Evil People
Aaron Robinson
How can such evil people exist? Mommy i'm scared.
Douglas A. Denny
> My neighboors use notepad.exe. ::shiver::
>
> How can such evil people exist? Mommy i'm scared.
Are you sure your aren't seeming dead people? Are you sure your not in
Hell with a UUCP connection(How else could he post?)? Although, Hell
couldn't be all that bad with a UUCP connection, but I digress.
--
Douglas A. Denny
de...@bigsuckingsound.com
http://www.bigsuckingsound.com
No! Sleep! till Brooklyn!
/"\
\ / ASCII RIBBON CAMPAIGN
X AGAINST HTML MAIL
/ \
Barry Fishman
> "Aaron Robinson" <ac...@NOSPAMpo.cwru.edu> writes:
>
>> My neighboors use notepad.exe. ::shiver::
>>
>> How can such evil people exist? Mommy i'm scared.
>
> Are you sure your aren't seeming dead people? Are you sure your not in
> Hell with a UUCP connection(How else could he post?)? Although, Hell
> couldn't be all that bad with a UUCP connection, but I digress.
M-x rant-mode
M-x babeling-mode
I remember when I used uucp on my home system. I had a real mail
address uucp!barry not a POP box on someone else's machine. I could
send and receive mail from home without having to bring up the "network"
and get it from a pop server. I could send and receive files from
work without having to sit there and ftp them myself and wait. I had
a real network, as opposed to the crap most people believe to be the
"Internet" they see at home.
I didn't have the "web", I had to rely on "news" and mailing lists.
No pretty advertisements that take forever to load, not display
correctly on my browser, and end up beings ads for things I don't
want. (But then I rely on news and mailing lists now, its just slower
since I have to go through a active (but faster) phone connection.)
With UUCP my phone connections were automatic and took place when I
didn't need my phone for other purposes. I got the information later
but less of my time was wasted.
Of course you can get call-waiting, a second phone line, dsl, or
road-runner so you can pay extra to help jam up the phone system at
peak times. But uucp sent things in off hours, when you otherwise
didn't need the phone.
This may seem silly to some, but at one time I managed to get my work
done sharing a system 1/600 the one I am using now at home. I don't
seem to find myself or anyone else 600x more productive. In fact some
old programs seem to take much longer to replace than their originally
design and writing. The big difference is that algorithms that were
known but too slow to use, are now plausible. Object design is an
improvement, but other (maybe more verbose) ways to do the same thing
have existed for some time, they just had different fancy names.
I don't mind people using notepad. I can deal with notepad files
a lot better than Microsoft-Word files.
I abhor the idea that Emacs has to have all the Microsoft word /
Netscape features which I use Emacs to get away from. Emacs has been
an extremely reliable program up to now. Adding all this point at
pictures and grunt stuff is senseless. The current "want to use
computers but are software illiterate" crowd will be dead in a
generation or two, so do we need another dark ages developing software
for them?
As it was worked out in the late stone age, languages are a lot better
at communicating information then pointing and grunting. People who
really need computers, (and not Internet appliances) will come up with
a richer means of communicating with their computer. This may be
super-common-lisp or some new imperative language.
I don't think it will be a normally spoken language. Even human to
human voice communication usually involve writing down ideas first so
that the ideas can be worked out clearly. I don't think you can
program computers with the level of information you hear on talk
radio. I don't think its worth the effort to make computers as
idiotic and prejudiced as humans. We need computers to help us, not
mouth our biases.
Sorry for the rant, but sometimes I feel the new FSF direction is
to start building Microsoft like tools as the have already done
unix like tools. I find more and more GNU stuff too complex to be
useful. Are our choices going to be GNOME or .NET. This leaves me
with xterm, fvwm, and Emacs 20.
Although free software movements typically have a hard time developing
new ideas, they won't survive relying on Microsoft for how things
should be done.
The new generation GNOME tools seem to create a lot of private
information which doesn't need to be kept, never mind left around in
plain text. Putting some of these things in a .private directory is
not security. Neither would be creating a Microsoft resource system.
Some things are more private than your password. Richard Stallman
talks about privacy a lot, and seems to understand it. Maybe he
should look closer to home. In his defense, I doubt he uses GNOME.
I have an increasingly complicated script to weed out all of the
senseless resource files in my home directory. Somehow they don't
seem to be missed when I login again. They do take up a lot of space
in an inconvenient place (all over my home directory) which makes my
quick backup procedures overly complex (avoiding them). Maybe
GNU/linux needs to enforce a ~/config directory to put this garbage.
On a happier note, no matter how frustrated I get with the direction
the GNU/linux world is going, at least it isn't Microsoft Windows,
Barry Fishman
Chong Yidong
> Netscape features which I use Emacs to get away from. Emacs has been
> an extremely reliable program up to now. Adding all this point at
> pictures and grunt stuff is senseless. The current "want to use
> computers but are software illiterate" crowd will be dead in a
> generation or two, so do we need another dark ages developing software
> for them?
(X)Emacs 21 being able to display pixmaps is a far cry from pointing
and grunting. This feature still seems clunky (or maybe I just don't
know how to use the functions correctly), but I think it has great
potential.
