Where do you RTFM ?
Martin Emrich
When newbies ask something, they are often asked to RTFM...
But where do I find the Information ? Very often, neiter #man foo nor
the contents of the /usr/doc/foo are very helpful. Where is aditional
documentation usually installed ? Or do you know a good
"getting-documentation-howto" ? (maybe even something in german.. would
be nice)
Thx
Martin
PS: Merry Christmas to all of you !
--
To UNSUBSCRIBE, email to debian-us...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Alec
> Hi All !
>
> When newbies ask something, they are often asked to RTFM...
>
> But where do I find the Information ? Very often, neiter #man foo nor
> the contents of the /usr/doc/foo are very helpful. Where is aditional
> documentation usually installed ? Or do you know a good
> "getting-documentation-howto" ? (maybe even something in german.. would
> be nice)
>
> Thx
>
> Martin
>
> PS: Merry Christmas to all of you !
Try typing "info:/" into konqueror. This will give you something to read
until 2003. Otherwise, RTFM is sometimes accompanied by DAFS (Do a f*
search), which should be directing you to, say,
http://groups.google.com/groups?group=comp.os.linux.misc
with all google's search capabilities.
HTH
Alec
Lev Lvovsky
linuxdoc.org for whole documentation
groups.deja.com for specific questions (this list posts to usenet as
muc.lists.debian.user methinks)
-lev
On Mon, 24 Dec 2001, Martin Emrich wrote:
> Hi All !
>
> When newbies ask something, they are often asked to RTFM...
>
> But where do I find the Information ? Very often, neiter #man foo nor
> the contents of the /usr/doc/foo are very helpful. Where is aditional
> documentation usually installed ? Or do you know a good
> "getting-documentation-howto" ? (maybe even something in german.. would
> be nice)
>
> Thx
>
> Martin
>
> PS: Merry Christmas to all of you !
>
>
>
--
personal site :: www.sonous.com
rave site :: raves.sonous.com
I'm a DJ! site :: djkgb.sonous.com
"Progess is the direct result of dissatisfaction." -Mark Rudholm
Osamu Aoki
> When newbies ask something, they are often asked to RTFM...
>
> But where do I find the Information ? Very often, neiter #man foo nor
> the contents of the /usr/doc/foo are very helpful. Where is aditional
> documentation usually installed ? Or do you know a good
> "getting-documentation-howto" ? (maybe even something in german.. would
> be nice)
Always:
http://www.debian.org/doc/ddp
After apt-get install doc-linux-text :
/usr/share/doc/HOWTO
or
http://www.linuxdoc.org
And my plug :)
http://qref.sourceforge.net/quick/
--
~\^o^/~~~ ~\^.^/~~~ ~\^*^/~~~ ~\^_^/~~~ ~\^+^/~~~ ~\^:^/~~~ ~\^v^/~~~
+ Osamu Aoki <deb...@aokiconsulting.com>, GnuPG-key: 1024D/D5DE453D +
+ My debian quick-reference, http://qref.sourceforge.net/quick/ +
dman
|
| When newbies ask something, they are often asked to RTFM...
|
| But where do I find the Information ? Very often, neiter #man foo nor
| the contents of the /usr/doc/foo are very helpful. Where is aditional
| documentation usually installed ? Or do you know a good
| "getting-documentation-howto" ? (maybe even something in german.. would
| be nice)
Sometimes a manpage is good. Sometimes /usr/share/doc/<foo>.
Sometimes install foo-doc package to get the documentation. google
can help a lot. linuxdoc.org has lots of howtos.
If you can't find the F{ine} Manual for a given piece of software or
task then just ask. It usually helps to mention where you have
already looked too.
-D
--
"He is no fool who gives up what he cannot keep to gain what he cannot lose."
--Jim Elliot
Brian Nelson
> On Mon, Dec 24, 2001 at 06:37:25PM +0100, Martin Emrich wrote:
> | Hi All !
> |
> | When newbies ask something, they are often asked to RTFM...
> |
> | But where do I find the Information ? Very often, neiter #man foo nor
> | the contents of the /usr/doc/foo are very helpful. Where is aditional
> | documentation usually installed ? Or do you know a good
> | "getting-documentation-howto" ? (maybe even something in german.. would
> | be nice)
>
> Sometimes a manpage is good. Sometimes /usr/share/doc/<foo>.
> Sometimes install foo-doc package to get the documentation.
And sometimes the package just has poor documentation. Some packages
just have the upstream documentation thrown in /usr/share/doc/, which
often conflicts with the actual Debian-specific configuration of the
package. It's frustrating, and there's not much you can do other than
ask around or read the source.
> If you can't find the F{ine} Manual for a given piece of software or
> task then just ask. It usually helps to mention where you have
> already looked too.
