to understand, a
figure is no of value whatsoever. it can be pretty, it can be redrawn a
lot easier by somebody who does _not_ understand, but understanding does
not come from figures.
it's harder to argue against a figure than a real argument. it's harder to
argue against a color than what it represents. for people who wish to
manipulate people, colors and figures are _the_ choice. for people who
wish to understand and study, _language_ is the choice.
the same is true for your ATC users. they wouldn't know what to do with a
white line or yellow line or red dot or whatever, _unless_ they had very
specific training in their jobs. case in point: I sat down in front of a
"modern" computer the other day and watched somebody else work. he had to
click on an amazing number of buttons with meaningless, itty-bitty "icons"
on them to accomplish a very simple task. I found myself frustrated and
impatient, because I was wasting a lot of time (and their money), on one of
those "user friendly" systems. as a counter-example, I had the pleasure of
dealing with secretaries as Oslo Stock Exchange who used Emacs (Gosling, I
think) on PRIME systems a few years ago. they wrote their own Lisp macros,
in fact, since the keyboard macro facility produced Lisp code for them to
edit. they were extremely efficient at everything they did. one time,
we're talking the early 90's here, they had to use a temp, one of those who
were efficient at "modern" computers. she got her own "modern" computer,
since she was unable to use the Prime systems, but still turned out to be
about one third as productive as the others, although she appeared to be
_working_ at least as hard, maybe harder, since she became frustrated that
her tools were not up to the speed that the other secretaries used.
I note in passing that you make a tremendous number of simple errors in
your writing. "your" instead of "you're", etc. English is not my native
language, but "Steven Jones" sounds English-like, even though you're in
France. I find it alarming that language skills are dropping so fast among
users of graphic user interfaces, and I'm wary of programmers who make a
lot of mistakes of the "lack of precision" kind in their other dealings.
Steve Jones - JON
Erik Naggum wrote:
>
> this could have been an interesting discussion had it been possible for
> Steven Jones to understand what I'm trying to tell him.
>
> * Steve Jones
> | Now your being silly, I have not said that _you_ need this or _you_ need
> | that, I am saying that for the _majority_ of users a GUI is a preferable
> | mode of interaction.
>
> when did the majority receive the right to speak for all? please try to
> understand that all I'm trying to do is to communicate to you that _not_
> _every_ user agrees with the "majority" view. incidentally, there never
> was a vote on this, and if there were, you would find that users buy what
> they think is necessary to get the job done, at the lowest cost, and with
> the least relevant blemishes and flaws. _millions_ of users have accepted
> that they lose work with Microsoft tools. should we then claim that the
> "majority" is willing to lose their work, so there is no need to improve
> the quality of the products?
>
People have paid up front my the million for Windows95, I don't think it
is a good product, but it is a better product than 3.1 and a much better
product than DOS. The interaction in 95 is simpler cleaner and quicker.
The old "Money talks bullshit walks" expression comes to mind. 95 is
still poor in comparison but better than the previous version and the
major + is the new interface.
> however, when you make universal statements like the very next statement:
>
> | Colours shouldn't be stuffed down peoples throats, but colours do help.
>
> then I must again object. the question is "_whom_ do colors help?" for
> what _reason_ do they help? _colors_, per se, do not help. colors per se
> are just photons in a different hurry.
>
They help people who see in colour. If colours have no information why
are Wasps and Bees the same colour while beetles are black. Why are
poisoned arrow frogs brightly coloured ? These things do not have a big
danger sign on them in English, but they do have the colours for
"Danger" all overthem.
If you want to see what colours mean etc, have a word with the Royal
College of Art who did the study on colour perception for the UKs new
ATC centre, they studied people, perception and vision and came up with
a complex set of rules and colours to best represent the data. Of course
they would be wrong as well according to you.
> you don't even seem to _register_ that I don't have a color display to
> begin with, so, no, "the links will be in a different color than the
> surrounding text" is a false assumption on your part. I seriously wonder
> how, in your world of pointing and colors, the idea of "false assumption"
> could ever be expressed, but you _assume_ that I have colors even though
> I'm actually telling you that in _fact_ I don't. why?
>
I've never said you have a colour display. Although yes I think that is
a bit strange, but each to their own. I don't claim that every single
person on the planet is the same that all want to use a computer this
way. I will have ever stick my neck out and say that people in general
want a colour display, and that people in general prefer a GUI to a CLI.
Why should the majority be told they are "wrong" because you would
prefer the old method of interaction. Do you still use a VT220 ?
> the rest of your message is equally out of touch with what I tried to tell
> you: that _not_every_ user will be a new user who does not understand
> computers (if you're looking for 70's cliche's, you're a master at using
> them -- today's users are far more educated in the workings of a computer
> than they were when this "oh, we must protect the ignorant users" argument
> was at its peak), that _not_every_ computer will be used by people who are
> expected to be unskilled for their job, that _not_every_ programmer has to
> work _only_ with the user interface. I'm quite certain that your world
> consists only of user interface programmers who deal with stuff where color
> helps, but I'm asking you only one thing: to remember, if not understand,
> that there are uses for computers where colors do _not_ help, where colors
> add to the cognitive load, where lots of gimmickry on the screen _reduce_
> productivity, etc. if users matter to you, _I_ will matter to you, and _I_
> say, as a long-time computer user, that what you say about all users is
> simply _not_true_ of _all_ users. if you ignore this fact, as you have
> shown an amazing willingness to do, you are not interested in users, you
> are interested in finding users who agree with you, or can be forced to
> agree with you, and then sell them things. well, I don't, and I won't,
>
Prat. No when I say "the majority of users" this does not mean "all
users" when I look into what is required for a specific task for a
certain enviroment I care only about a limited set of users. To say
that I am not interested in the User is foolish, that is all I give a
shit about. The interface is there to provide the best representation
to the User of what they need. There is NO WAY that a modern ATC system
would be conceived of without the use of a complex GUI. This is done to
improve the controllers lot, reduce their work load and to alert them to
problems quicker. You CANNOT do this as quickly without a GUI. In fact
without a GUI you can't show a Radar display, with a modern GUI you
don't have to show just dots on the screen you can show more text,
present information and alerts by using colours.
We don't protect ignorant users, we allow people to become users without
the crap. Why do you need to be able to write a driver for a HD to
install one, the nightmare that used to exist on lots of platforms is
slowly disappearing. You can use the computer as a tool, just like all
the other tools you use. Users aren't stupid, in many specialised cases
they know a damned sight more about the task than the people programming
it. To say that a GUI is provided for an ignorant user is rubbish.
Jobs and Wosniak weren't morons and yet they created a GUI and used a
GUI. The people at MIT are certainly not morons and yet they support and
use X like most of the Universities. GUIs give more power and more
control to the user.
> you mention CLI's and think DOS or Unix (!) is the best CLI to compare
> against. it is obviously the best when you wish to let colors and GUIs
> win, which is just what they did in the DOS world: because the DOS CLI is
> so crummy. Unix came later, not because vendors were slower or it was
> harder to do or anything like that, but because users saw no need for it.
UNIX came after DOS ? Ummm check your history books I think... wasn't
UNIX an early 70s product ?
> if you are honest and actually want a useful comparison, look at TOPS-20's
> command interface. yes, TOPS-20 is defunct, but you can read about it, and
> the interface lives on in Cisco's routers, in Kermit, in lots of places.
> should a command line interface be able to help the user? DOS doesn't
> think so. Unix' shells don't think so. I think so. and COMND does.
> should a program and the command line interpreter work together? DOS
> doesn't think so. Unix' shells don't think so. I think so. and COMND
> does. etc.
>
You can actually do lots of interaction from the Unix Shell to running
programs. The CM system we use here has a CLI and a GUI interface,
everything is possible from both sides, UNIX is very powerful from this
point of view.
> whether you get a menu by typing `?' or clicking on a mouse is irrelevant.
> whether you speak an actual _language_ to the computer _is_ relevant. I
> want to use a _language_ to talk to the computer, and I want it to be the
> best language that each of us can deal with (i.e., I'm willing to learn
> something new in order to be better at my job, and I don't accept the
> argument that professionals in any discipline should demand _not_ to learn
> to use their tools), such that we achieve maximum information with the
> least amount of effort.
Learning to use a tool and using how the tool works are very different
things. there is no need to learn how a computer works to use an editor.
Your job is computers so you need to learn about them and adapt
(something you don't appear to be doing). A scientist from another area
doesn't need to understand the assmbler a Fortran program will create
when doing his FFTs and these days he won't even need Fortran he buys
his off the shelf package that allows him to do his job better
(something you nor I could do). Accountants don't need to understand
spreadsheets to use one, but here is a case of a HUGE increase in
performance for a worker who previously spent hours doing the
calculations by hand. A computer is better at that job than they are,
and providing them with a quicker interface for data entry (a GUI)
simplifies their job further so they can care about other tasks and also
save the company money.
> GUIs lose _big time_ in this regard, except for
> some specialized environments, where the information has already been
> exchanged, such as in training air traffic controllers. colors may help in
> reminding people or stirring them into action. if they were to handle a
> new situation, or get new regulations or whatever, I hope they can still
> get information in a _language_.
>
RUBBISH. A live ATC enviroment will use the same colours/interface as
the simulator, thats the point of a trainer, it should be the same as
the real thing. When new regulations came in their would be a change to
the interface and they would be trained up on it. They would be told of
the new regs and they would probably see this on paper, but operationaly
it means nothing, there is no "Help" button that desribes the current
action in text.
If you want a general advantage of a GUI take the games industry, here
is a totally client driven industry that seems to belive not just in not
CLI but in 3/4D interaction and colours by the bucket load. Is Doom
better than the Origonal Adventure (on UNIX enviroments) ?
> take syntax highlighting -- it wouldn't help if the user doesn't know the
> language and the meaning of the colors, it confuses users greatly when
> there are mistakes in it, it's very CPU intensive, and it is very wasteful
> for people who don't make syntax mistakes in the first place. in other
> words, it's a beginner's support tool, and that beginner will probably
> never acquire the proficiency in the language that another beginner would
> that did not have such "crutches" to lean on. but more beginners who don't
> know what they're doing means more "jobs" and those of us who earn more on
> a two-month project than they do slaving all year round, can reap even
> higher rates when there are fewer "competitors" at our level.
CPU cycles mean less and less each year. This years machines are the
supercomputers of 10 years ago (almost). As you appear to be perfect in
never doing a typo or failing to close brackets or missing a trailing
comment in C you don't need these things. For those of use who type at
more than a char per min and will be prone to the odd error that can be
picked up by syntax highlighting (which is less CPU intensive than
compiling) it adds quite alot. It is not a beginners tool, it does not
mean the user does not know the language, it is not a "crutch", it is a
tool pure and simple. Hey so is LISP or C if you don't need tools like
compilers why not just start coding in HEX ? There are alot of people
out there earning quite high rates who use tools to aid them rather than
seeing new developments as "unpure" and against the will of the great
VT220. This is clearly just an attempt to insult my abilities, I have no
worries on that front, you clearly will never be competing in the same
job market as I will, until yours dies that is.
>
> | This is just an editor, the war you seem to be wageing against GUIs and
> | useability is over. The majority of people like GUIs, they don't like
> | CLIs.
>
> I want the _choice_ of a non-mouse-infested environment, I want the
> _choice_ of a command line, I want the _choice_ of a high-contrast screen
> like a black-and-white screen. _YOU_ want to deprive me of those choices.