Currently, whenever I need to view an image, I have to run a viewer
program like xv. It pops up a new x-window, and I have to reach for my
mouse to delete it when I'm done. It would be nice to be able to
manage images in Emacs instead, since Emacs actually has a sane
windowing system.
-- Yidong
Barry Fishman
> (X)Emacs 21 being able to display pixmaps is a far cry from pointing
> and grunting. This feature still seems clunky (or maybe I just don't
> know how to use the functions correctly), but I think it has great
> potential.
I was only commenting on the point and grunt part. If you are having
problems with a point and grunt interface, don't feel alone. I have
problems with it also. I draw my adverse conclusions about it by
looking over the shoulders of other software developers who like and
use MicroSoft Windows every day. I just look at the time and
concentration it requires to do everyday tasks. They are held back by
it, they just don't realize it. When I find things that seem simpler
for them, I can usually make up the disadvantage by quickly writing a
short sh or perl script and putting it in my ~/bin directory.
Point and grunt interfaces only allow a fixed set of activities.
I may sound old and crotchety, but I usually take to new ideas
enthusiastically. One can't exist as an engineer without a strong
desire for learning new things. I read new computer language manuals
to relax.
> Currently, whenever I need to view an image, I have to run a viewer
> program like xv. It pops up a new x-window, and I have to reach for my
> mouse to delete it when I'm done. It would be nice to be able to
> manage images in Emacs instead, since Emacs actually has a sane
> windowing system.
>
> -- Yidong
I have trouble commenting on XEmacs specifically, but what you say
seems logical. The xv pop-up is an anoyance for me most of the time.
But xv has a whole relm of capabilities for dealing with images that
are probably impractical to put in Emacs. But you are right, most of
the time the image belongs in the displayed document.
I have made many attempts to build the newer versions of XEmacs and
never got something I had much faith in. Originally it was the
problem of finding all the image conversion software it needed and
getting the librarys correct (on my Sun computer). Now, under
GNU/Linux, its more a matter of reconciling the general purpose e-lisp
setup in XEmacs with what I have build up over many years for GNU
Emacs. It never seems worth the effort when 99.9% of my time in Emacs
is in programming and reading mail/news, and XEmacs is not always
buildable on the platform I am using. I have found GNU Emacs always
simple to build, even under Windows/NT.
XEmacs seems to have done a good job in managing the outragous
complexity involved in trying to include so many e-lisp packages, but
it is somewhat a lost cause for me. It seems tailored for new users.
The packages I use, I already have, and some of them are quite
incompatable with those distributed by XEmacs. I do keep uptodate,
but the XEmacs packages seem a bit different than the ones I see
distributed separately.
I started building a XEmacs setup from scratch, but I just can't stand
the customize package! Its not too bad when you look at the package's
customize code in one window, my GNU Emacs start up code in another,
the Info page in another, and create the needed XEmacs code in a 4'th.
I just can't sustain this for 500 lines of GNU setup code, all the
while trying to keep GNU'ism and XEmac'ism straight.
I like knowing that I can build the software I use from source. I
find it gives me a better feeling about the quality of that software.
Easy to build code is usually better written. (Can you say POSIX) Over
the years, I find that my investment in software is sometimes far
greater than the attention span of the maintainers, or the life of the
computer I am using. Under these constraints XEmacs and GNOME are too
fragile to use. For example, I can use SPMS on a ESV computer without
problems. I couldn't even guess where one could currently find the
source for SPMS on the net.
I complain because even in my sheltered environment, I find myself
inundated with Microsoft/Word documents and web pages with much of the
content hidden in Javascript or Shockwave generated images. This
stuff may be fun for a 10 year old or most marketing people who have
the same maturity, but it gets boring very quickly.
Using just words you have expressed your opinions, and I think you
have made yourself perfectly clear. What is wrong with that, I don't
think a few screen dumps and HTML buttons would have helped. All my
comments were based only having different experiences and biases. We
didn't need special EMAC-21 capabilities.
Maybe when GNU Emacs 21 is no longer proprietary, I'll try to get my
Emacs setup able to handle all three.... Or not.
Barry Fishman
David B Terrell
>Currently, whenever I need to view an image, I have to run a viewer
>program like xv. It pops up a new x-window, and I have to reach for my
>mouse to delete it when I'm done. It would be nice to be able to
>manage images in Emacs instead, since Emacs actually has a sane
>windowing system.
hit 'q' in the xv window to make it go away.
I really have nothing else to contribute to this thread.
--
David Terrell | "Instead of plodding through the equivalent of
Prime Minister, NebCorp | literary Xanax, the pregeeks go for sci-fi and
d...@meat.net | fantasy: LSD in book form." - Benjy Feen,
http://wwn.nebcorp.com | http://www.monkeybagel.com/ "Origins of Sysadmins"
Dave Pearson
> Currently, whenever I need to view an image, I have to run a viewer
> program like xv. It pops up a new x-window, and I have to reach for my
> mouse to delete it when I'm done. [SNIP]
You don't have to reach for your mouse. xv has a keyboard interface. Hit
'q'.