--
Brian Nelson <nel...@bignachos.com>
BigN...@jabber.org
http://bignachos.com
Linda Laubenheimer
Alec wrote:
>
> Try typing "info:/" into konqueror. This will give you something to read
> until 2003. Otherwise, RTFM is sometimes accompanied by DAFS (Do a f*
> search), which should be directing you to, say,
> http://groups.google.com/groups?group=comp.os.linux.misc
> with all google's search capabilities.
What I tend to say is STFW - Search The Friggin' Web - via google,
with the package/problem and OS (version) as keywords. Search web
pages and groups (usenet). I also RTFM in various books (you know,
those paper and ink things) that are published on the subject by
reputable companies. My computer book expenditures are quite high,
but that's me and my addiction.... ;)
Linda
--
Linda J Laubenheimer
UNIX Geek, Sysadmin, Bibliophile and Iconoclast
http://www.modusvarious.net/ - consultants available
http://www.laubenheimer.net/ - personal demo site
http://www.geocities.com/laubenheimer/ - web design gaffes (I wouldn't
disgrace a real ISP with these) and rants about bad design.
Karsten M. Self
--sgneBHv3152wZ8jf
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Mon, Dec 24, 2001 at 06:37:25PM +0100, Martin Emrich (debianml@emmes-wor=
ld.de) wrote:
> Hi All !
>=20
> When newbies ask something, they are often asked to RTFM...
I consider this acceptable only IITTNTRFMTFR [1]
> But where do I find the Information ? Very often, neiter #man foo nor
> the contents of the /usr/doc/foo are very helpful. Where is aditional
> documentation usually installed ? Or do you know a good
> "getting-documentation-howto" ? (maybe even something in german..
> would be nice)
Assuming you know a command:
$ command --help
$ command -help
$ man command
=20
Rarely:
$ info command
I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
basic concept still sucks. It should be scrapped for a searchable
format based on HTML, XHTML, or preferably something like DocBook
capable of creating multiple output formats.
Local docs would be far more useful with a good search engine _specific
to online help_. AFAIK, there's not a good solution to this for Debian,
RH used to be somewhat better in this regard.
If you don't know the command but have an idea of the problem area
concept:
$ apropos concept | less # often augmented with a grep filter
$ locate concept
Google. Surfraw rocks, and rocks harder with w3m:
$ google <search term>
Ditto: Google groups. I still forget this exists.
HOWTOs. I have these installed locally:
http://localhost/doc/HOWTO/en-html/
http://localhost/doc/HOWTO/en-html/mini
Mailing lists and boards I hang at. Typically debian-user, svlug, and
zIWETHEY: http://z.iwethey.org/forums/render/board/show?boardid=3D1
Occasionally some chats and/or MUDs I hang at.
Books:
http://kmself.home.netcom.com/Linux/FAQs/linux-books.html
My mind. Both for knowledge and to extract clues from the information
available.
Peace.
----------------------------------------
Notes:
1. If it tells the newbie the right fucking manual to fucking read.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--sgneBHv3152wZ8jf
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8J7U8OEeIn1XyubARApVuAJwLgkRevTwAg1b3Wx9qWN1lVpzaBwCdG6Rm
4eEKmn8MMl3Ln+W/WOI9IFs=
=6vMG
-----END PGP SIGNATURE-----
--sgneBHv3152wZ8jf--
Faheem Mitha
On Mon, 24 Dec 2001, yugami wrote:
> www.linuxdoc.org
> www.linuxnewbie.org
Also http://newbiedoc.sourceforge.net/ (Debian-specific)
Faheem.
Jesse Goerz
> Hi All !
>
> When newbies ask something, they are often asked to RTFM...
>
> But where do I find the Information ? Very often, neiter #man
> foo nor the contents of the /usr/doc/foo are very helpful.
> Where is aditional documentation usually installed ? Or do you
> know a good "getting-documentation-howto" ? (maybe even
> something in german.. would be nice)
>
> Thx
>
> Martin
>
> PS: Merry Christmas to all of you !
Try
http://newbiedoc.sourceforge.net/general/help-system.en.html
I need to update that but it will get you started.
There is a package available called newbiedoc but I believe it's
only available in testing.
If you're running stable take a look at the dhelp and dwww
packages which are available in stable. Or just do:
apt-get install dhelp dwww
After you install these packages you may wish to put your
browser start page as (or create a bookmark):
file:///usr/share/doc/HTML/index.html
which is the index to dhelp. I had to do this because just
typing dhelp at the command line didn't work. I believe there
is a missing symlink or something.
Just type dwww at the command line to get dwww.
HTH,
Jesse
Gary Turner
>on Mon, Dec 24, 2001 at 06:37:25PM +0100, Martin Emrich (debi...@emmes-world.de) wrote:
>> Hi All !