> _YOU_ want to take away what makes _me_ productive, and you think you have
> the right to do it, don't you? who do you think is waging a war? YOU are,
> against the productivity of people who don't agree with you.
>
Where did I say I wanted this of you ? I can say that the majority of
users want colour and want a mouse. You can stick with todays technoloy
for as long as you want (or until your service agreement gives up) new
interfaces are there as an improvement for the majority. More money
gets put into research to help the majority and so the developments suit
them.
> wars have _casualties_. democracy was invented to stop the majority from
> completely destroying the minority. wars takes the shape of _ethnic
> cleansing_, and you, Steven Jones, want to rid the world of all the users
> who either cannot or will not use your colorful GUIs. I want the _freedom_
> to choose the interface between me and the program. you want to deprive me
> of that freedom. who do you think is waging a war here?
>
Moron. My ideal is that everyone should be able to use a computer for
every task that a computer can do, without ever having to pick up a
manual. A computerised cook-book in the kitchen, a computerised voice
activated lighting system. Computer links for sound and vision to your
friends. All the libraries of the world at you fingertips. The Media
YOU want when you want it and how you want it. More power to the user
not less. GUIs don't take away power (using Motif/X you can code it to
have a CLI if you so wish) they give more power. The power user can do
more in a GUI than in a CLI just because it is a GUI. Do you use X on
your machine or any windowing enviroment ? Is that better than a Vt220 ?
> [..snip..]
> I want choice. you want a single way all users should interact
Which is what I want, the right for people not to be limted to the 70s
and to be able to use the computer without learning bizare languages
like sed and awk like I had to.
> with the computer, just like that William H. Gates III fellow. you can
> keep your colors and GUIs and whatever you want, but you can be dead
> certain that if you wish to exclude people based on handicaps or wish to
> dictate working conditions for your users, you will face very serious
> lawsuits. if you want to listen to your users, you listen to users who
> have the guts to stand up to "trends" and hyped-up nonsense, not just the
> users who have been beaten to death by your bludgeoning arguments and can
> only wimper about _which_ color they prefer.
>
Now thats offensive. We are talking here about the majority of users.
For impaired users the questions are different, for a paraplegic user a
touch screen GUI has helped alot, for the Blind there is research going
on into how to covert flat 2D images into 3D images on the fly. These
are different issues that need to be addressed as well. I do not
advocate having a picture only page on the WWW as this does limit the
interaction from people using braille terminals. These developments
will come, and you can bet money they won't be limited to a CLI.
> but you don't even understand what I'm objecting to, do you, Steven Jones?
> you think I'm objecting to the use of colors in Air Traffic Control, don't
> you? you don't understand that I'm not working as an Air Traffic
> Controller, but as an author, a programmer, a lecturer, and I find myself
> significantly reduced by so-called "audio-visual aids", not because it is
> not well made when I make them, not because it is not easier to look better
> and more "professional" with it, not because it might well help a student
> or reader or two get the point quicker, but because I spend _five times_ as
> much time preparing material for those "multimedia shows" as I do with a
> blackboard and chalk, and it costs a lot of _real_ money to make them. on
> top of this, I get through less material in an hour, because people don't
> _understand_ from figures, they have to be guided to understand _both_ the
> figures _and_ what they represent. _after_ they have understood, a figure
> can indeed be benefitial, but _before_ they know what to understand, a
> figure is no of value whatsoever. it can be pretty, it can be redrawn a
> lot easier by somebody who does _not_ understand, but understanding does
> not come from figures.
>
Whereas I working as a Software engineer find the opposite. My Father
who lectures in Computing finds the opposite, people who give
presentations as a living find the opposite. I can knock up a
presentation at the same pace I can write it, you appear to be using the
wrong tool, or maybe in the wrong way.
> it's harder to argue against a figure than a real argument. it's harder to
> argue against a color than what it represents. for people who wish to
> manipulate people, colors and figures are _the_ choice. for people who
> wish to understand and study, _language_ is the choice.
>
It is hard to argue against images. Someone says "I never did it Gov"
the prosecution shows a picture of him doing it. Horses for courses,
why not just admit that there is power in GUIs that pictures do have
information just from being a picture.
> [.. examples of GUI v not ..]
>
Yes and on the other hand .... we can all come up with these sort of
examples, and the user who clicked on lots of buttons to do a simple
task was using a bad interface. This is not because it was a GUI this
is because it was bad. The one thing about GUIs that is truly awful is
that people code for them that do not understand the biology and
psychology behind them the "I like it that way" crew are the problem not
the GUI. The GUI gives them too much power, with a CLI they wouldn't
have this power. This can then be moved to the User, a well defined
system will allow the user to all that they did do and more, where
before something was mapped to F7 it should still be possible to do it
in F7 but it should also be possible to drag a selected region and drop
it straight to the printer.
There are more tools these days and until programmers realise that they
can't just "do it my way" any more there will be problems. On the
bright side this is beginning to happen, research is going on into what
is good and what is bad. For the poweruser you should be able to change
everything, for the novice give them the best default.
> I note in passing that you make a tremendous number of simple errors in
> your writing. "your" instead of "you're", etc. English is not my native
> language, but "Steven Jones" sounds English-like, even though you're in
> France. I find it alarming that language skills are dropping so fast among
> users of graphic user interfaces, and I'm wary of programmers who make a
> lot of mistakes of the "lack of precision" kind in their other dealings.
Nice snipe :-) Yes my written english is awful, my writing is worse, of
course you don't have dyslexia do you ? I have never created a problem
that took more than a day to fix, how many others can say the same ?
Steve Jones
Erik Naggum
* Hrvoje Niksic
| It was/is quite trivial to check (there are NEWS files, and similar).
| XEmacs developers regularly check for new features in GNU Emacs, and
| do their best to implement them compatibly.
so you assume guilt on your opponent's part because you think he neglects
out of ill will to do something you _have_ to do because _you_ decided to
split out to begin with? such is not justice.
I am left to assume that you have no _knowledge_ of that which could be the
basis for your accusations. I assume further that the accusations are
baseless and groundless and that similar accusations will be of similarly
little value if the first question is not asked, namely "how do you know it
was intentional?". remember, XEmacs ceased to _exist_ as an input source
after what the XEmacs folks did to some Emacs folks. you complain about it
when it suits you and ignore the fact when that suits you. do you know why
I don't trust XEmacs folks? I don't know when they are honest and when
they are dishonest. I don't know when what they say is propaganda for the
masses (e.g., marketing) or developer-to-developer "internal information".
in the light of the complaints about how Emacs does not adopt new stuff in
XEmacs, I'd like to see the date IN THE PAST, for the implementation of
char-tables and bool-vectors in XEmacs. they were added as basic types to
Emacs 1995-09-22. what's your explanation for _those_ not being in 19.14?
| teach-extended-commands-p was in 19.13, which was released before that
| date. It was probably in 19.12 too.
you mean, like, there _are_ no ChangeLogs in XEmacs? why can't you _know_
the _exact_ date it was added, like I can by reading the Emacs ChangeLogs?
I found it incongruous that XEmacs came without ChangeLogs. I'm inclined
to think that this is because they wish to hide information.
Erik Naggum
* Holger Schauer
| But to get back to Emacs/XEmacs discussion: neither does a good job at
| adapting to the user - this is left to him.
how _could_ a program adapt to the user without the user communicating his
wishes to the program? of course customization it is left to the user.
anything else means the programmer and the user are the same person.
those who don't want to do the work themselves should read the SERVICE file
and find somebody who can help fix their problems. for an experienced
Emacs hacker, most customization questions can be answered in seconds.
| "They did it wrong" "You did it wrong" I don't care who did what wrong.
| Obviously neither side WANTS to work with the other because of "You did
| it wrong" "No, you did" and "You insulted me" "You did first". And this
| is the incredibly stupid reason for two different versions, isn't it ?
no, that is not the reason for the two different versions.
the way that they (XEmacs) did things _after_ they had decided to split
(for entirely different reasons) was sometimes wrong and just plain stupid,
so the split grew in size. the split didn't _occur_ because of it, a merge
is instead being _blocked_ because of it.
David Kastrup
Steve Jones - JON <Steve...@eurocontrol.fr> writes:
> Erik Naggum wrote:
>
> > I prefer people who can communicate and think in words. I have
> > been to France once (Aix-en-Provence, a lovely little town, and Marseilles,
> > a not so lovely city). I had to point at things because my French wasn't
> > up to speed. I made an intense effort to learn to _speak_ French, and
> > after two weeks, I could speak it well enough to discuss the best packaging
> > to send home a bunch of books I had bought in one of the excellent
> > bookstores in Aix, after having spent a day in delighted (albeit patient)
> > discussion with the bookstore owner. the thrill of being able to speak a
> > new language was just exhilarating. unfortunately, I can't speak French, I
> > can't even write French, anymore, but I can still read French, and I read
> > German, as well. reduced to pointing, I feel like an illiterate
> moron.
[...]
> I don't want point and click to be the only interaction with the
> computer. But I would prefer to be able to click on a connection to a
> friend if they are at their machine and talk to them (using a colour
> camera) than to type.
Even more important: if Erik would have been *prohibited* using hands
and feet and pointing in his interactions with the French, it would
have been much harder for him to learn French. Pointing might be a
primitive mechanism not to be preferred to verbal use, but it is
immensively helpful for *learning* more advanced forms of
communication. A user having pull-down menus with annotated key
bindings will learn to use the proper key sequences to use for typical
tasks much *faster* than a user having only a reference manual. In
addition, the user might learn about commands he did not know before.
This is not about user interface efficiency, it is about *learning*
efficiency. Even if one would agree with Erik that the learned user
would not need any menus and other bric-a-brac a modern user interface
offers to the incredibly stupid and arrogant and whatever not XEmacs
users and other lower life forms, one should try to offer the utmost
possibilities of learning to leave that state of superhumanness.
Pointing devices, menus, click interfaces, visual aids and color to
point out important things to people not yet knowing where to look for
importancies help to sooner leave this state of desolation where the
user gets into panick even for the sort of helpful things that happen
if you type several backspaces.
Christopher Davis
EN> == Erik Naggum <nob...@naggum.no>
EN> I want the _choice_ of a non-mouse-infested environment, I want the
EN> _choice_ of a command line, I want the _choice_ of a high-contrast
EN> screen like a black-and-white screen.
XEmacs works without a mouse, it works from a command line (assuming M-x
counts as one :-), and it works fine on black & white screens.
It even does bold/inverse/underline on TTYs, which (IIRC) Emacs 19.34 does
not yet do.
Holger Schauer
>>"EN" == Erik Naggum schrieb am 09 Nov 1996 13:09:01 +0000:
EN> statements about _all_ users. I object to this because it is
EN> _not_ universalizable, and the programmers and other users who do
EN> _not_ improve their productivity with _lots_ of irrelevant
EN> sensory input, new and application-specific symbolic languages
EN> (icons), etc, should not be sacrificed to the "let's pretend all
EN> people are the same" that greatly improves the profit margin of
EN> companies who can sell billions of copies of the same program all
EN> over the globe, instead of having the machine adapt to the user.
Aha - this sounds different from your previous postings. But to get
back to Emacs/XEmacs discussion: neither does a good job at adapting
to the user - this is left to him. Still my point is that if one WANTS
to have fancy buttons and the like he can have it with XEmacs but he
can't with Gnu Emacs (yeah, I know, "Gnu Emacs could have that, too, if
only the XEmacs ...." sschnnn ... author fallen asleep).