--
Dave Pearson: | lbdb.el - LBDB interface.
http://www.davep.org/ | sawfish.el - Sawfish mode.
Emacs: | uptimes.el - Record emacs uptimes.
http://www.davep.org/emacs/ | quickurl.el - Recall lists of URLs.
Kai Großjohann
> * Chong Yidong <yid...@stanford.edu>:
>
>> Currently, whenever I need to view an image, I have to run a viewer
>> program like xv. It pops up a new x-window, and I have to reach for
>> my mouse to delete it when I'm done. [SNIP]
>
> You don't have to reach for your mouse. xv has a keyboard
> interface. Hit 'q'.
Focus-follows-mouse aficionados might have to reach for the mouse to
give focus to the xv window. I think those people should get a real
window manager, which groks both focus follows mouse _and_ allows to
automatically give focus to new windows.
kai
--
~/.signature: No such file or directory
Dave Pearson
> Focus-follows-mouse aficionados might have to reach for the mouse to give
> focus to the xv window. I think those people should get a real window
> manager, which groks both focus follows mouse _and_ allows to
> automatically give focus to new windows.
And, perhaps, can be extended using a handy language such that you can use
the keyboard to move the mouse, move windows, or pretty much anything else
you care to hack?
--
Dave Pearson: | festival.jl - Make sawfish talk.
http://www.davep.org/ | keydrag.jl - Drag windows from keyboard.
Sawfish: | sawfish.el - Sawfish mode for emacs.
http://www.davep.org/sawfish/ | uptimes.jl - Record sawfish uptimes.
Jym Dyer
=v= I learned about Alt-TAB from a Dave Barry article in which
he described a Windows 95 demo that was supposed to eliminate
the need for Alt-TAB and replace it with the mouse. A few weeks
later I found a Windows box running and hit Alt-TAB and saw what
it did. I immediately made my X-Windows manager work the same
way. It's the only good thing I ever got from Windows.
=v= I pop up xv and even Netscape from Emacs, so it's nice to be
able to Alt-TAB over to them to make them go away, so that I can
continue to dedicate the entire screen to the greater glory of
The One True Editor.
<_Jym_>
Chong Yidong
> able to Alt-TAB over to them to make them go away, so that I can
> continue to dedicate the entire screen to the greater glory of
> The One True Editor.
I used to bind Alt-TAB to switching windows, too. Unfortunately, I
found that it was not worth the loss of lisp-complete-symbol,
ispell-complete-word, complete-symbol, and so on. Nowadays, I use
Alt-[Windows Key] for switching windows instead.
-- Yidong
Dave Pearson
> I used to bind Alt-TAB to switching windows, too. Unfortunately, I found
> that it was not worth the loss of lisp-complete-symbol,
> ispell-complete-word, complete-symbol, and so on. Nowadays, I use
> Alt-[Windows Key] for switching windows instead.
I suppose it depends on what you mean when you say "Alt-TAB". The key marked
"Alt" on my keyboard is actually the Meta key, and:
,----[ C-h k M-TAB ]
| M-TAB runs the command lisp-complete-symbol
| which is an interactive compiled Lisp function in `emacs-lisp/lisp'.
| (lisp-complete-symbol &optional PREDICATE)
| [SNIP]
`----
The key that is Alt here is the one with a little window on it (as you might
gather, I'm working on a "PC" with a "PC keyboard"). Alt-TAB (Window-TAB) is
used to cycle thru windows.
Charles Sebold
[quite a few things I find myself agreeing with, followed by...]
> I abhor the idea that Emacs has to have all the Microsoft word /
> Netscape features which I use Emacs to get away from. Emacs has been
> an extremely reliable program up to now. Adding all this point at
> pictures and grunt stuff is senseless. The current "want to use
> computers but are software illiterate" crowd will be dead in a
> generation or two, so do we need another dark ages developing software
> for them?
This being alt.religion.emacs, maybe I can get away with saying this
here...
This reminds me of something I said a few months ago, on gnu.emacs.help
if I remember correctly. People were suggesting changes to a number of
defaults (and I even thought that one of them could stand to have been
changed, and it is changed in Emacs 21 pretest), but these changes were
being suggested just because people would expect it to work differently
than it does. Now, personally, I have found that most of the defaults
make sense on some level, and changing them is not only easy, but also a
useful exercise for the Emacs and/or Lisp newbie.
I argued that maybe it's better to keep the bar a little higher, rather
than lowering it to the typical new user. Others argued that this would
keep new users from sticking with Emacs (or at least GNU Emacs). I
responded that maybe this will weed out the people who wouldn't have
stuck with it anyway, and lower the number of FAQ's asked on the
newsgroup, etc. People who want the defaults to make sense from the
typical lower-end user perspective, or who want shiny things, bells, and
whistles, might end up with XEmacs (not an insult to XEmacs but it does
go in for bells and whistles), or might fall through the cracks and not
use Emacs at all. So what, I said - people like that just decrease the
signal-to-noise ratio in the *.emacs.* newsgroups, and they wouldn't end
up using it anyway, probably, unless they use it like Notepad. They
probably would never contribute to the cause unless they recognized the
power behind Emacs, and to see it you have to get your hands dirty, and
learn how to (setq) and even (defun) a few things.