>>
>> When newbies ask something, they are often asked to RTFM...
>
>I consider this acceptable only IITTNTRFMTFR [1]
>
Thanks, Karsten. When I query the list, it's because I couldn't find
the answer in the manual nor in the 8 or 10 books on my desk. Worse, I
found the answer when I didn't need it and can't find it when I do.
We newbies often need a pointer to guide us. RTFM is not a pointer.
RTPM, Read This Particular Manual, is.
I do not address the case of the freeloader who has not made any effort
in his own cause.
<snip>
>
>
>Peace.
>
>----------------------------------------
>Notes:
>
>1. If it tells the newbie the right fucking manual to fucking read.
Now off to Mom'n them's for xmas eve dinner. My best wishes to
everyone.
gt
Yes I fear I am living beyond my mental means--Nash
Osamu Aoki
> dman <dsh...@rit.edu> writes:
> > On Mon, Dec 24, 2001 at 06:37:25PM +0100, Martin Emrich wrote:
> > | When newbies ask something, they are often asked to RTFM...
> > |
> > | But where do I find the Information ? Very often, neiter #man foo nor
> > Sometimes a manpage is good. Sometimes /usr/share/doc/<foo>.
> > Sometimes install foo-doc package to get the documentation.
>
> And sometimes the package just has poor documentation. Some packages
> just have the upstream documentation thrown in /usr/share/doc/, which
> often conflicts with the actual Debian-specific configuration of the
> package. It's frustrating, and there's not much you can do other than
> ask around or read the source.
Sometimes, true. In that case:
# vi /etc/apt/sources.list
... activate deb-src lines for unstable
# apt-get update
# apt-get source packagename
# mc
Then you can read all the sources :)
--
~\^o^/~~~ ~\^.^/~~~ ~\^*^/~~~ ~\^_^/~~~ ~\^+^/~~~ ~\^:^/~~~ ~\^v^/~~~
+ Osamu Aoki <deb...@aokiconsulting.com>, GnuPG-key: 1024D/D5DE453D +
+ My debian quick-reference, http://qref.sourceforge.net/quick/ +
Carel Fellinger
...
> I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
> basic concept still sucks. It should be scrapped for a searchable
> format based on HTML, XHTML, or preferably something like DocBook
> capable of creating multiple output formats.
I don't get this, just as I don't understand people bashing Stallman.
What is it in the program that is so horrible to ***DESPISE*** it?
It *is* searchable from within info, it *has* several output formats
as it is LaTeX based. And the basic concept seems valid.
Okee, the interface is not to everybodies likening:), so improve.
Okee, the translation into other formats has its problems:), so improve.
But the search works for me, and the idea that there is more to documenting
a program then merely listing what options it has seems okee. And going
through a tree like doc structure is quit common these days.
So what is it that makes you (and others) react so vehemently?
--
groetjes, carel
dman
| On Mon, Dec 24, 2001 at 03:07:41PM -0800, Karsten M. Self wrote:
| ...
| > I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
| > basic concept still sucks. It should be scrapped for a searchable
| > format based on HTML, XHTML, or preferably something like DocBook
| > capable of creating multiple output formats.
|
| I don't get this, just as I don't understand people bashing Stallman.
| What is it in the program that is so horrible to ***DESPISE*** it?
Personally it is the emacs-centric interface. I don't get it. (I did
use emacs for a while, but I remember hardly any commands, vim was
just easier for me to remember). Also, when I have tried info a few
times, I see a highlighted word (link) that I want to follow, info
tells me there is no link there.
In short, I haven't been able to get it to work. 'man' works though.
(and I hate those manpages that say "this manpage sucks cause I [the
author] like info better")
-D
--
All a man's ways seem innocent to him,
but motives are weighed by the Lord.
Proverbs 16:2
John Hasler
> Personally it is the emacs-centric interface.
What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch, and ENTER?
I use both info and Emacs and don't find info Emacs-like at all when not
run from inside Emacs.
--
John Hasler
jo...@dhh.gt.org
Dancing Horse Hill
Elmwood, Wisconsin
Karsten M. Self
--wAI/bQb0EMvlZCHl
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 02:44:17AM +0100, Carel Fellinger (cfel...@iae.nl)=
wrote:
> On Mon, Dec 24, 2001 at 03:07:41PM -0800, Karsten M. Self wrote:
> ...
> > I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
> > basic concept still sucks. It should be scrapped for a searchable
> > format based on HTML, XHTML, or preferably something like DocBook
> > capable of creating multiple output formats.
>=20
> I don't get this, just as I don't understand people bashing Stallman.
> What is it in the program that is so horrible to ***DESPISE*** it?
- It attempts to replace, not augment, an existing, established,
viable, useful, and effective standard. This is almost always a bad
idea. The far better tack: provided augmented functionality. If
your solution is compelling enough, people will transition.