EN> I'm reminded of the old philosophical discussion about swans.
EN> "all swans are white", was the proposition after someone had only
EN> seen white swans. "no, that's not true", would be a legitimate
EN> objection by one who had seen a black swan. if, at this point,
EN> some _idiot_ should object with "are you saying that NO swan is
EN> white!?", all hope of resolution would evaporate. and that is
EN> _exactly_ what it is like to tell people that "no, not all people
EN> find color and graphics and lots of icons and mouse clicking to
EN> be beneficial".
Sorry, Erik, but look at your own postings and perhaps you will notice
that you also use the same tone: Steve Jones was the one who said that
one can benefit from "point-and-click" and you told him he could go to
hell if he would force you to adopt his belief (while he at best
modestly tried to convince you).
EN> optimize too much for X, and you lose the necessary flexibility.
Why and in which respect ? Now that XEmacs can be used on ttys I can't
see a loose in flexibility (aside from size and speed - although for
me that's not the big deal).
EN> be easy to add). however, they did it wrong, and we now have two
EN> Emacsen where the rest of Emacs is also different. how
EN> _incredibly_ stupid!
"They did it wrong" "You did it wrong" I don't care who did what
wrong. Obviously neither side WANTS to work with the other because of
"You did it wrong" "No, you did" and "You insulted me" "You did
first". And this is the incredibly stupid reason for two different
versions, isn't it ?
Holger
--
holger_schauer :-
mail_address("Holger....@gmd.de"),
project("BGP-MS/AVANTI, GMD Sankt Augustin, FIT.MMK"),
www_home_page("http://www.uni-koblenz.de/~schauer/index.html").
(^:= A donkey came to my office. It had a theory about people anaphora...
Christopher Davis
EN> == Erik Naggum <nob...@naggum.no>
EN> `suggest-key-bindings' was implemented 1995-10-02, as per the ChangeLog.
EN> when was `teach-extended-commands-p' implemented? I can't find any
EN> ChangeLogs in the 19.14 distribution. (and whoever decided to abuse the
EN> -p convention for a _variable_?)
I don't know who put it in originally, but etc/NEWS does note that it's a
feature that was added in Lucid Emacs 19.9. 19.9 is *very* old, though
(unfortunately) there's no really good way to calibrate the date; it's of
roughly the vintage of VM 5.40 (also included in that release), though.
Linh Dang
>>>> "Rick" == Rick Baartman <baar...@alph04.triumf.ca> writes:
Rick> Naggum sez:
>> a word says more than a thousand images.
Rick> Indeed. This contradiction of a what most people take as a
Rick> truism nicely sums up your philosophical position. All kinds of
For some, a word says more than an icon ...
... but for most, an icon is easier to sell !
L.D.
Rick> things follow from this. DOS is better than Windows, emacs is
Rick> better than Xemacs, ... Fine. I can except that you have this
Rick> point of view and won't endeavour to change it. But keep in
Rick> mind that most people _have_ gone over to using
Rick> GUIs. Therefore, your position that GUI designers are foisting
Rick> this on a public that doesn't need it (if only everyone would
Rick> learn how to communicate with their computers in text) is the
Rick> arrogant and intolerant position. Not the other way around.
Rick> Just a casual observation: Those people using the strongest
Rick> language are the intolerant ones. Check back on this thread and
Rick> see which side is using the strongest language. -- rick
--
=====================================================================
Linh Dang Nortel Technology
Member of Scientific Staff Speech Recognition Software
li...@nortel.ca
=====================================================================
Erik Naggum
* Steve Jones
| If colours have no information why are Wasps and Bees the same colour
| while beetles are black. Why are poisoned arrow frogs brightly coloured
| ? These things do not have a big danger sign on them in English, but
| they do have the colours for "Danger" all overthem.
I can only ask you to please observe that you are taking naturally occuring
things, properties, objects, etc, and transferring any conclusions you make
about them to images and abstract notions on a computer screen. it is not
enough that you show that color is useful for animals (!), or that people
point at physical objects. you must show that the transfer of these
qualities to the representational domain is valid. we may well point when
we speak. from there to moving a "mouse" in order to move a pointer on a
screen, in order to point, is a non-trivial transfer. it needs to be
proved that this transfer is not distorted by other factors. no such proof
has been forthcoming from you, only more examples whose validity has
already been challenged. you don't pay very much attention, do you?
| Nice snipe :-) Yes my written english is awful, my writing is worse, of
| course you don't have dyslexia do you ? I have never created a problem
| that took more than a day to fix, how many others can say the same ?
wow, the dyslexia argument! what an utterly _pathetic_ excuse! "I'm
sorry, sir, I can't write well because my eyesight is such that I can't
actually see the screen two feet in front of me." (that's true of my
eyesight, actually.) I was diagnosed with mild dyslexia when I was a kid.
I worked _very_hard_ to get rid of it. today, I write nearly flawlessly in
two different languages, and well enough to draw compliments from literate
natives in another. there is mounting evidence to suggest that dyslexia is
mainly a consequence of the way written language is taught in our schools.
all evidence suggests that it can be cured.
in fact, the reason researchers now believe dyslexia occurs mainly in boys
is that the way we teach writing, letter by letter, creates a barrier to
the "whole word" approach that children normally would make on their own if
it were not for the fact that some anal-retentive school teachers insist
upon the letters. teaching meaning instead of spelling, building
associations between the word and the meaning, instead of between the word
and its spelling, reinforces the fact that "your" and "you're" are
different _concepts_. the "whole word" approach tries to work in terms of
_images_ of the word, instead of images of their components.
(lack of conceptual thinking _skills_ is a growing problem, but there is an
_inverse_ correlation between dyslexia and conceptual thinking skills, e.g,
the average IQ for dyslexics is well above the average of the population,
and math skills and programming skills are generally dependent on the same
factors that give rise to high IQ scores.)
Erik Naggum
* Holger Schauer
| Emacs (in it's current state) doesn't offer you the possibilities I like
| (!) to have in XEmacs (and please, Erik don't answer: do it for Emacs
| again - I have to get MY work done, not yours. That's why I use XEmacs,
| not Gnu Emacs).
"MY work" vs "your work"?
Emacs and free software in general were supposed to help people work in a
community, where each programmer could create software for himself as well
as others who would like to use it, and they should not have to do the same
work all over again. when you try to make this "my work" vs "your work",
you have pin-pointed the problem that I suggest in the XEmacs camp: that of
conflict, competition, reinvention, and reimplementation, the very
antitheses of free software and cooperation to solve common problems.
| You don't have to understand how an engine works in order to drive your
| car (though it can be of help certainly).
sigh. this is _such_ a bogus argument. I wonder what marketing genius
invented it, and for what purpose. "you don't have to understand how
thermodynamics works in order to dress for the cold."
if you can't drive a car, you're scheduled for heavy fines or jail time if
you try it in regular traffic, even if you don't hurt anybody. if you
can't fix a modern, high-end car's internal computer systems, you're in
good company. somewhere between these two extremes, the need to understand
your equipment takes second place to just driving the damn thing. however,
the need to understand things is stays with you for a very long time.
it is not a sign of professionalism to blame one's tools, but it is even
worse not to know them to the degree where they can be used appropriately.
if you want quality in your work, you learn to use your tools well. there
simply is no other way.
if you are a _consumer_ (you know, the kind of people mass marketing
supposedly works on), there is perhaps legitimacy in the claims about "ease
of use" and such. if you are a professional user, there isn't. just like
programming is hard so using the program can be simpler and safer, a
professional user necessarily has to do a lot of uncomfortable work to get
quality results. now, it should be reasonably comfortable, but you can't
make an omelet without breaking eggs, and you can't produce sleek cars in
open fields, surrounded by dancing young girls and a lovely little brook --
you need a dirty, noisy, dangerous and restricted environment that is
conducive to efficiently producing cars that are clean, silent, and safe,
suitable to the kind of marketing that includes open fields, dancing young
girls and a lovely little brook. it shouldn't be needlessly had, but if
you believe the marketing that is aimed at consumers, and you're a producer
(of anything), you have been _duped_, big time.
know why I hate marketing (and marketing people)? they instill people with
hopes so high and images to beautiful that there is nothing but cynicism
and disillusionment left after they have done their job (i.e., enough
people bought the product). in our marketing economy, it is impossible to
sell a solution to a problem without overstating a hundred times how good
it's going to feel, and without associating some trivial thing like shampoo
with the kind of hair you will never even lay your eyes on. it wasn't the
government that created Newspeak, it was the marketing departments. it is
not a solution to tell people to divide by a hundred to get to the real
message -- it's precisely that which is the problem.
| So, Erik and any other visual-people-haters: if you don't like pictures,
| colors, anything else which people like to use to get THEIR work done,
| don't use them. But stop hindering people from using them (not to
| mention insulting).
do I hate "visual people"? no, I don't. I hate the obnoxious marketing
campaigns and the idiots who claim that _only_ "visual people" should be
allowed to use modern computers. don't you even see the difference?
and just how have I been hindering any people in doing anything? this is
the kind of pungent bovine manure you get from _idiots_ who don't even try
to read what they're commenting on.
Roger Espel Llima
[ comp.unix.admin removed from newsgroups list ]
In article <328735...@eurocontrol.fr>,
Steve Jones - JON <Steve...@eurocontrol.fr> wrote:
>If you want to see what colours mean etc, have a word with the Royal
>College of Art who did the study on colour perception for the UKs new
>ATC centre, they studied people, perception and vision and came up with
>a complex set of rules and colours to best represent the data. Of course
>they would be wrong as well according to you.
Whatever people who have studied color have to find, *I* too find
colored displays (esp. various colors inside the same text, as in syntax
highliting) more of an annoyance than a help.
>way. I will have ever stick my neck out and say that people in general
>want a colour display, and that people in general prefer a GUI to a CLI.
>Why should the majority be told they are "wrong" because you would
>prefer the old method of interaction. Do you still use a VT220 ?
I don't know whether he does, but I know that *I* use Linux consoles in
text mode a lot, and xterms under X quite a lot too, using them as text
consoles. It's simply a matter of efficiency: moving your hand from the
keyboard to the mouse and back is a waste of time and an annoyance, so
for tasks where this is appropriate (and I'm prefectly willing to admit
that air trafic controlling isn't one) you'll be more productive using a
CLI or some other keyboard-based interface. Whatever your majority of
users wants, all I'm saying is that this possibility should still exist
in new systems and new software. After all, on a well designed system,
the two can coexist.
[ stuff about ATCs ]
If I remember right, we were talking about editors to begin with. Guess
what one of the reasons for me to drop emacs/lemacs was? Yep, that it
would open its own window and make me move to it (and yes, I did know
about "-nw", but not being particularily interested in programming my
editor in lisp or reading my mail in it, I went for something that loads
quickly and just does its job efficiently: vi/vim).
>We don't protect ignorant users, we allow people to become users without
>the crap. Why do you need to be able to write a driver for a HD to
>install one, the nightmare that used to exist on lots of platforms is
>slowly disappearing.
Using the keyboard to control an editor is not akin to writing your own
device drivers in assembler, even if you'd like the former to be as much
'a thing of the past' as the latter. The difference is that once you've
learned to use your editor this way, you can be *more* productive than
before.
>You can actually do lots of interaction from the Unix Shell to running
>programs. The CM system we use here has a CLI and a GUI interface,
>everything is possible from both sides, UNIX is very powerful from this
>point of view.