The nicest thing I was called was "elitist." Private emails were a bit
more vitriolic.
I don't want Emacs development to dry up but it doesn't seem like it's
about to dry up from lack of interest, either. My point was, that we
are actually doing users a disservice by lowering the computer to their
level; their productivity (if not their whole thought process) will be
enhanced if they rise to the level of good, useful software. My own
programming skills have improved since I started picking up Emacs Lisp,
and I don't get to use Lisp at all outside of Emacs. We have manuals
and tutorials to help them get to the level they need to achieve; they
don't have to blackbox their way through Emacs.
Apparently this is an unpopular notion, though. It certainly isn't as
sexy as building an Outlook-clone or a file manager that shows you the
first few characters of your text files in their icons (Evolution and
Nautilus, respectively, from GNOME).
> Although free software movements typically have a hard time developing
> new ideas, they won't survive relying on Microsoft for how things
> should be done.
Amen. It's time to really dig into some new ideas that are not
point-and-click related. I wish I had some to contribute, but
unfortunately I am more of a command-line sort of guy and I'm relatively
happy with shell prompts (and Eshell, woo-hoo!). All the noise from
Eazel before they folded really made me think that they were going to be
the third wave or whatever of the GUI, and maybe do something great and
perhaps unheard-of with it - and what we seem to have gotten is Windows
Explorer plus a couple of neato hacks (judging by the Nautilus I got
with Debian Woody a few weeks ago).
Excel would seem to be the one place where Microsoft really has
something approaching best-of-class in applications (leaving for the
moment the VB fiascoes), and Gnumeric is doing quite well in catching
up; my comments are not directed at that particular GNOME project,
although perhaps others have a different opinion of that. But in every
other way, it seems like the desktop projects are the blind following
the blind. They have some value because some people have to have
Windows technology cloned for whatever weird personal or business
reasons, but the way they have swung into center stage in GNU/Linux is a
shame.
> On a happier note, no matter how frustrated I get with the direction
> the GNU/linux world is going, at least it isn't Microsoft Windows,
I think the same thing (although in my mind I would add "or Mac OS,"
since I came from a mostly-Macintosh background - which might account
for my problem with point-and-click and the unholy mess that it has
become).
--
Charles Sebold 21st of Sivan, 5761
Random Answer to an Emacs Very Frequently Asked Question:
gnu.emacs.help is for GNU Emacs, comp.emacs.xemacs is for XEmacs.
--
Operating System, n.:
Emacs bootloader.
Douglas A. Denny
Charles Sebold <cse...@ezl.com> writes:
> This reminds me of something I said a few months ago, on gnu.emacs.help
> if I remember correctly. People were suggesting changes to a number of
> defaults (and I even thought that one of them could stand to have been
> changed, and it is changed in Emacs 21 pretest), but these changes were
> being suggested just because people would expect it to work differently
> than it does. Now, personally, I have found that most of the defaults
> make sense on some level, and changing them is not only easy, but also a
> useful exercise for the Emacs and/or Lisp newbie.
Being relatively new to Emacs I must add that it has not been an easy
transition. I am now a True Believer, but I have also been kicking
around UN*X for almost ten years and resisted Emacs because I heard
that it was a cult! (and the fact that it looked extremely (overly)
complex, when most of the time all I needed was to send some mail and
edit a few files...)
So I have been using Emacs-lite editors like Jove and Joe and vi when
there was nothing else around. The reason I started to use Emacs was I
had a reason called LaTeX. When you start writing 300 page documents
in LaTeX you need a good editor, and Emacs is the best around for
this.
Getting to the point:
I consider myself an advanced computer user. I am not a strong
programmer, I can read and modify code but I couldn't write a cc
variable-array to save my life. One way to avoid changing things from
The Way they Are to something a little nicer for new-bies, is a novice
mode. It might do some things that new-bies might expect like
line-wrapping (although I now prefer M-q) and turning colorization
on.. and the like. It has worked for other programs like Pine in the
past, it could work for Emacs.
-Doug
Tilman J. Rothe
> On 18 Sivan 5761, Barry Fishman wrote:
> I argued that maybe it's better to keep the bar a little higher, rather
> than lowering it to the typical new user. Others argued that this would
> keep new users from sticking with Emacs (or at least GNU Emacs). I
> signal-to-noise ratio in the *.emacs.* newsgroups, and they wouldn't end
> up using it anyway, probably, unless they use it like Notepad. They
> probably would never contribute to the cause unless they recognized the
> power behind Emacs, and to see it you have to get your hands dirty, and
> learn how to (setq) and even (defun) a few things.
> about to dry up from lack of interest, either. My point was, that we
> are actually doing users a disservice by lowering the computer to their
> level; their productivity (if not their whole thought process) will be
> enhanced if they rise to the level of good, useful software. My own
> programming skills have improved since I started picking up Emacs Lisp,
> and I don't get to use Lisp at all outside of Emacs. We have manuals
> and tutorials to help them get to the level they need to achieve; they
> don't have to blackbox their way through Emacs.