- It's (largely) bound to a specific viewer. Which, if you don't use
emacs, isn't particularly usable, and is about as intuitive
as...well, emacs. This has changed as additional viewers are
avilable (e.g.: pinfo -- based on lynx...but, of course, I don't
care for lynx's keybindings, and use w3m instead....), but it's not
fully remedied the problem. The authoring markup is distinct from
both manpages (groff) and DocBook (SGML).
- It fragments the documentation effort. GNU favors info. Debian
favors manpages (I've written on this in the past, references
supporting both claims are available).
- Info and man serve different functions. Man is meant to be a quick
reference, putting the most likely needed information -- command
arguments and use -- first. For Info, you've got to dig to get to
this. It's like the "entrance tunnels" that some web wanks were
advocating for websites for about fifteen minutes in 1997. Like I
want to open four pages to get to your site.... Info puts shit in
the way of what I'm looking for.
- It splits a programs docs into multiple pages. In electronic
format, single, large, documents which can be searched through
readily (gee: that sounds like a manpage under 'less') is more
useful than a set of smaller docs which can't be as readily scanned.
- GNU project manpages frequently include a passage:
This man page is not kept up to date except when volunteers want
to maintain it. If you find a discrepancy between the man page
and the software, please check the Info file, which is the
authoritative documentation.
[Taken from the gcc manpage]
...followed by dire warnings that the manpage may not be updated,
etc., etc. At which point the pitiless reader turns to the info
document...which in many cases is a copy of the same manpage (now
presented in an unfamiliar document viewer). Houston, we've got a
problem.
> It *is* searchable from within info, it *has* several output formats
> as it is LaTeX based. And the basic concept seems valid.
The basic concept is valid. However, what seems to have happened in the
world is that we got several rival document and hypertext languages in
about the same five minute period of 1987. Tim Berners-Lee happened to
win the race. Ironically, man pages translate quite well to HTML, and
DocBook / LDP has largely filled the niche Info is more suited to:
providing a more substantial, book-styled, document suitable for
browsing rather than a quick reference.
Info _was_ a really good idea at its time. Binding it to an emacs-style
editor made a bit of sense. But things have changed, the computing
audience has exploded beyond Richard's dreams (both in general, and for
GNU/Linux specifically), and I'd venture to suggest that most people
using GNU/Linux either don't know emacsen or don't use them as their
principle editing environment. This is no longer a principally
technical community, even on the "techincal" platform.
> Okee, the interface is not to everybodies likening:), so improve.
No. Divorce. =20
Thou shalt render content and presntation asunder. It was a fatal sin
in 1987 when Info was developed. It remains a sin.
> Okee, the translation into other formats has its problems:), so
> improve. =20
No. Realize you've been one-upped and passed by.
Man works for simple docs. Use man2html to present man pages via a web
interface. Convert the Info content to one or the other. ***AND BACK
FILL THE !@#$%^&*() MAN PAGES YOU'VE DEPRECATED FOR THE PAST FIFTEEN
YEARS***.
Hmm...actually, dwww and info2www seems to answer a bit of this issue.
I've never really played with dwww a whole mess, it actually seems to
answer a number of the issues I've got.
> But the search works for me
I've got _no_ idea how to access this search function. The _only_ time
I use the info browser is...when I'm using info. Whereas with man, I
use the same less pager that I'm using to read any other text file. I
hit manpages probably a few dozen times a day. I'd be (un)lucky to hit
info as many times per month.
> , and the idea that there is more to documenting a program then merely
> listing what options it has seems okee. =20
There is _one_ problem I consistently see with manpages: a lack of
examples. A manpage should explain _what_ the function does, provide a
brief synopsis of _why_ (e.g.: when do you want to use foo), and then
***SHOW YOU HOW***. See my earlier comment today WRT setting the system
clock with 'date'. Syntax isn't presented in the manpage. Remedying
this single failing of man pages would do no end of good.
> And going through a tree like doc structure is quit common these days.
>=20
> So what is it that makes you (and others) react so vehemently?
Um. Answered? ;-)
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--wAI/bQb0EMvlZCHl
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KALZOEeIn1XyubARAo47AJ4yA1CSseEZSF2IJi3fHSbH71sspACePinT
FfFnmQpQLmKgfC/P1EyXKCw=
=Qn/G
-----END PGP SIGNATURE-----
--wAI/bQb0EMvlZCHl--
Karsten M. Self
--Ie5iOtK4e9kgqh2F
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Mon, Dec 24, 2001 at 08:57:27PM -0600, John Hasler (jo...@dhh.gt.org) wro=
te:
> dman writes:
> > Personally it is the emacs-centric interface.
>=20
> What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch, and ENTER?