Right.
>For those of use who type at
>more than a char per min and will be prone to the odd error that can be
>picked up by syntax highlighting (which is less CPU intensive than
>compiling) it adds quite alot.
You sound like you know what's good for everyone, and that's what I
don't like. I have no problem with editors offering syntax
highlighting, but I know that *I* find my bugs faster from what the
compiler tells me than by trying to figure out what made my editor
indent the next line weirdly, so *I* pick an editor that doesn't do this
kind of thing. As long as you don't try to tell everyone that such an
editor is inferior and a thing of the past, I have nothing to complain
about.
>Which is what I want, the right for people not to be limted to the 70s
>and to be able to use the computer without learning bizare languages
>like sed and awk like I had to.
But now there's perl to replace them! :-)
>It is hard to argue against images. Someone says "I never did it Gov"
>the prosecution shows a picture of him doing it. Horses for courses,
>why not just admit that there is power in GUIs that pictures do have
>information just from being a picture.
Can't you get it in your head that the two can be appropriate for
different things, *and* that different people can handle one or the
other more easily? There are things for which images are *the* way,
and there are things for which words are *the* way, and there are thing
for which both can work, and I know that *I* will prefer the words. If
you'll prefer the pics I'm not going to take them from you, but don't
try to tell me that the pics are universally better.
-Roger
--
#!/usr/local/bin/perl # e-mail: roger.es...@ens.fr
s**QJwEw=QGv:vt=PGwEvvvEwt=OGUtMGUt=N\@:tKG:A=O> G:GNb,&+ 1%" Z)2" j&zz,+ Y+1*
,s^$^&E["+0,/0%&- [y*-y&\$+!=Pt:S=QtUSQM%11-RGG444F"##F,/\$Gz)2"/&zz,+F%1*)=^,
s@$@Q SF=QSUUt=PSF:t=OSJ>=NSL>PN/,\$"/F"0-")F))&*yX"+0F#/=NtUUGNtUUG=OtGPtG=@,
y,= :-|"-9,\n "-|, ,s,&,\\"\\,g,s,;,/"/,g,s,",===,g,s,(\d),' 'x$1,eg,print $_;
Darin Johnson
>I can only ask you to please observe that you are taking naturally occuring
>things, properties, objects, etc, and transferring any conclusions you make
>about them to images and abstract notions on a computer screen. it is not
>enough that you show that color is useful for animals (!), or that people
>point at physical objects.
I would think such things are obvious. I'm not talking 64K of color,
but even 3 colors, black, white, and red, is a major advance.
Even crusty old MVS made use of colors for decades.
For instance, you scroll through a list of computer events, yawn yawn,
they all look alike, except for different machine names and so forth,
then one jumps out at you in RED, and you *notice* it. this is a highly
appropriate use of color.
In a compiler, if you have colors for various syntactical units,
you can see immediately if you've forgotten to close a comment, and
can see in short order which comment it is.
In short, colors give quick and easy to use information. You can
get the same information through black and white text only, but it's
not as quick since you have to read it. The keyword is "quick".
Such things are important, even if you don't wish to use them
personally. Which is all this is, a silly "I don't us colors, so
support for colors is wasted" argument.
Similar arguments for the mouse, although in this case I see too much
tendency to mouse-only interfaces to the exclusion of all else (you
couldn't have a color-only interface :-). Heck, big red buttons are
nice and convenient, but you can't build your entire interface from
big red buttons.
--
Darin Johnson
da...@connectnet.com
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> e.g., I understand that ange-ftp was handled by RMS in a way that made Andy
> Norman sorely pissed at him. consequently, efs will not make it into
> Emacs. this fact alone tells me that I should be careful not to draw
> conclusions before I have heard both sides. in this case, I have not even
> heard _one_ of the sides properly. I also don't know anything about the
> timing of the supposed disappearance of these features. it seems to me,
> however, that this particular argument has been used for at _least_ three
> years. its relevance may have expired.
>
> however, is it in fact true that such support has disappeared? where can I
> find the source that Andy Norman gave to RMS, and where is the first
> version where the support has supposedly disappeared? and what is the
> state today?
Just what I have gathered from some communications: what RMS did with
ange-ftp while integrating it into Emacs 19 was to throw out all
functionality for accessing non-Unix ftp servers: you cannot use
ange-ftp as included in Emacs 19 to access VMS ftp servers, or several
other brands. The argumentation was, as far as I remember, that RMS
would not keep code in he could not test himself, and anyhow
considered support for non-Unix systems a non-priority.
efs will run with Emacs, but due to the massive changers with bad
consequences for the users RMS did to ange-ftp, Andy Norman is not
likely to sign the copyright to efs over to the FSF, and this probably
will make it unlikely that efs will ever be included in a standard
distribution of Emacs. There are no hostilities involved here, but
just differences in opinion how packages contributed should be treated
to the best of other users.
This is what I gathered from several posts over here, and is just my
personal impression. I hope I have not misrepresented something.
Steve Jones - JON
Erik Naggum wrote:
>
> * Steve Jones <Steve...@eurocontrol.fr>
[.. agreed that colour is not a _universal good_]
>
> | If colour doesn't help why do humans have colour vision? Why are traffic
> | lights Red/Amber/Green what information is inherent in these colours?
>
> if you wish to universalize your argument, you need more than that. a
> simple counterargument: I have over 1800 books, of which 50 or so have
> color illustrations. am I unable to learn from the rest, or enjoy them?
> how come I find the imagery in good books better than the imagery in
> movies? clearly, some people are more verbal than visual. you, like most
> of the people who are unable to deal with natural variation in human
> preferences, want to stuff colors down my throat, and I have to pay for it
> in reduced resolution, more expensive displays, not to mention _larger_
> displays to handle 19" or 21" screen real estate. have you asked me? no,
> you _argue_ against _my_ preferences. your arrogance is unbelievable!
>
Now your being silly, I have not said that _you_ need this or _you_ need
that, I am saying that for the _majority_ of users a GUI is a preferable
mode of interaction. Colours shouldn't be stuffed down peoples throats,
but colours do help. Why are the books you read only in B/W ? Well
because they are printed using a technology that has its foundation over
500 years ago (2,000 if your Chinese). Browse some HTML pages, the
links will be in a different colour/font than the surrounding text.
This gives additional information about the text. I can think of
several examples where a book would be improved by HTML. Emma by Jane
Austen for instance requires several footnotes to inform the reader of
the difference in word and phrase usage between the time the book was
written and the modern day. Several authors are using modern printing
methods to present the information in new ways (eg Irvin Welsh in
Maribou Stork). Colour is one of these new ways.
As an example I'll use the Air Traffic Control world, in which we have
HUGE screen real-estate (2kx2k displays). Tracks are displayed in a
basic format with similar information, an Emergency track is then
coloured Red, a Conflict Track in white (with a line between the two
tracks). Radar Proximity warnings in yellow, point-outs in blue. They
only use a limited number of colours (around 8) but these provide huge
amounts of information to the user that could not be presented as
quickly (and if your plane is hi-jacked you want the controller to know
ASAP) in other ways.
> | For most modern systems the appearance is the content, ...
>
> exactly! is this a good thing? no. this caters to a class of people who
> obviously are unable or unwilling to handle words. with all the lack of
> contents in published material these days, _all_ they have to speak for
> them is appearance. this has been a recognized problem on the Internet
> (WWW) for at least two years, now. do you wish to tell me that I _must_
> prefer your colorful, contentless, "modern" pages over an information-rich
> "dull" page? if so, go to hell.
No this is what I'm saying. Go to the Electronic Telegraph
http://www.telegraph.co.uk they have an excellent Electronic newspaper
which uses colour in a minimal way (eg headlines in yellow main text in
different colours, links in red etc). There are alot of pages that are
rubbish on the WWW and those that don't know how to use colour are the
worst. Colour can however add extra information that words cannot in
that same time/space. Well done it is much better than B/W, badly done
it is worse.
>
> | The Human Computer Interface is the key part of modern systems, XEmacs
> | has its problems and Emacs is not a bad interface but I would say that
> | the XEmacs trend is more in line with what will produce better
> | productivity.
>
> "better productivity"? (whose marketing department are you paid by?)
> that's nothing short of a load of bullshit. better productivity depends on
> the task and on the person. you ignore the _invidual_ completely, to be
> replaced by your obnoxious categorization of what is best for humans as
> such. humans don't exist as such, they exist as individuals. like any
> good dictator, however, some people would like to take away the working
> environments that make "deviants" more productive, in order to make the
> vast masses more productive.
This is a world conspiracy against you personally is it ? Find a
secretary, tell her she has to use an old Typewritter again, that she
can't use Word any more because you say so. The reason the market you
long for is dying is that people with little or no computer knowledge
are being required to use computers every day and they have cash by the
bucket loads. The computer is the most powerful tool of the 20th
century why should this be limited to the "few". If you cannot adapt to
the new world, or cannot see how anyone else can work quicker with a
modern system over on 10 years old I pity you. This is computing, it is
now in a rapid phase of expansion, the challenges are ahead and they are
tougher than those that came behind. Gone are the days when a sad bloke
sat in a room on his own able to code anything to any standard. GUI
engineers must understand how people work, how the majority of people
work, you may not be in the majority and you therefore see the world
going against you, so be it.
<Phew wipes sweat of fingers>
>
> | The natural urge of people is to point, the mouse is the interaction tool
> | that allows the user to do this.
>
> do you actually _believe_ this "natural urge" stuff? my "natural urge" (if
> I have any that exceed sex and hunger in complexity) is to speak or write.
Okay heres a test, someone asks you which box to pick up out of three.
You want the middle one, do you use you had to point while saying "the
middle one" ? Maybe you don't most people will, people speak with their
hands quite a lot.
>
> | A secretary, a physist, a chemist and an English student have no need to
> | know about the complexities of using a computer system to be able to use
> | what is an editor.
>
> your saying so does not make this true, no matter how hard you try to make
> it so. the fact is that this _stupid_ line of argumentation has kept
> computing away from those who could have benefited from it decades ago,
> because what you are _really_ saying is that "you poor idiots, computers
> are still too hard to understand for your puny little brains to be useful
> to you". the _belief_ that computers are any more complex than, say, cars
> and driving in heavy city traffic is so incredibly condescending to those
> users that I find myself speechless when faced with the incredible stance
> "we know what's good for you, just you sit still until we fix it for you".
Do you think that all software engineers need to know about SR,GR how to
dope a transistor, how to mine silicon (or however you get it), how to
distribute computers BEFORE they are allowed to touch one ? Of course
not. A computer for a secretary should be designed for their job not
for ours. It is not saying they are stupid (hell I type at only 40wpm)
or that computers are amazingly complex, it says that here is an
amazingly powerful tool that can help you in your job. Do BIC patronise
everyone of use when we use their pens because they are better than
using a quill pen but we don't know exactly how they make them ?
Anyone should be able to sit down at a Computer and do some basic tasks
(WP, accounts etc) without having to study manuals for days. Why is
this wrong ? This was much less likely with DOS or from the UNIX cli
than it is with modern GUIs. They don't dictate half as much as CLIs
how someone should do a task. Macros in even Word allow you to speed up
your task as you want. Motif allows you to configure just about anything
on the system. I can't see how they limit people, I can see how they
empower them.
>
> you live in your graphical user interface world with colors and idiots for
> users.