In principle, I share your opinion. However, I did have to blackbox my
way through emacs to use it like a better notepad.
Please don't get me wrong, I'm a fairly advanced user, have edited
most files in /etc, frequently write scripts to automate various
tasks, send more bug reports than queries for assistance.
Firstly, I haven't found a readable lisp intro anywhere. Yes, there is
a lips-beginners-manual, but its way to eloquent to read, and other
things assume you already know lisp. Finally, I figured out very basic
lisp syntax by reading portions of the source.
Then, emacs does have "info". After glancing at it, it's way far much
to read all of it and it doesn't provide hands-on answers to my
problem. It doesn't even provide usable answers if I have very
specific questions. So I looked for tutorials. The
emacs-beginners-HowTo is a very good first reading, but after that I
was stuck. The tutorials I found on the net only explained a few
exotic features. (I didn't search too hard, as I usually do this kind
of thing when I'm offline for the day.)
Having copy&n&pasted for a while, I finally got fed up because nothing
worked the way it should and its very hard to debug things that don't
give error messages nor have a usable manual. I'm running gnus in
default mode now.
The result of all this is much mis-investment of time - I'll have to
move back to emacs anyway sooner or later because its more flexible.
In summary, emacs really needs some kind of introductory tutorial that
resides somewhere between afio man page and win95 help. It should
explain things from ground up, but shouldn't assume you are vaguely
familiar with (unix) computers (and computer-like reasoning) at
best. Only then you can safely leave some easy defaults as an exercise
to the reader.
PS please mail me a copy of replies, I'll be off usenet for some time.
Paul Jarc
> Firstly, I haven't found a readable lisp intro anywhere.
Have you seen the Emacs Lisp Intro info file (not to be confused with
the Elisp Reference Manual info file)?
paul
Charles Sebold
> Being relatively new to Emacs I must add that it has not been an easy
> transition. I am now a True Believer, but I have also been kicking
> around UN*X for almost ten years and resisted Emacs because I heard
> that it was a cult! (and the fact that it looked extremely (overly)
> complex, when most of the time all I needed was to send some mail and
> edit a few files...)
Don't kid yourself, "cult" is the best word I can find to describe this
bunch. And I have been completely brainwashed. *smiles*
> I consider myself an advanced computer user. I am not a strong
> programmer, I can read and modify code but I couldn't write a cc
> variable-array to save my life. One way to avoid changing things from
> The Way they Are to something a little nicer for new-bies, is a novice
> mode. It might do some things that new-bies might expect like
> line-wrapping (although I now prefer M-q) and turning colorization
> on.. and the like. It has worked for other programs like Pine in the
> past, it could work for Emacs.
I hesitate to suggest this (given what I just said), but this always
seemed to me to be the call for something like EMacro. I haven't used
it, but its author has claimed for a long time that it does precisely
what you suggest, setting up some fairly sane defaults and adding a
little functionality that wasn't there before. But I like the fact that
it is not distributed with Emacs; a user then at least has to figure out
how to load that, before losing touch with Lisp for a while (until they
find the need to customize further, at any rate). Maybe that's a good
compromise between our positions.
I suppose it could even be distributed as part of Emacs, as long as it
wasn't enabled by default. That way a vanilla Emacs would still work
like it does now, since I have heard good reasons for most of the more
daunting defaults (like font-lock being off, and line-wrapping). And a
simple line, near the beginning of the manual (or the end of the
tutorial) would tell a user how to turn on the novice settings (by
adding (load 'novice) or something like that) to your .emacs.
--
Charles Sebold 21st of Sivan, 5761
Random Answer to an Emacs Very Frequently Asked Question:
Your .emacs is either in C:\ or in the directory of %HOME% under Win32.
--
"I decry the current tendency to seek patents on algorithms. There
are better ways to earn a living than to prevent other people from
making use of one's contributions to computer science."
-- Donald E. Knuth, The Art of Computer Programming, Vol. 3
Barry Fishman
> Charles Sebold <cse...@ezl.com> writes:
>
>> On 18 Sivan 5761, Barry Fishman wrote:
>
>> I argued that maybe it's better to keep the bar a little higher, rather
>> than lowering it to the typical new user. Others argued that this would
>> keep new users from sticking with Emacs (or at least GNU Emacs).
The above are not my comments, but Charles's. Mine are below.
With the current focus on getting new people to use Emacs, I thought I
would discuss how an older Emacs user got started. People like me use
Emacs because it makes many tasks easier for us. We are not looking
to make our work harder.
I started using Emacs when I needed to work on a large program and
found the time spent learning a new editor was much less that the
problems of trying to do the editing in multiple terminal windows
running vi.
At that time I was pretty good at vi, I just felt it was more a line
by line editor. Even though tags actually worked better with vi, the
reasons for having tags worked very poorly. I still use vi for editing
things like tabular data files. I even have vim setup to colorize
elisp files correctly.
I found the emacs tutorial, enough to get started, and I knew enough
lisp to add some unbundled packages I needed. New editors mean new
key bindings and its always a pain to learn them. So you just want to
get as much benefit from that learning process as you can.