How about the fact that NPU have no relationship to your _own_ path
through the documentation tree (as they would in, say, a web browser,
which is, along with 'less', the most common text-reading environment
most of us know.
I constantly find myself surprised with where info wants to take me, and
perplexed at how to get back to where I wanted to be.
> I use both info and Emacs and don't find info Emacs-like at all when
> not run from inside Emacs.
It's rather more like emacs than, say, less.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--Ie5iOtK4e9kgqh2F
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KANuOEeIn1XyubARAlOkAJ9jFwhIIjevDn6xqKJbzt66p5AaiwCfUGYf
BmPf+RpEyg5IS5gM6EeiWwY=
=YjRX
-----END PGP SIGNATURE-----
--Ie5iOtK4e9kgqh2F--
Christoph Simon
"Karsten M. Self" <kms...@ix.netcom.com> wrote:
> - It attempts to replace, not augment, an existing, established,
> viable, useful, and effective standard. This is almost always a bad
> idea. The far better tack: provided augmented functionality. If
> your solution is compelling enough, people will transition.
Right, but doesn't apply here. Info does augment manpages. First it
fulfills the purpose of the manpage, giving exact and complete
information through the reference sections, and then it augments the
limited sections of a manpage by free chapters which try to explain
the issue in human language, often with tutorials and examples.
> - It's (largely) bound to a specific viewer. Which, if you don't use
> emacs, isn't particularly usable, and is about as intuitive
> as...well, emacs. This has changed as additional viewers are
> avilable (e.g.: pinfo -- based on lynx...but, of course, I don't
> care for lynx's keybindings, and use w3m instead....), but it's not
> fully remedied the problem. The authoring markup is distinct from
> both manpages (groff) and DocBook (SGML).
Inconsequent. Man requires groff and a pager. I can't read SGML or XML
(but hate them), specially when created by some `intelligent tool'
with no human digestable linebreaks. You need a tool to write _and_ to
read them. You can write and read info files in any text viewer/editor
including less. Of cource, then you'll lose the cross references, but
this those wouldn't work with XML either. Man pages stop being useful
when they get too long. Try locating a short word in man bash. Also,
IMHO, I don't think a slash for searching forward and a question mark
for back is intuitive, at least not more than Ctrl-S (search) Ctrl-R
(reverse search). Also, I'll always consider more intuitive and human
readable \TeX syntax than .groff, .gro^_off-text or</XML>.
> - It fragments the documentation effort. GNU favors info. Debian
> favors manpages (I've written on this in the past, references
> supporting both claims are available).
Probably true, but people still can choose, unless they get
consistenly bashed for it.
> - Info and man serve different functions. Man is meant to be a quick
> reference, putting the most likely needed information -- command
> arguments and use -- first. For Info, you've got to dig to get to
> this. It's like the "entrance tunnels" that some web wanks were
> advocating for websites for about fifteen minutes in 1997. Like I
> want to open four pages to get to your site.... Info puts shit in
> the way of what I'm looking for.
Both, man and info have a learning curve. That you forgot about yours
concerning man doesn't make this untrue. One of the declared aims of
info is to provide a frame to write introductions or tutorials which
wouldn't fit well into a man page, because that is limited to a
reference manual. Info also provides that. If I already know a program
but need to remember some option or syntax, I'll use for instance
info's command reference, but if I don't know the program, and may be
really don't have an idea what it is about, I'll read the info from
the beginning to the point where I got the idea.
> - It splits a programs docs into multiple pages. In electronic
> format, single, large, documents which can be searched through
> readily (gee: that sounds like a manpage under 'less') is more
> useful than a set of smaller docs which can't be as readily scanned.
Try Ctrl-S, which will search through all files. Don't want to type
info? You also can use (z)grep (same work but less useful).
> - GNU project manpages frequently include a passage:
>
> This man page is not kept up to date except when volunteers want
> to maintain it. If you find a discrepancy between the man page
> and the software, please check the Info file, which is the
> authoritative documentation.
This is the choice of the program's authors. We'll have to respect
it. And some will be more happy (me) than others (you) :-)
>
> [Taken from the gcc manpage]
>
> ...followed by dire warnings that the manpage may not be updated,
> etc., etc. At which point the pitiless reader turns to the info
> document...which in many cases is a copy of the same manpage (now
> presented in an unfamiliar document viewer). Houston, we've got a
> problem.
This is the choice of debian maintainers.
> The basic concept is valid. However, what seems to have happened in the
> world is that we got several rival document and hypertext languages in
> about the same five minute period of 1987. Tim Berners-Lee happened to
> win the race. Ironically, man pages translate quite well to HTML, and
> DocBook / LDP has largely filled the niche Info is more suited to:
> providing a more substantial, book-styled, document suitable for
> browsing rather than a quick reference.