Umm ATC controllers are idiots are they ? Of course they are not. I am
an idiot for using a GUI ? Of course not. Jobs and Wosniak were working
with morons too. The majority of people are morons are they ? Step out
of your shell and smell the modern world. The computer will become less
a box and a keyboard and more a standard mode of interaction (eg talking
instead of typing).
> I prefer people who can communicate and think in words. I have
> been to France once (Aix-en-Provence, a lovely little town, and Marseilles,
> a not so lovely city). I had to point at things because my French wasn't
> up to speed. I made an intense effort to learn to _speak_ French, and
> after two weeks, I could speak it well enough to discuss the best packaging
> to send home a bunch of books I had bought in one of the excellent
> bookstores in Aix, after having spent a day in delighted (albeit patient)
> discussion with the bookstore owner. the thrill of being able to speak a
> new language was just exhilarating. unfortunately, I can't speak French, I
> can't even write French, anymore, but I can still read French, and I read
> German, as well. reduced to pointing, I feel like an illiterate moron.
> that feeling carries over into computing. pointing is for people who have
> yet to discover _thought_. in my view. you obviously disagree, but you
> won't see me agree to your "natural urge" bullshit any time sooner than you
> stop universalizing that bullshit to include me and millions of other
> people who feel _disenfranchised_ by the now point-and-click "interaction"
> you want to make the universal mode of communication with a computer.
>
Bollocks, sorry but that is total rubbish. You consider your view as
superiour to all others and will not even consider another point of
view. I don't want point and click to be the only interaction with the
computer. But I would prefer to be able to click on a connection to a
friend if they are at their machine and talk to them (using a colour
camera) than to type. For discussions like this text is more
appropriate as it gives more time to think and deliberate rather than
reducing to a "Does" v "Does not" session. To say that people who point
don't think is small minded to an amazing degree, on a VT220 it wasn't
an option so it wasn't good ? The millions you claim who share your
view that the old ways were best I have not met. But the millions who
use and buy GUI products everyday seem to back me up. There are more
people on Usenet now (for better or worse) than there ever have been, as
a result of the "point-and-click" power of modern interfaces, some of
these have added alot (especially in the science groups).
> this has nothing to do with any Emacsen, anymore, but I just want you (and
> others) to know your "natural urge" nonsense is disputed by _individuals_
> who don't actually _like_ your "interaction for dummies". and for those
> who are inclined to point and not think: this is an example of how one
> person (namely, I) think and have preferences, not a universalizable
> argument about what Emacs users prefer over what XEmacs users prefer.
> however, I'm still inclined to believe that XEmacs users (not the least
> after reviewing the comments I have received), are less verbal and more
> visual than Emacs users, are less interested in studying manuals and
> learning _languages_ (case in point: XEmacs calls its Lisp an "API", Emacs
> calls its Lisp a programming _language_). David Hughes felt _insulted_ by
> this, for God knows what reason. observation it is, and it has been
> confirmed by this debate, and in no way disputed.
>
I am indeed a visual person, although I am realy a visual-audio based
person in that I remember most that I have seen and most that is said.
I have read a huge amount, an editor has nothing to do with this. It
could be said therefore that if you are a typical Emacs user (which I
doubt) then the Emacs user is sitting in about 1978 saying that the
"Star project at Xerox ? Huh nothing will come out of that". This isn't
true, neither is your generalisation of XEmacs users. This is just an
editor, the war you seem to be wageing against GUIs and useability is
over. The majority of people like GUIs, they don't like CLIs. Do you
recommend the pen over a CLI ? Afterall that had a simple interface and
no frills at all.
> a word says more than a thousand images.
Have you seen picture of the Jews during the holocaust ? Or football
fans as their teams scores ? Or of Linford Christie when he crossed the
line to take Gold in Barcelona ? The Mona Lisa ? The Last Supper ? The
starving in Ethopia etc etc etc.
>
> exercises for the visually inclined: illustrate "appreciation", "humor",
> "software", "education", "inalienable rights", "elegance", "fact".
>
This is just stupid, describe the Sistene Chapel, a Rembrant self
portrait, a bird in flight, the sea in a storm with out refering to any
image real or remembered.
And have you ever seen the Far Side cartoons, they sometimes have no
words and yet have lots of "humour". A photo of an audience on its feet
appluading a conductor would be "appreciation", Software is an abstract
concept so is only really a word, a class of pupils studying or
University students graduating is not "education". "Inalienable rights"
could be the Civil war picture of the first black battalion (I forget
the name, Missori 54th ?) walking past some cheering freed slaves.
"Elegance" could be a painting of the 18th Century of some great lady
(the one on the Oxford classics version of Emma would be a prime
example). Fact too is an abstract concept and therefore hard in images.
> #\Erik
> --
> Please address private replies to "erik". Mail to "nobody" is discarded.
RSI here I come.
Steve Jones
Eurocontrol Experimental Centre
Mike Hamrick
Roger Llima wrote:
> Whatever people who have studied color have to find, *I* too find
> colored displays (esp. various colors inside the same text, as in syntax
> highliting) more of an annoyance than a help.
From the Jargon file:
:angry fruit salad: n. A bad visual-interface design that uses too
many colors. (This term derives, of course, from the bizarre
day-glo colors found in canned fruit salad.) Too often one sees
similar effects from interface designers using color window systems
such as X; there is a tendency to create displays that are
flashy and attention-getting but uncomfortable for long-term
use.
A friend of mine, an acomplished C programmer, uses Xemacs with syntax
highlighting. Whenever I stop by his office to check out a peice of
code he's working on I say something like "Mmmm, fruity." This seems
to annoy him. He's pretty much a color freak, his X environment,
which he is constantly dinking with, is 16bpp for a reason -- he uses
most of them. I don't know if this makes him more productive, but it
does seem to make him happy.
Holger Schauer
>>"h" == hacksaw schrieb am 08 Nov 1996 20:52:08 -0500:
h> Erik Naggum <nob...@naggum.no> writes:
>> not me. that's the point I'm trying to make here. color is not a
>> universal good, the way some people believe it is. I don't to
>> stuff that needs color. I do stuff that is _hampered_ by color.
h> Fine, so turn it off. Choice is the universal good.
Sic ! This is the point. Emacs (in it's current state) doesn't offer
you the possibilities I like (!) to have in XEmacs (and please, Erik
don't answer: do it for Emacs again - I have to get MY work done, not
yours. That's why I use XEmacs, not Gnu Emacs). I re-iterate it: if
the folks from the Gnu Emacs team and from XEmacs form up to unite the
two versions I will be happy to be the first one to use it - until
then I will use the one I prefer.
>> | The Human Computer Interface is the key part of modern systems,
>> XEmacs | has its problems and Emacs is not a bad interface but I
>> would say that | the XEmacs trend is more in line with what will
>> produce better | productivity.
h> I'd be hard pressed to agree with that one. No one interface is
h> good for everyone. For instance, an iconic interface totally sucks
h> for the blind, who need text that can be read to them by the
h> computer.
This is certainly correct - but one (the programmer) has to create an
interface which should be usable to most. And if he does his work
reasonably he also enables the user to change the interface to fit his
(the user's) needs. But I, too, believe that most people are helped by
having a visual interface in the first place (and I don't use the
icons in Gnus).
h> A mouse is kind of stupid for a lot of things, to be sure. Menu
h> bars add questionable benefit to productivity. Never the less, the
h> combination is incredibly useful for learning about the
h> capabilities of the program. Ideally one day we will be able to
h> *say* any of:
h> Move this paragraph here
h> Move the second paragraph to after the third
h> Stick line 2 before line 12
h> Bag the first second and fourth paragraphs, as well as this line
Please: don't ! This would only be of use if everyone of us would sit
alone and isolated in his own cell without anybody else trying to get
their work done. Have you sit in a room with 5 other people of which
two are talking while you try to hack some problem into the keyboard ?
>> | A secretary, a physist, a chemist and an English student have no
>> need to | know about the complexities of using a computer system
>> to be able to use | what is an editor.
>> your saying so does not make this true, no matter how hard you try
>> to make it so. the fact is that this _stupid_ line of
>> argumentation has kept users that I find myself speechless when
>> faced with the incredible stance "we know what's good for you,
>> just you sit still until we fix it for you".
This is simply nonsense. Go and ask some physist if he WANTS to know
the internals - in general he doesn't. It is fine that if he WANTS to
gain knowledge of the internals (or complexity or whatever) that he
can do so, but to force him learning about it before he can do
anything will stop anybody from using computers. You don't have to
understand how an engine works in order to drive your car (though it
can be of help certainly).
>> this has nothing to do with any Emacsen, anymore, but I just want
>> you (and others) to know your "natural urge" nonsense is disputed
>> by _individuals_ who don't actually _like_ your "interaction for
>> dummies". and for those who are inclined to point and not think:
>> this is an example of how one
h> Wow, what a nice prejudice against the visually oriented. It may
h> come as a shock to you, but some of the greatest minds in history
h> and in computing, who are fantastic writers and thinkers, have
h> been known to prefer to point at things instead of describing
h> them. That way, they don't have to waste time formulating the
h> words of how to describe a paragraph, and can instead concentrate
h> of the contents of the paragraph as a set of concepts.
I second that. Some say: you can't think of something which you can't
put in words. A lot of pictures and a whole lot more of music can not
be put in to words or if you try it doesn't give you an impression for
real. And believe me, I can think of music and much more and I don't
have to use words for that.
So, Erik and any other visual-people-haters: if you don't like
pictures, colors, anything else which people like to use to get THEIR
work done, don't use them. But stop hindering people from using them
(not to mention insulting).
Holger
Per Abrahamsen
[ Followups redirected to alt.religion.emacs ]
David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
> The argumentation was, as far as I remember, that RMS
> would not keep code in he could not test himself, and anyhow
> considered support for non-Unix systems a non-priority.
No, he wouldn't *port* the part of ange-ftp that he couldn't test
himself from the Emacs 18 overload mechanism to the Emacs 19 hook
mechanism. He did request for volunteers to port the remaining code
soon after the first public release of Emacs 19, but at that point the
people with a direct interest in the code were already too pissed.
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> * Steve Jones
> | Now your being silly, I have not said that _you_ need this or _you_ need
> | that, I am saying that for the _majority_ of users a GUI is a preferable
> | mode of interaction.
>
> when did the majority receive the right to speak for all?
Nobody said that. It is easy to configure a graphical Emac with colour
support in a way as to display no graphic menus and icons, and no
colours. So an Emacs including these features can be configured to
act so that *all* are satisfied with the behaviour it shows for each
of them. An Emacs without this support can't.
When did you receive the right to speak for all?
Erik Naggum
* David Kastrup
| When did you receive the right to speak for all?
*sigh*. I have not spoken for anybody else. pay attention, dammit! I'm
objecting to be told what I should prefer, when I don't. there's a sizable
minority who do not prefer the "modern" user interfaces, but who are being
neglected because obnoxious GUI fanatics such as Steven Jones can't handle
the fact that people are not _identical_ and _must_ deal with things in
_exactly_ the same way, which would have made it possible to make user
interfaces for the "model A human", instead of letting individual users
choose their interface. I don't speak for the minority, either. _I only
represent the fact that they exist. OK?
well, I _implicitly_ speak for all humans when I say: "hold it! we're all
unique! you can't speak about universal preferences." if you object to
this, it's like the scene from Monty Python's Life of Brian where Brian
shouts "you're all unique!" and someone shouts "I'm not!".
as I have said numerous times already, I want programs to be written such
that they speak a reasonably complete _language_ that other programs can
talk, and then there's a program that talks to the user and handles the
user interface aspects. this is in contrast to the user interfaces that
permeate every aspect of a program, as is often the case in "modern"
software. then _more_ people could be productive with their computers,
instead of unproductive with somebody else's idea of what should have made
them productive.
when _did_ you think I was speaking for all, David?