I still haven't bothered trying to learn all the key bindings. For
rarely used commands its simpler doing a M-x and using command name
(with completion) to enter the command. If their is a quick key
binding Emacs will then tell me about it, and if I use the command
often enough, I might start using it. There are so many ways of
finding commands interactively that I rarely need to go into info to
find it in the manual. Anyway most of the key bindings I use are
automatic and I could only recall them by watching my fingers perform
them on a keyboard.
I think the lisp code in Emacs is very well documented, and where it
is confusing, you can look about in the elisp directorys to see what
is going on. Documentation can always be improved. I found it much
easier to follow that any other software system. My biggest problem
with a lot of software is that the documentation doesn't begin by
telling you what its for. Emacs at least doesn't have this problem.
The copyrights may come first but then packages have a "Commentary"
and most explain in detail how to set them up.
Now, one can add menus and customization packages so that more people
can start editing without key bindings, but in a sense they would not
be using basic Emacs, and the certainly wouldn't be as productive.
All this does is allow Emacs to emulate another kind of editor, except
that Emacs would run much slower then other editors of that type. Is
this the best way to *sell* Emacs? People can run a vi emulation in
emacs already, but I never found anyone that really did it, since they
can run vi itself.
The amount of lisp needed to be productive in Emacs is very small.
Rather than trying to convince people to run a window based editor in
Emacs, wouldn't it be better if they just spent the time learning some
lisp. If they don't choose to learn Emacs afterwords, then probably
Emacs isn't the editor for them. There are plenty of menu based
editors around the are not just emulations. Most are free. Meantime
they have gained something by learning a imperative programing
language which they might otherwise overlook. They would have some
appreciation of a new language and means of expression, which is far
more valuable than another set of editor key/mouse bindings.
With some knowledge of lisp, Emacs becomes an extendable editor. One
can tune it to do the things you want to do more efficiently. Other
editors can only do what their authors decided you should be able to
do.
The Unix blood in me, tells me that in general its better to
have a lot of simple tools that work together. A mail reader, a news
reader a c-language editor, a java-editor, all separate programs.
Emacs is not such a tool. It is quite the opposite. It does all
these things in one place, but that is not why it is useful. It was
incredibly written with major and minor modes so it can do all
'flavors' of tasks and then 'mix-in' an uncountable number of
abilities that can be applied across the tasks.
After many years I find I am still improving my environment. I can
now translate the comments in a program from French to English. What
other editors do this? What other editors would want to? I only got
it because I was having trouble reading the SuSE news group. I got
the elisp to use in gnus but a side effect was then it would work
anywhere. It was there when I was looking at some scheme code written
in Quebec.
This is an alternate way (from unix) where the whole can be greater
than its parts.
It doesn't take much lisp to assemble the environment you want, but
without some knowledge of it, you are missing out on a great deal.
People learn SH, AWK, PYTHON, C, C++, Java, XSL, and even a
language as difficult as English to do their work. Compared with
*any* of them, the required e-lisp is simple.
Does this make me an elitist. You may say secretaries can't learn
lisp. (Isn't that elitist.) I'm suprised so many have mastered vi. I
see secretaries wasting their time fixing their PC hardware problems
because some companies don't realize that that they are a too
expensive resource for that. If the system administrators set up the
computers and the software people either tailored a Emacs environment
to meet the secretaries needs (or supplied an appropriate editor)
things would go much smoother.
As supposed non-tecnical people end up fixing their PCs, you might
find if they start with an emacs mode tailored for their expected
needs (and I don't mean key bindings), they will soon be hacking their
.emacs files to improve things in ways that other editors could not be
changed.
Barry Fishman
Greg Menke
> Does this make me an elitist. You may say secretaries can't learn
> lisp. (Isn't that elitist.) I'm suprised so many have mastered vi. I
> see secretaries wasting their time fixing their PC hardware problems
> because some companies don't realize that that they are a too
> expensive resource for that. If the system administrators set up the
> computers and the software people either tailored a Emacs environment
> to meet the secretaries needs (or supplied an appropriate editor)
> things would go much smoother.
Though its sometimes tempting to discount end-users and/or secretaries
from a technical standpoint, its an oversimplification to do so. I
worked on a pretty large accounting system based on VB and Access
(sorry- I wasn't using Linux back then...), and the 3 accountants the
project was written for taught themselves SQL so they could get into
the database and help out with validations, queries, reports and data
input. We had to set up naming conventions so their queries and
tables wouldn't mess up those used by the system.
Back in their mainframe days, the systems people never gave them the
opportunity to actually work with the system. They really liked doing
so; though they weren't the type to go messing about in stuff we asked
them to not get into either- so we could trust them.
Of course, then there are the other users who should never be left
alone with a computer...
Gregm
Douglas A. Denny
> The Unix blood in me, tells me that in general its better to
> have a lot of simple tools that work together. A mail reader, a news
> reader a c-language editor, a java-editor, all separate programs.
This statement has been bouncing around in my head for a couple of
days now..
Emacs is not an application, it is a way of life. Not an application
but a meta-operating system. One that allows me to view the world in
new and weird and wonderful ways.