If I can open info on any terminal, I do prefer it over lynx and would
never want to have to wait for mozilla starting up, (which is possibly
not even installed on such machine). This sounds to me like when I had
to interview a guy for a job, who told me that he uses Star Office to
edit configuration files (my question was expecting vi, emacs,
etc).
It is obvious that man pages translate easy to HTML: they are so
simple and limited, that they translate well to anything, including
info.
> Info _was_ a really good idea at its time. Binding it to an emacs-style
> editor made a bit of sense. But things have changed, the computing
> audience has exploded beyond Richard's dreams (both in general, and for
> GNU/Linux specifically), and I'd venture to suggest that most people
> using GNU/Linux either don't know emacsen or don't use them as their
> principle editing environment. This is no longer a principally
> technical community, even on the "techincal" platform.
I'm happy belonging to a minority.
[...]
> I've got _no_ idea how to access this search function. The _only_ time
> I use the info browser is...when I'm using info. Whereas with man, I
> use the same less pager that I'm using to read any other text file. I
> hit manpages probably a few dozen times a day. I'd be (un)lucky to hit
> info as many times per month.
Try Ctrl-H to get help in info. I use info and man. My biological RAM
is very limited, but it's still enough to deal with both. Oh, and info
info works as well as man man.
> There is _one_ problem I consistently see with manpages: a lack of
> examples. A manpage should explain _what_ the function does, provide a
> brief synopsis of _why_ (e.g.: when do you want to use foo), and then
> ***SHOW YOU HOW***. See my earlier comment today WRT setting the system
> clock with 'date'. Syntax isn't presented in the manpage. Remedying
> this single failing of man pages would do no end of good.
Some manpages do, but they have not been designed for that, nor is it
their purpose. Manpages with examples miss their basic aim: to should
provide a quick but complete reference. No space for examples (nor is
there a traditional secion for examples). It would render the pages
much longer and more difficult to get an overview. There is no index
in man. Info is designed to deal well with this. Using several levels of
indices, references and cross-references, you can locate quickly any
information you are looking for without reading the whole thing once
more (you did read it the first time, did you?) OTOH, often I try
--help before using man, if I just need a reminder.
> > So what is it that makes you (and others) react so vehemently?
>
> Um. Answered? ;-)
No.
--
Christoph Simon
cic...@kiosknet.com.br
---
^X^C
q
quit
:q
^C
end
x
exit
ZZ
^D
?
help
.
Karsten M. Self
--IS0zKkzwUGydFO0o
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 03:54:36AM -0200, Christoph Simon (ciccio@kiosknet.=
com.br) wrote:
> On Mon, 24 Dec 2001 20:38:49 -0800
> "Karsten M. Self" <kms...@ix.netcom.com> wrote:
>=20
> > - It attempts to replace, not augment, an existing, established,
> > viable, useful, and effective standard. This is almost always a bad
> > idea. The far better tack: provided augmented functionality. If
> > your solution is compelling enough, people will transition.
>=20
> Right, but doesn't apply here.=20
Wrong.
If your explicit policy is "deprecate manpages in favor of info pages",
you're replacing functionality. I've documented this from the gcc
manpage, it's a typical example.
GNU Info is used to replace manpages. It leaves packages with
incomplete, outdated, or missing man pages. This is a Bad Thing=AE.
> > - It's (largely) bound to a specific viewer. Which, if you don't use
<...>
> Inconsequent. Man requires groff and a pager.=20
A pager is a non-specific reader.
> I can't read SGML or XML (but hate them), specially when created by
> some `intelligent tool' with no human digestable linebreaks.=20
I'm not talking about the composition tools, I'm talking about the
_viewing_ tools. Non sequitur.
<...>
> Man pages stop being useful when they get too long. Try locating a
> short word in man bash.=20
/\<at\>
Or, for a context search:
man bash | col | grep -w at
=2E..returns 81 lines, which can then be looked at fairly readily or
searched with additional context:
/character at point
> Also, IMHO, I don't think a slash for searching forward and a question
> mark for back is intuitive, at least not more than Ctrl-S (search)
> Ctrl-R (reverse search).=20
It's not intuitive, that's somewhat my point. But it *is* learned,
particularly if you're using more/less and/or a vi clone frequently, or
other tools (e.g.: w3m) which use vi keys.
As it happens, info allows use of vi keys (didn't know this until
today), and the C-s and C-r keys are among the emacs bindings I do
recall more readily. FWIW, I use emacs mode for command line editing in
bash. I'm not fully locked into one mode of thinking. _But_, what I
_don't_ know are multi-line and multi-buffer navigation keystrokes,
which are necessary in info.