Darin Johnson
>*sigh*. I have not spoken for anybody else. pay attention, dammit! I'm
>objecting to be told what I should prefer, when I don't. there's a sizable
>minority who do not prefer the "modern" user interfaces, but who are being
>neglected because obnoxious GUI fanatics such as Steven Jones can't handle
Look, XEmacs, just like the Emacs it branched from, does not force you
to use colors. In fact, by default I think you don't get syntax
highlighting.
Yes, some packages come up with icons and colors by default. This is
because for the majority of users, this is what is wanted. You
solution seems to be, use no enhancements if at least one user does
not want them. This is absurd. Should GNUS pop up a window asking if
you want enhancements or not the first time it is run? If so, then
you should go write it (thus supporting the free software model of
supplying code rather than whining about it). Or perhaps VM should
start with a disclaimer that "Some people do not want to use icons or
colors, if you wish to use these novice features, you must first learn
to Emacs Lisp, at which point you won't be a nogice so it's all moot
anyway".
You can make XEmacs look identical to GNU Emacs (oops, I meant to say
Emacs) from a user interface perspective.
>well, I _implicitly_ speak for all humans when I say: "hold it! we're all
>unique! you can't speak about universal preferences." if you object to
>this, it's like the scene from Monty Python's Life of Brian where Brian
>shouts "you're all unique!" and someone shouts "I'm not!".
People are not identical, yes, that's true and I stand by that myself
quite strongly. But XEmacs is not like Windows, you can change it
from being novice friendly to being expert friendly. You however, are
wanting to treat everyone identical by denying features that you
personally don't want even though you aren't forced to use them. But
just because everyone is unique doesn't mean that there aren't
features that a majority of users want.
If in the name of uniqueness we reduce Emacs to the lowest common
denominator, we have nothing less. If everyone is a prime number,
there is no lowest common denominator. We might as well not have an
Emacs at all, since someone somewhere may not like its defaults!
That's the reason you don't like XEmacs with icons and colors, because
they're sometimes the default! Get real, and stop with all this
arrogance. XEmacs is not on a holy war against you personally, your
views are supported just fine. XEmacs developers are working hard,
just like Emacs developers, to make an improved product that allows
unique people to use it their own way, and they don't need someone
with who is also working towards the same goal to stab them in the
back. This would be an American soldier attacking the Brits in WWII
instead of going after the Nazi's.
(there I said, Nazi, the thread must end)
--
Darin Johnson
da...@connectnet.com
Mark Hood
Steve Jones - JON <Steve...@eurocontrol.fr> writes:
> If you want a general advantage of a GUI take the games industry, here
> is a totally client driven industry that seems to belive not just in not
> CLI but in 3/4D interaction and colours by the bucket load. Is Doom
> better than the Origonal Adventure (on UNIX enviroments) ?
The best computer game ever written is Nethack, and the preferred interface
is the keyboard and text output. Doom is fun for about 10 minutes; Nethack
is a life-long obsession :)
I have to add myself to the class of users that finds gratuitous colors and
inefficient GUI's evil. Once I've started Emacs up I generally park the
mouse and never touch it again except to click the occasional URL or
hypertext cross-reference (I'm not an anti-GUI purist).
I also seem to be one of the few people I know who doesn't suffer from some
sort of RSI after nearly two decades of programming (knock wood).
-- Mark Hood
Steinar Bang
>>>>> Jamie Zawinski <j...@netscape.com>:
> By the way, the -p convention has been used on variables since the
> dawn of time. TV::WHO-LINE-JUST-COLD-BOOTED-P. ZWEI::*CASE-REPLACE-P*.
> UCL::DWIMIFY-PACKAGE-P. And so on. Where've you been?
And here I've been thinking all this time, that it signified
"predicate"... Silly me!
- Steinar
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> * David Kastrup
> | When did you receive the right to speak for all?
>
> when _did_ you think I was speaking for all, David?
When you answered to the claim that a majority would perhaps prefer to
have the option to use menus and mouses and colours while they are
learning Emacs, and you said (verbatim)
When did the majority receive the right to speak for all?
While you forbid the majority to have a choice, you demand that *you*
not only get a choice, but *just* what you want, and all others should
also get *just* what you want, and not a choice.
--
David Kastrup Phone: +49-234-700-5570
Email: d...@neuroinformatik.ruhr-uni-bochum.de Fax: +49-234-709-4209
Institut f=FCr Neuroinformatik, Universit=E4tsstr. 150, 44780 Bochum, Germa=
ny
David Kastrup
Mark Hood <ho...@eng.sun.com> writes:
> Steve Jones - JON <Steve...@eurocontrol.fr> writes:
> > If you want a general advantage of a GUI take the games industry, here
> > is a totally client driven industry that seems to belive not just in no=
t
> > CLI but in 3/4D interaction and colours by the bucket load. Is Doom
> > better than the Origonal Adventure (on UNIX enviroments) ?
Well, at least watching it being played gets my stomach involved quite
a bit more.
> I have to add myself to the class of users that finds gratuitous colors a=
nd
> inefficient GUI's evil. Once I've started Emacs up I generally park the
> mouse and never touch it again except to click the occasional URL or
> hypertext cross-reference (I'm not an anti-GUI purist).
Yes, but for *learning* users colors, mouses, menus etc. are helpful.
Emacs is near to unusable for newbies, unless they vigorously plan to
invest a lot of time to *first* become an adept of Emacs, *then* a
user of it. One does not get hooked easily to a tool if every time
you try it you totally f*** up. I know a good programmer and computer
scientist who is loath to try again using emacs for exactly that
reason.
> I also seem to be one of the few people I know who doesn't suffer from so=
me
> sort of RSI after nearly two decades of programming (knock wood).
Well, programmers type quite a lot less than average typists.
Sometimes they think. At least some of them do.
Ian G Batten
In article <328735...@eurocontrol.fr>,
Steve Jones - JON <Steve...@eurocontrol.fr> wrote:
> Date: Mon, 11 Nov 1996 15:17:58 +0100
> If you want to see what colours mean etc, have a word with the Royal
> College of Art who did the study on colour perception for the UKs new
> ATC centre, they studied people, perception and vision and came up with
> a complex set of rules and colours to best represent the data. Of course
> they would be wrong as well according to you.
There's a rather heavy irony to people in the ATC trade lecturing the
rest of us on how to design clear and effective interfaces, when it
appears that more than three hundred people died because the self-same
trade can't agree on units to represent height between one airspace and
the next.
ian
Boyd Roberts
In article <328735...@eurocontrol.fr>, Steve...@eurocontrol.fr says...
>
>I will have ever stick my neck out and say that people in general
>want a colour display, and that people in general prefer a GUI to a CLI.
You're probably right, but for the wrong reason. Most people are stupid
enough to choose something that looks good over something that is good.
Those people are just wrong. Those people would build a scientific
discipline based on ascetics, not correctness.
I would argue that a better interface is a more efficient interface,
for input as well as output. A colour display or GUI may look better,
but it does not guarantee a good interface. A good interface is relies
on good design.
--
Boyd Roberts <bo...@france3.fr> N 31 447109 5411310
``Not only is UNIX dead, but it's starting to smell really bad.'' -- rob
Jo Meder
Erik Naggum <nob...@naggum.no> writes:
> * David Kastrup
> | When did you receive the right to speak for all?
>
> *sigh*. I have not spoken for anybody else. pay attention, dammit!
That's you speaking to yourself, I suppose ...
> I'm objecting to be told what I should prefer, when I don't.
... because nobody wants to tell you what you "should prefer". It's
all in the part of Kastrups posting you thoughtfully didn't quote:
David Kastrup wrote
(<m2loc7b...@mailhost.neuroinformatik.ruhr-uni-bochum.de>):
"It is easy to configure a graphical Emac with
colour support in a way as to display no graphic
menus and icons, and no colours."
So what are you complaining about? You complain that there is a
program, lets call it "XEmacs" for the moment, which can be used with a
GUI *and* without one.
You have the choice, so choose and stop whining about some noexistant
persons trying to force something on you.
To sum it up:
1.) Nobody tries to force you to use such a flexible piece of
software. You don't want it? Don't use it.
2.) Even if you were forced to use XEmacs, you'd be free to configure
it into not using any GUI-elements.
So there you are. Anything you wanted and much more for people who
want more.
--
j...@delorges.in-berlin.de -- Berlin, Germany
Marcus Daniels
>>>>> "KB" == Karsten Ballueder <kar...@pimiraculix.physik.uni-karlsruhe.de> writes:
KB> What keeps annoying me as a user, are the incompatibilities
KB> between the two emacsen. To me it seems that, instead of talking
KB> with each other on how to at least keep them compatible, the two
KB> camps fight each other with heated rethoric.
More annoying still: the irrelevant whinings of lusers complaining about
these debates getting in the way of their consumption.
Barry A. Warsaw
>>>>> "JZ" == Jamie Zawinski <j...@netscape.com> writes:
JZ> By the way, the -p convention has been used on variables since
JZ> the dawn of time. TV::WHO-LINE-JUST-COLD-BOOTED-P.
JZ> ZWEI::*CASE-REPLACE-P*. UCL::DWIMIFY-PACKAGE-P. And so on.
JZ> Where've you been?
It's also in a sci fi short story, I forget by who (Heinlein?), and I
forget when (mid-70's?). At one point in the story the female
computer expert is wearing a shirt that says "screw p" and the male
antagonist is pondering what that means. I think it's described in
the story that the "p" stands both for predicate and it also kind of
looks like a `?'.
Anybody remember the details about that short story?
Karsten Ballueder
Can't we just stop this heated rethoric?
From a user's point of view, I just don't care why the two emacsen are
different and I tend to blame both teams.
After long having been happy with CLIs, regarding GUIs as something
stupid, slow and simply unnecessary, it happened that I changed my
mind. However, who are we to decide what others should use?
What I want is the freedom to choose whether I want to use a GUI or a
CLI. And, I want my editor to support both.
As Emacs takes quite some time to learn for the beginner, especially
if you want to customize it, I happened to look at it again and again
and never got convinced enough to switch from vi. Then some day, along
came XEmacs and suddenly I had a powerful editor which was easy to use
as well. As a novice to emacs, I could mess around in the menus and
was greatly impressed by the wonderful "vm" mailer.
Now, of course I can have all this in GNU emacs as well. And now,
after having been using XEmacs for months, it is no longer a problem
for me to do so. But the point is, that without XEmacs, I had never
looked at using emacs again.
What keeps annoying me as a user, are the incompatibilities between
the two emacsen. To me it seems that, instead of talking with each
other on how to at least keep them compatible, the two camps fight
each other with heated rethoric. That's ridiculous!
Why don't you just spend the time on useful things?
If you need ideas what do to with your spare time, here is my
wishlist:
- make emacs/xemacs more compatible
- make emacs/xemacs multithreaded (why can't I read mail, while gnus
looks for my news)
- make xemacs faster (it is _so_ slow! maybe I should look at
GNU/emacs? :-)
Please, don't get me wrong. I am _not_ saying that any of the emacsen
is better, nor am I in any position to criticise your work. I am just
a simple user who got fascinated on how much energy you waste on this
stupid thread.