M-x all-hail-emacs
Cheers
-Doug
Shyamal Prasad
Barry> The Unix blood in me, tells me that in general its better
Barry> to have a lot of simple tools that work together. A mail
Barry> reader, a news reader a c-language editor, a java-editor,
Barry> all separate programs.
Barry> Emacs is not such a tool. It is quite the opposite.
The Unix blood in me tells me that Emacs is a very good example of
this. It has a lot of simple tools that work together: a few mail
readers (VM, Gnus), a news reader (Gnus), a c language editor (C
mode), a Java editor, a file manager, a source browser etc. and they
are all separate programs ;-)
Emacs is not a tool. It's an environment.
M-x all-hail-emacs
Barry Fishman
I agree.
[I came not to bury Emacs but to honor it.]
But that isn't what makes it special (for me). At the crudest level
Emacs is a lisp environment specialized for manipulating buffers of
text and displaying them. One can write all kinds of independent
programs in it. And these programs are portable to all the platforms
Emacs runs. (Unlike java which is supposed to do this but *can't* for
some very fundamental reasons.)
To me, the thing that makes Emacs unique, is that you are really running
different faces of the same program. These `programs' can share code
(and not just minor modes) in a way that unix/windows shared libraries
can't. I'm writing a message in gnus message mode but I still have
access to a plethera of features that were actually written for other
modes.
People using windows, or gnome (as I am) get a slew of programs which
are written (using libraries) to seem like one environment. This is
just the results of their authors' pre-selection of those feature which
applications would share, and maybe an API to make that simpler.
Emacs does not play such games. Emacs programs share features because
the author of each mode (in a sense) didn't explicitly de-select the
feature. Emacs is the union of its feature rather than a series of
programs which share an intersection of features. In Windows/Gnome
environments users can add programs but the shared features are always
fixed (unless you want to rewrite *all* the applications).
Now with an upgrade of display engine, there is interest in making
Emacs simpler to learn by the masses. But which masses? Java went to
the point of supplying emulations of all the common OS environments.
This just makes testing its applications Look-and-Feel almost
impossible. Write once, run anywhere, but the screen may not be
readable everywhere.
People have always been drawn to Emacs because what it could do, not
how it looked. I really don't think putting a prettier face on it
would make it much more used. A fancy cover on a thick book may
improve sales, but would many more people read it? When it comes to
performing real work, people have to find some benefit from the tool.
Why use emacs for a web browser if its always possible to find faster
web browsers which have all the latest plugins. The benefit in emacs
is what special about it, and you can't appreciate that until you put
in some work.
The `market' for Emacs is not going to change with a fancy front end.
Most likely as people become more computer literate (in the real
sense), its use will grow. But why do we feel so concerned that
others use it? We don't make money by selling it to people.
Unfortunately, part of the concern is that without a strong demand,
many companies would ban it from their computers. It was outlawed
from most of the computers at a compony I recently worked (Advanced
Information Systems Group, Longwood FL) but I don't think they are
alone. There is a push toward PCs and a growing fear of what software
people put on them. They seem to ignore the fact that the most
dangerous software comes with the box.
I don't want to change the world to my beliefs, I just don't want to
be persecuted for them.
Barry Fishman
[ Sorry for all these religious tracts, but I am in the
process of reevaluating how I design software and for me that requires
a lot of talking. I just writing here to keep my friends and avoid
being committed. ]
Alex Schroeder
> But why do we feel so concerned that others use it? We don't make
> money by selling it to people. Unfortunately, part of the concern
> is that without a strong demand, many companies would ban it from
> their computers.
Good points you raise in your post. This last one, however, is
bewildering (to me). I am always the first to say that I couldn't
care less how many people used Linux or Emacs or whatever. As long as
it is good enough. And that means it has to attract a larger
developper base. And the assumption goes that increasing the user
base will also increase the developper base, ultimately improving our
software. I don't know wether that is a valid assumption. However,
that is the only reason I see for adding newbie-friendly features to
Emacs (such as custom, which I used to hate but which I now use quite
frequently).
Alex.
--
http://www.geocities.com/kensanata/emacs.html
http://www.emacswiki.org/
Hans O. Lowe
> but a meta-operating system.
Hallelujah! Here is my testimony to the power of such a meta-operating
system.
I started a new job recently where we are each given a cute little
Windows 98 / Novell work station with limited network access.
My job entails quite a bit of text editing for the purposing of writing
scripts for data processing. The tool I was given to do the job is none
other than the Almighty Notepad. Web access is limited to within the
intranet, and 'cmd' is disabled. After a couple weeks at the job, I was
faced with the decision of either laughing at them and walking out, or
smuggling in some tools of my own. I chose the latter.
I found out that whoever blocked the internet access came from the school
that think internet==www. Only www is blocked. So, I ftp'd and installed
emacs. With ange-ftp, now I can access my other files and tools from my
other accounts outside of the company. The next thing I did was to
download eshell. These, along with dired and other elisp packages, allow
me to create an environment with which I am somewhat comfortable. The
ability to write scripts in Elisp also allows me to customize this Windows
98 machine without the use of Visual Basic and such, which were
deliberately made unavailable anyway.