The point isn't which mode is "more intuitive", it's whether or not the
commonly used tools allow the option to select an environment that's
comfortable to the user. Info perforce constrains this choice. It
doesn't matter if you prefer searching with '/', with C-s, or by
touching your tongue to your nose, Info restricts the tools available
to you.
> > - It fragments the documentation effort. GNU favors info. Debian
> > favors manpages (I've written on this in the past, references
> > supporting both claims are available).
>=20
> Probably true, but people still can choose, unless they get
> consistently bashed for it.
You can only choose if you're writing the documentation on your system.
As it is, Debian prefers manpages:
Debian Policy, 13.1, Manual Pages:
Each program, utility, and function should have an associated manpage
included in the same package. It is suggested that all configuration
files also have a manual page included as well.
<...>
Even though the GNU Project do not in general consider the lack of
a manpage to be a bug, we do.
> > - Info and man serve different functions. Man is meant to be a quick
> Both, man and info have a learning curve.=20
***ACCESSING*** a man page -- the task of opening and navigating the
document itself -- doesn't have a _separate_ learning process.
Accessing an info page does.
_Understanding_ a man page is admittedly an art. Ditto an Info page.
Info throws two hurdles at the user, man only one.
> One of the declared aims of info is to provide a frame to write
> introductions or tutorials which wouldn't fit well into a man page,
> because that is limited to a reference manual.=20
This is a supplemental function. This documentation shouldn't attempt
to replace something it can't: a basic usage reference.
> > This man page is not kept up to date except when volunteers want
> > to maintain it. If you find a discrepancy between the man page
> > and the software, please check the Info file, which is the
> > authoritative documentation.
>=20
> This is the choice of the program's authors.=20
For someone crowing over the advantages of Info and its benefits in
conveying information, you seem to have a fundamental comprehension
problem:
This is _NOT_ a statement of author's choice, it's _GNU's POLICY_.
> > ...followed by dire warnings that the manpage may not be updated,
> > etc., etc. At which point the pitiless reader turns to the info
> > document...which in many cases is a copy of the same manpage (now
> > presented in an unfamiliar document viewer). Houston, we've got a
> > problem.
>=20
> This is the choice of debian maintainers.
Looks to me more like someone dropped the ball. "Gee...GNU doesn't do
man pages, so we'll note that in the man page, but I don't feel like
writing an info page....".
If you want to press the point, I'll try to raise examples, others are
similarly welcome to provide same.
> > The basic concept [of Info] is valid. However, what seems to have
> > happened in the world is that we got several rival document and
> > hypertext languages in about the same five minute period of 1987.
> > Tim Berners-Lee happened to win the race. Ironically, man pages
> > translate quite well to HTML, and DocBook / LDP has largely filled
> > the niche Info is more suited to: providing a more substantial,
> > book-styled, document suitable for browsing rather than a quick
> > reference.
>=20
> If I can open info on any terminal, I do prefer it over lynx and would
> never want to have to wait for mozilla starting up
Shared perspective. Though you've other alternatives for text-mode
browsing: links, and w3m (latter is my pref).
> It is obvious that man pages translate easy to HTML: they are so
> simple and limited, that they translate well to anything, including
> info.
Yeah. Funny, when I was a kid, we called that KISS....
> > Info _was_ a really good idea at its time. Binding it to an
> > emacs-style editor made a bit of sense. But things have changed,
> > the computing audience has exploded beyond Richard's dreams (both in
> > general, and for GNU/Linux specifically), and I'd venture to suggest
> > that most people using GNU/Linux either don't know emacsen or don't
> > use them as their principle editing environment. This is no longer
> > a principally technical community, even on the "techincal" platform.
>=20
> I'm happy belonging to a minority.
How about providing choices to the rest of us.
IIRC, there's a web browser built into emacs. You can also browse
various files from within emacs. I'm not asking for access modes to
tools such as info to disappear. I'm asking that the more fundamental,
open, convention of the Unix manual pages be respected by FSF.
> > I've got _no_ idea how to access this search function. The _only_ time
<...>
> Try Ctrl-H to get help in info.=20
Thanks.
> Oh, and info info works as well as man man.
Yes. But 'man info' is rather useless (a neutered manpage is presented),
while 'info man' provides a full-featured explanation, in a single page,
of the 'man' command.
> > There is _one_ problem I consistently see with manpages: a lack of
> > examples. A manpage should explain _what_ the function does,
> > provide a brief synopsis of _why_ (e.g.: when do you want to use
> > foo), and then ***SHOW YOU HOW***. See my earlier comment today WRT
> > setting the system clock with 'date'. Syntax isn't presented in the
> > manpage. Remedying this single failing of man pages would do no end
> > of good.
>=20
> Some manpages do, but they have not been designed for that, nor is it
> their purpose. Manpages with examples miss their basic aim: to should
> provide a quick but complete reference.=20
You are grossly misinformed.