Thanks to both camps anyway for the best editor(s) I know!
--
|/ |>
|\arsten |>allueder e-mail:kar...@pimajestix.physik.uni-karlsruhe.de
The KBackup Homepage: http://www.uni-karlsruhe.de/~Karsten.Ballueder/
Do "finger kar...@pimajestix.physik.uni-karlsruhe.de" for PGP-key.
Erik Naggum
* David Kastrup
| While you forbid the majority to have a choice,
I know you're German, but nobody has been forbidding anybody anything
anywhere in these discussions. please pay attention.
Erik Naggum
* Jo Meder
| So what are you complaining about? You complain that there is a program,
| lets call it "XEmacs" for the moment, which can be used with a GUI *and*
| without one.
you should have noticed that the discussion on user interfaces had moved
away from just Emacs/XEmacs and into a more general domain. your summary
is remarkably stupid, considering what the arguments have been here. that
"pay attention" was directed to fools who insist on commenting when they
have not read anything of the actual discussions. but that's generally
what happens to a debate on USENET. the fools come in when a general
understanding, if not agreement, has emerged among the earlier debaters,
and we get such things as "where have you been"-style bickinger over the
-p convention instead of intelligent arguments.
Christopher Davis
EN> == Erik Naggum <nob...@naggum.no>
EN> how _could_ a program adapt to the user without the user
EN> communicating his wishes to the program?
Gnus 5 has adaptive scoring, which doesn't require the user to explicitly
communicate his wishes to the program. Perhaps code could be developed
for XEmacs that would ask the user if they wanted toolbars and/or menubars
turned off ("you've been using XEmacs for 15 days and haven't used the
toolbar yet; perhaps you'd rather reclaim the screen space?").
Emacs has long had something like this (disable-command) for protecting
functions which might confuse novices, but which experts may want enabled.
--
Christopher Davis <c...@kei.com> <URL: http://www.kei.com/homepages/ckd/ >
"I conclude that the CDA is unconstitutional and that the First Amendment
denies Congress the power to regulate protected speech on the Internet."
-- Judge Stewart Dalzell in _ACLU v. Reno_
Ken Bibb
>You're probably right, but for the wrong reason. Most people are stupid
>enough to choose something that looks good over something that is good.
These are not diametrically opposed. Something can look good and
be good at the same time.
>Those people are just wrong. Those people would build a scientific
>discipline based on ascetics, not correctness.
>I would argue that a better interface is a more efficient interface,
>for input as well as output. A colour display or GUI may look better,
>but it does not guarantee a good interface. A good interface is relies
>on good design.
The more efficient interface is the one which provides the easiest-to-
read text. For some systems, that will be anti-aliased, variable
width fonts. On others, it will be a default screen font (since that's
the only thing there ;)
If you don't like anti-aliased fonts then don't use them--your choice.
But for many people, it's the best looking text around.
--
Ken Bibb "If the boundary breaks I'm no longer alone
kb...@arastar.com Don't discourage me
kb...@best.com Bring out the stars/On the first day"
kb...@csd.sgi.com David Sylvian--"The First Day"
Roger Espel Llima
In article <61g22do...@anthem.cnri.reston.va.us>,
Barry A. Warsaw <bwa...@cnri.reston.va.us> wrote:
>It's also in a sci fi short story, I forget by who (Heinlein?), and I
>forget when (mid-70's?). At one point in the story the female
>computer expert is wearing a shirt that says "screw p" and the male
>antagonist is pondering what that means. I think it's described in
>the story that the "p" stands both for predicate and it also kind of
>looks like a `?'.
>
>Anybody remember the details about that short story?
That was "Press Enter", I'm not sure about the author but it might be
John Varley. The male isn't an antagonist btw (unless we're talking
about another story, but I do remember something along the lines of a
"screw-p")... the female character is a hacker working for the police to
find out what happened, and the guy is the neighbor.
-Roger
--
e-mail: roger.es...@ens.fr
WWW page & PGP key: http://www.eleves.ens.fr:8080/home/espel/index.html
Des Herriott
In article <30569043...@naggum.no>,
Erik Naggum <nob...@naggum.no> writes:
> I know you're German, but nobody has been forbidding anybody anything
> anywhere in these discussions. please pay attention.
Oh, please. These discussions are pretty worthless when they degenerate
to this sort of mud-slinging. Take it to email, guys.
Just so I'm vaguely on-topic: I like both variants, but I use XEmacs
more, because it's prettier, and easier to configure for someone like
me who doesn't grok Lisp. GNU Emacs seems faster and more compact but
not as friendly or polished. All IMHO, of course.
--
Des Herriott
d...@corp.netcom.net.uk
David Kastrup
Erik Naggum <nob...@naggum.no> writes:
> * David Kastrup
> | While you forbid the majority to have a choice,
>
> I know you're German, but nobody has been forbidding anybody anything
> anywhere in these discussions. please pay attention.
Why so shy? In your private Email reaction you have just now called me
"Nazi" for my differing opinion. (BTW, it is rumoured that this word
will end a thread. Let's hope so).
It is sad that you have no more arguments left than those. Not that
this is of any relevance, but I am a born citizen of the U.S.A., and
my father was about 10 years old when WWII ended.
Anyhow, since you'll need to have the last word anyway (or you'll
choke), this is the last you'll hear from me about this thread.
Holger Schauer
>>In article <30567260...@naggum.no>, Erik Naggum <nob...@naggum.no> writes:
EN> * Holger Schauer | Emacs (in it's current state) doesn't offer
EN> you the possibilities I like | (!) to have in XEmacs (and please,
EN> Erik don't answer: do it for Emacs | again - I have to get MY
EN> work done, not yours. That's why I use XEmacs, | not Gnu Emacs).
EN> "MY work" vs "your work"?
EN> Emacs and free software in general were supposed to help people
EN> work in a community, where each programmer could create software
EN> for himself as well as others who would like to use it, and they
EN> should not have to do the same work all over again. when you try
EN> to make this "my work" vs "your work", you have pin-pointed the
EN> problem that I suggest in the XEmacs camp: that of conflict,
EN> competition, reinvention, and reimplementation, the very
EN> antitheses of free software and cooperation to solve common
EN> problems.
Sorry, Erik, but you got me completely wrong. It is much more simple
than what you say: I don't want a competetion - I just wanted to point
out that I have to do my work and you have to do yours (and maybe we
can help making other people work and ourselves' a bit easier.) If you
want to say that I belong to the marketing people this is nonsense.
After all we get paid to do our work which is unrelated to who uses that
stuff.
EN> it is not a sign of professionalism to blame one's tools, but it
EN> is even worse not to know them to the degree where they can be
EN> used appropriately. if you want quality in your work, you learn
EN> to use your tools well. there simply is no other way.
Agreed. But if I can do my job (in XEmacs) with the help of the GUI
fine and don't need to learn every command of e.g. Gnus by heart but
can do the stuff with point-and-click, why the hack shouldn't I use it
(as you seem to say).
EN> results. now, it should be reasonably comfortable, but you can't
EN> make an omelet without breaking eggs, and you can't produce sleek
EN> cars in open fields, surrounded by dancing young girls and a
EN> lovely little brook -- you need a dirty, noisy, dangerous and
EN> restricted environment that is conducive to efficiently producing
EN> cars that are clean, silent, and safe, suitable to the kind of
EN> marketing that includes open fields, dancing young girls and a
EN> lovely little brook. it shouldn't be needlessly had, but if you
EN> believe the marketing that is aimed at consumers, and you're a
EN> producer (of anything), you have been _duped_, big time.
Come on, Erik. In XEmacs you can tell the young girls to go away but
in Emacs you won't see a trace of them. This is the difference in my
view. You seem to say that one *must* learn anything about a tool if
he wants use it professionally, and that this learning should be prior
to using the tool, right ? I (and many others I think) believe that
this is wrong: I prefer a lower learning curve which allows me to
start using simple things and learning the deep complexities when I
have to. This certainly does not imply that I prefer people driving
cars without a licence :-), learning is necessary but I can drive my
car without knowing (in the details) how the car works.
[bashing of marketing deleted]
EN> | So, Erik and any other visual-people-haters: if you don't like
EN> pictures, | colors, anything else which people like to use to get
EN> THEIR work done, | don't use them. But stop hindering people
EN> from using them (not to | mention insulting).
EN> do I hate "visual people"? no, I don't. I hate the obnoxious
EN> marketing campaigns and the idiots who claim that _only_ "visual
EN> people" should be allowed to use modern computers. don't you
EN> even see the difference?
Erik: please delete the "and" between "Erik" and
"visual-people-haters" and insert an "or", alright ? And I *never*
said that _only_ visual people should use modern computers or just
point-and-click interfaces are allowed. Instead I said that
point-and-click helps at first for many and should therefore IMHO some
kind of default which can/should/must be adopted by the individual
user.
EN> and just how have I been hindering any people in doing anything?
EN> this is the kind of pungent bovine manure you get from _idiots_
EN> who don't even try to read what they're commenting on.
Alright, you don't hinder any people from anything - but you created
the impression that you believe that every GUI user is some kind of
idiot and that you believe that Emacs should never-ever be adapted to
support these poor souls. And this is what I _wanted_ to express and
which I probably did not state clearly enough. And one point left:
why do you react that harsh on an otherwise modest posting ? This is
not necessary, is it ? Put down the flame-thrower and perhaps we will
all get something valuable out of this lengthy discussion.
Holger
--
*** Holger Schauer :-
mail_address(sch...@infko.uni-koblenz.de),
www(http://www.uni-koblenz.de/~schauer).
(-: I think sex is better than logic but I can't prove it. :-)
Holger Schauer
>>In article <30567839...@naggum.no>, Erik Naggum <nob...@naggum.no> writes:
EN> * David Kastrup | When did you receive the right to speak for
EN> all?
EN> *sigh*. I have not spoken for anybody else. pay attention,
EN> dammit! I'm objecting to be told what I should prefer, when I
EN> don't. there's a sizable minority who do not prefer the "modern"
You only spoke for yourself and state that any individual should
prefer (and be able to do so) what-ever he likes and that you (IYHO)
don't like GUIs (and don't want anybody to force you to use one). This
is fine and I agree with you (although I do use GUIs - i.e. X). But
the important point is that you don't say (or at least you just said
it implicitly) "Okay, go ahead, build your GUI but leave me with my
idea of environment" Instead you explain why *YOU* don't like GUIs and
Steve Jones tries to explain why many users prefer GUIs. I can't find
that he wants you to use a GUI.
Ray Auchterlounie
Erik Naggum <nob...@naggum.no> wrote:
[...]
>I want the _choice_ of a non-mouse-infested environment, I want the
>_choice_ of a command line, I want the _choice_ of a high-contrast screen
>like a black-and-white screen. _YOU_ want to deprive me of those choices.
>_YOU_ want to take away what makes _me_ productive, and you think you have
>the right to do it, don't you? who do you think is waging a war? YOU are,
>against the productivity of people who don't agree with you.
[...]
I don't think anyone in this thread has been advocating _forcing_
people to use colours and/or GUIs, they want choice, just like you.
Unfortunately you, and some, at least, of the GNU Emacs developers
seem to want Emacs to offer only the choices they agree with.