I started using Emacs years ago as a Unix tool. Now it has allowed me to
achieve much productivity in a non-Unix environment. "Meta-operating
system" is a truly descriptive term for Emacs.
----------------------------------------------------------------------
The opinions expressed here do not represent those of my computer.
----------------------------------------------------------------------
Nix
> Firstly, I haven't found a readable lisp intro anywhere. Yes, there is
> a lips-beginners-manual, but its way to eloquent to read, and other
I don't think that means what you think it means.
If you mean `too verbose to read', then consider that verbosity is much
better than the alternative. (Besides, in the lispref itself, you have a
terse introduction --- that also doesn't assume you know Lisp.)
Also, may I recommend the book _Structure and Interpretation of Computer
Programs_?
> things assume you already know lisp. Finally, I figured out very basic
> lisp syntax by reading portions of the source.
This really is not the way to learn programming languages :( it's the
way to learn their idioms, but only after you've grokked the syntax and
the rest of the basics.
> Then, emacs does have "info". After glancing at it, it's way far much
> to read all of it
That's why it's divided into sections.
(Also, Emacs is not an editor for people who don't like to read
documentation. It's complex; you must invest effort into learning it.)
> and it doesn't provide hands-on answers to my
> problem.
That's because it's not a telepathic system. How are the authors
supposed to know what problems you will have in the future? The set is
unbounded.
This is what Usenet is for.
> It doesn't even provide usable answers if I have very
> specific questions.
Personally, I have found Emacs's and XEmacs's help systems to be the
best of any programs I have ever used, anywhere. If I'm completely lost,
it's rare indeed that a search over the Emacs info files or `M-x
apropos-documentation' don't yield something useful.
> So I looked for tutorials. The
That's a weak point. You're meant, I think, to work through the manual
from the start to about the middle, then hop about; certainly that's how
I learned Emacs.
> Having copy&n&pasted for a while, I finally got fed up because nothing
> worked the way it should and its very hard to debug things that don't
That's because you didn't read the manual.
> give error messages nor have a usable manual. I'm running gnus in
> default mode now.
Oh, this is the *Gnus* manual you're talking about? That's a reference
manual only, I'm afraid, and quite a terse one; there is no user manual
yet that I know of. It's definitely a need; you're the fourth person
I've heard complain about this in as many weeks.
> The result of all this is much mis-investment of time - I'll have to
> move back to emacs anyway sooner or later because its more flexible.
Using Emacs is a misinvestment of time since you'll have to move back to
Emacs? I profess confusion.
> In summary, emacs really needs some kind of introductory tutorial that
> resides somewhere between afio man page and win95 help. It should
i.e., somewhere between a terse reference manual and the most worthless
pile of wreckage masquerading as a help system that I have ever seen?
Well, you're welcome to write one, but since Emacs already has new user
documentation, detailed reference documentation and documentation for
(almost) every one of its functions and variables, I don't think you'll
find many other people willing to spend their time on it.
> explain things from ground up, but shouldn't assume you are vaguely
> familiar with (unix) computers (and computer-like reasoning) at
If someone's not familiar with computer-like reasoning, they shouldn't
be trying to use a text editor whose primary power lies in its
programmability; they should be willing to learn to program, either
first or in parallel with learning Emacs.
Emacs is not MS Word.
> PS please mail me a copy of replies, I'll be off usenet for some time.
*sigh* if you insist... please reply to the group, though.
--
`This will immediately become a flamewar.' --- Mark Mitchell
Kai Großjohann
> Oh, this is the *Gnus* manual you're talking about? That's a
> reference manual only, I'm afraid, and quite a terse one; there is
> no user manual yet that I know of. It's definitely a need; you're
> the fourth person I've heard complain about this in as many weeks.
Please make a sacrifice at the altar of Lars, but then come to
my.gnus.org quickly and help improve the user manual.
Thanks!
Hans O. Lowe
> Also, may I recommend the book _Structure and Interpretation of Computer
> Programs_?
"Simply Scheme" is an excellent intro to Lisp, and a "prequel to SICP," as
the authors called it.
Nix
> On 28 Jun 2001, $}xinix{$@esperi.demon.co.uk wrote:
>
>> Oh, this is the *Gnus* manual you're talking about? That's a
>> reference manual only, I'm afraid, and quite a terse one; there is
>> no user manual yet that I know of. It's definitely a need; you're
>> the fourth person I've heard complain about this in as many weeks.
>
> Please make a sacrifice at the altar of Lars, but then come to
> my.gnus.org quickly and help improve the user manual.
I would do this, if I could. However, I'm much better at reference
manuals --- which Gnus already has, and an excellent one too --- than at
user manuals. I just don't think the right way :(
(Neither, I expect, will anyone who learns new languages by reading the
language standards.)
raj
>able to Alt-TAB over to them to make them go away, so that I can
>continue to dedicate the entire screen to the greater glory of
>The One True Editor.
Amen !
Er... that should have been ....
Emacen !
raj
>able to Alt-TAB over to them to make them go away, so that I can
>continue to dedicate the entire screen to the greater glory of
>The One True Editor.
Amen !
Er... that should have been ....
Emacen !
---------------------------------------------------
VI VI XIII
Roman Neighbour of the Beast