Example 1: Every other X program I use allows me to change the look of
scrollbars (if it has them). Emacs does not, no amount of fiddling
with Xresources or widget libraries will change it, it's been fixed.
This allows the developers the choice of using a flat black-and-white
scrollbar which will look fine on your black and white screen.
It also denies other users the _choice_ of a colour 3D-look scrollbar
to match their other programs. Providing this _choice_ would _not_
deny you the choice of a flat black-and-white scrollbar.
Example 2: Face support on terminals - something I've been looking for
since it was done for X. About 19.25 I found it, someone had written
it and shared it in the spirit of the GPL. It was never accepted by
the Emacs developers, presumably it didn't fit the grand design, Emacs
developers were upset at the prospect of faces/colours on terminals
and so the rest of us were denied the choice.
I do not wish to deny you the choice of a black-and-white screen etc.,
but adding a choice of colours does not restrict your choice of black
and white, insisting that black and white should be the only choice,
because that is all you have, does restrict others who want to use
colour (for whatever reason - people are different).
All this is need not be a problem, as explained in the GNU Manifesto:
...a user who needs changes in the system will always be free to
make them himself, or hire any available programmer or company to
make them for him. Users will no longer be at the mercy of one
programmer or company which owns the sources and is in sole position
to make changes.
Under the GPL they can, and indeed should, pass on those changes to
others that they might benefit from them. So are new versions born, to
be different things to different users with different requirments.
Yet, when this happens to GNU / FSF software the authors start yelling
that competition is bad and forked versions are evil etc. etc.
Apparently users who make modifications are supposed to argue with the
"developers" until they either convince them that the modifcations are
"the right thing" for the "one true version", or get bored and go and
do something productive. At no stage should unofficial versions be
released to offer users the _choice_ of using them.
You say elsewhere in the thread:
XEmacs continues to draw away programmers from the Emacs community
who _could_ have implemented this in Emacs. "we have XEmacs, so why
bother"
Emacs could use the XEmacs code, just as other GNU projects use code
from elsewhere - it is GPL, and that is what the GPL is for after all.
If the code/features/ideas don't fit the Emacs grand design then they
would have been rejected anyway - at least they can be used in XEmacs.
I am considering installing XEmacs, partly because of it's improved
tty support. I am not going to work on implementing this in Emacs, not
because "I have XEmacs so why bother", but because it WAS impemented
(at least in part) for Emacs and WAS REJECTED.
ray
--
Ray Auchterlounie <r...@kythera.demon.co.uk>
"Forty Two! Is that all you've got to show for
seven and a half million years' work?"
Ian G Batten
In article <30566575...@naggum.no>, Erik Naggum <nob...@naggum.no> wrote:
> ChangeLogs in the 19.14 distribution. (and whoever decided to abuse the
> -p convention for a _variable_?)
Quite a lot of people, it would appear.
violet:/usr/local/gnu/lib/emacs/19.28/lisp 08:41:51 (751)
$ egrep -l '\((defvar|setq) [a-z][^ ]*-p ' *.el
cc-mode.el
cl-extra.el
cl-seq.el
completion.el
cust-print.el
dired-x.el
ediff.el
etags.el
font-lock.el
gud.el
hideif.el
imenu.el
ispell.el
ispell4.el
makefile.el
menu-bar.el
mh-e.el
mh-utils.el
netunam.el
page-ext.el
regi.el
reporter.el
rnews.el
scheme.el
supercite.el
tar-mode.el
texnfo-upd.el
time-stamp.el
tpu-edt.el
tpu-mapper.el
vc.el
vi.el
vt-control.el
xscheme.el
ian
D. Dale Gulledge
In article <328735...@eurocontrol.fr>, Steve...@eurocontrol.fr says...
>
>I will have ever stick my neck out and say that people in general
>want a colour display, and that people in general prefer a GUI to a CLI.
You're probably right, but for the wrong reason. Most people are stupid
enough to choose something that looks good over something that is good.
Those people are just wrong. Those people would build a scientific
discipline based on ascetics, not correctness.
I would argue that a better interface is a more efficient interface,
for input as well as output. A colour display or GUI may look better,
but it does not guarantee a good interface. A good interface is relies
on good design.
Efficiency is certainly a very important criterion. When we are talking about
efficiency in a user interface, there are other factors besides the number of
keystrokes or mouse clicks worth considering. The most critical, IMHO, is how
easy it is to remember all of the commands you need. For the casual user, a
GUI is probably more efficient because it presents the selection of commands
visually in an obvious place. At some point, for those of us who rely on a
tool extensively (I currently have 7 windows open, 5 are GNU Emacs frames, one
is a console xterm and the other is Z-Mail), minimizing the number of
keystrokes and avoiding the mouse entirely becomes the most efficient
interface. There is a difference in usage pattern driving the choice.
Now if we look at the market for commercial software, we should all have
noticed that new releases come out with an alarming frequency, and they
sometimes have interface changes. This means that users are replacing their
entire tool set every few years and learning new tools constantly. Having an
obvious interface makes a program more markettable under these conditions.
While Emacs is not commercial software and is not designed for the same uses
as so many of the GUI tools out there, it runs on many of the same machines.
Therefore, as you pointed out, people notice the look first because it is just
another window among many.
I suspect that the criteria used by programmers to select their tools will
always differ from our customer's. And we are not alone in this. I have a
good friend who does freelance proofreading and typesetting. She does all of
her initial editing using XYWrite because it is powerful and fast. When she
is done with that she imports the documents into whichever desktop publishing
tool she is using.
I have chosen Linux, X, Emacs and TeX at home over flashy WYSIWYG tools. For
me they are practical. I am doing heavily customized tasks that no tools come
packaged out of the box to do. These tools provide both the mechanism to do
the customization and excellent documentation for how to do it. I've written
a major mode for Emacs, a macro file for TeX that interprets a different
character set, and keyboard bindings for Emacs to allow me to type it
efficiently. I went out and found fonts for it on the net and installed them.
These are not typical end user tasks. I don't expect shrink-wrapped software
to do them. If all I wanted to do was type memos in a single Western European
language (ISO 8859/1 character set), any word processor ought to do the trick.
If it doesn't these days, then it is defective, or the platform it is running
on is.
You have to choose the tools to suit the task you are doing and the pattern of
your usage. For infrequent users, who are in the majority, obvious
standardized menus are great. For frequent users, the eat more screen space
and CPU overhead than they are worth. It really isn't a matter of
intelligence (unless you choose a tool for the wrong reasons and get the one
that is least efficient for you).
- Dale
--
D. Dale Gulledge, Software Engineer
Directory & Operator Services, Nortel.
Rochester, NY
Warning: Bicyclist in mirror is moving faster than you think.
Erik Naggum
* Ray Auchterlounie
| I don't think anyone in this thread has been advocating _forcing_ people
| to use colours and/or GUIs, they want choice, just like you.
if they (XEmacs folks, GUI designers, etc) were in favor of choice, they
would have separated out the functionality from the user interface, and
built an interface to the functionality apart from the user interface.
they didn't. (some think that OLE or DDE (or whatever TLA is in vogue
today) is the way to go. it isn't.)
| Unfortunately you, and some, at least, of the GNU Emacs developers seem
| to want Emacs to offer only the choices they agree with.
how the hell did this "seem to be"? GNU Emacs would like to use the XEmacs
display engine, as has been said many, many times. this is impossible, for
all sorts of reasons, most of them having to do with randomly breaking
working code while writing it in the first place, and the rest with sloppy
copyright handling.
I have said so many times that it would border on spam to repeat it, that
what I think is really wrong with XEmacs was that they went off and _broke_
everything just to get their favorite user interface. I DO NOT CARE _WHAT_
INTERFACE THEY USE. I care about internals, usability to myself (just like
any other user), and I care about the Emacs Lisp programming language. I
see in Emacs a fundamental care for the programming _language_. I see in
XEmacs a fundamental attitude change towards "API"s and "trends" that have
shown _no_ real value for free and shared software, but _lots_ of easily
observable drawbacks.
| I do not wish to deny you the choice of a black-and-white screen etc.,
| but adding a choice of colours does not restrict your choice of black and
| white, insisting that black and white should be the only choice, because
| that is all you have, does restrict others who want to use colour (for
| whatever reason - people are different).
sigh. I give up. it isn't _my_ problem that fools like you have to put
words in my mouth and tell me that I "insist that black and white should be
the only choice". I NEVER HAVE, AND I NEVER WILL insist on that. OK? can
you at least try to _remember_ that, now?
| Yet, when this happens to GNU / FSF software the authors start yelling
| that competition is bad and forked versions are evil etc. etc.
your summary of this is as good as you imputing to me that insist that
black and white be the only choice. in other words, you're lying, and you
know it. (or you're an idiot, your choice.) competition _is_ bad, but for
an entirely different reason than you think. competition requires that two
or more teams implement the same functionality to "compete". this is not
only bad, it's monumentally stupid.
| Emacs could use the XEmacs code, just as other GNU projects use code
| from elsewhere - it is GPL, and that is what the GPL is for after all.
pay attention, dammit! if you have to make summaries of things you don't
understand, at least keep them to yourself. Emacs _cannot_ use the XEmacs
code. you don't _know_ that the code is GPL'ed just because people say so.
an employee is not allowed to give away his time, and any work for hire, or
what could be construed to be work for hire, requires that the employer
explicitly disclaim copyright or otherwise release the copyright on the
work performed by one of the employees. if this is not done, the GPL is in
peril, because somebody could claim copyright on parts of the purportedly
GPL'ed code. if you don't understand this much, at least keep quiet about
what could or could not be done.
| If the code/features/ideas don't fit the Emacs grand design then they
| would have been rejected anyway - at least they can be used in XEmacs.
where and when did you learn of the grand design? you have talked about
twice, now, only in negative terms. is it because you think it sounds
suitably sarcastic to talk about you use it? or is it because you would
like to share some of your grasp of that grand design?
| I am considering installing XEmacs, partly because of it's improved tty
| support. I am not going to work on implementing this in Emacs, not
| because "I have XEmacs so why bother", but because it WAS impemented (at
| least in part) for Emacs and WAS REJECTED.
do you know this? no, you don't. was it rejected? no, it wasn't. it
could not be implemented in Emacs because some idiots in the XEmacs camp
had stomped all over the code just to implement it. somebody is trying to
port this XEmacs stuff back into Emacs, but what do the XEmacs folks do?
grin? laugh? ridicule Emacs? continue their marketing campaigns against
Emacs for not having what they have? or all of the above? did you say
"benefit other users"? think again. XEmacs does not benefit users more
than in the _present_, while free software tries to set a standard for how
software should be shared and developed by lots of people in _cooperation_.
_this_ is the conflict.
Erik Naggum
* D. Dale Gulledge
| Now if we look at the market for commercial software, we should all have
| noticed that new releases come out with an alarming frequency, and they
| sometimes have interface changes. This means that users are replacing
| their entire tool set every few years and learning new tools constantly.
| Having an obvious interface makes a program more markettable under these
| conditions.
hmm. having a truly _learnable_ interface (i.e., a real language) would
not cause small changes in the user interface to break what people had
learned, so they wouldn't be in need of "easy to use" interfaces in order
to argue for rapid changes. e.g., for all its negative aspects, the DOS
command language remained essentially the same for 15 years. the Windows
stuff has changed and will continue to change so frequently that it would
be a colossal waste to study it. likewise, the Unix command languages have
remained largely the same for 15 years, too.
languages last longer than fashion, be it clothes or GUIs.