Traditionally, man pages, particularly for complex commands, would
include a few examples drawn from tasks commonly (or usefully) performed
with the command. Debian lacks strongly in this regard. I also keep an
OpenBSD box here, and would point to, for instance, man (1) find as an
example of a well-written manpage with some good examples.
OpenBSD's manpages area available online at:
http://www.openbsd.org/cgi-bin/man.cgi
> No space for examples=20
I'm sorry. You're absolutely right. We simply must save that virtual
paper by keeping manpages as short as possible.
<cough>bollux<cough>
> (nor is there a traditional secion for examples).=20
EXAMPLES
> It would render the pages much longer and more difficult to get an
> overview.=20
And somehow this isn't a problem for Info?
The synopsis and description (options summary) comes first. Examples
come later.
> There is no index in man.=20
/^[A-Z]
> Info is designed to deal well with this.
The man page format is a loosely structured, but nonetheless structured,
document. The user is welcome to use available tools of her choosing to
parse, search, scan, summarize, or simply read, the document.
> Using several levels of indices, references and cross-references,=20
=2E..which I, and the newbie, frankly find confusing.
> you can [...] quickly=20
=2E..get utterly and completely lost in the Info hierarchy...?
Oh, I'm sorry. I'm afraid I interrupted. You were saying?....
> any information you are looking for without reading the whole thing
> once more (you did read it the first time, did you?)=20
No. I'm scanning for data. The structure of the Info document, the
"indices, references, and cross references" frankly get in the way. I
_can't scan the document quickly, in one sweep_. Instead, I've got to=20
navigate multiple pages, which _always_ gets me lost.
> OTOH, often I try --help before using man, if I just need a reminder.
Ditto. See my starting post in this thread.
> > > So what is it that makes you (and others) react so vehemently?
> >=20
> > Um. Answered? ;-)
>=20
> No.
My $DEITY! I've just spent thirty minutes talking to the wall!!
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--IS0zKkzwUGydFO0o
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KFNDOEeIn1XyubARAtGpAKCAvcMqMj618yHPDOS94HBrB7QntQCfduK/
G3Ul5Uj02ikJ/o+661LVPLo=
=GT2t
-----END PGP SIGNATURE-----
--IS0zKkzwUGydFO0o--
Erik Steffl
>
> on Tue, Dec 25, 2001 at 03:54:36AM -0200, Christoph Simon (cic...@kiosknet.com.br) wrote:
...
> > One of the declared aims of info is to provide a frame to write
> > introductions or tutorials which wouldn't fit well into a man page,
> > because that is limited to a reference manual.
>
> to replace something it can't: a basic usage reference.
exactly. IMO info is a big failure, it tries to replace something it
cannot replace by it s definition (it clearly has different purpose than
the manpage) and info reader sucks big time according to opinion of
fairly huge number of people. yes, there are other ways to read info
pages but all references are alays 'run info'. the default reader should
be something usable.
info as such might be useful, man page clearly cannot cover ALL the
documentation needs. why didn't they think of an easy way to create a
man page from info? a sort of 'executive summary'?
erik
Karsten M. Self
--4Ckj6UjgE2iN1+kY
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 02:52:46AM -0800, Erik Steffl (ste...@bigfoot.com) =
wrote:
> "Karsten M. Self" wrote:
> >=20
> > on Tue, Dec 25, 2001 at 03:54:36AM -0200, Christoph Simon (ciccio@kiosk=
net.com.br) wrote:
> ...
> > > One of the declared aims of info is to provide a frame to write
> > > introductions or tutorials which wouldn't fit well into a man page,
> > > because that is limited to a reference manual.
> >=20
> > This is a supplemental function. This documentation shouldn't attempt
> > to replace something it can't: a basic usage reference.
>=20
> exactly. IMO info is a big failure, it tries to replace something it
> cannot replace by it s definition (it clearly has different purpose
> than the manpage) and info reader sucks big time according to opinion
> of fairly huge number of people. yes, there are other ways to read
> info pages but all references are alays 'run info'. the default reader
> should be something usable.
>=20
> info as such might be useful, man page clearly cannot cover ALL the
> documentation needs. why didn't they think of an easy way to create a
> man page from info? a sort of 'executive summary'?
Bingo.
I've discussed this issue in the past with Brad Kuhn (haven't quite
worked up to bringing it up with Stallman), and I copied Brad on my
entry on this thread.
There is a man-like summary which can be extracted from the Info format
and turned into something remotely resembling a man page, somewhat in
the way a cat might be said to resemble a horse: neither can operate a
can opener.
There are worse legacy standards stuck within computer concepts than the
man page. I suspect one of the attractions of Info is that it is
directly translateable into book formats, which themselves provide
revenues to the FSF. Again, it would seem that adopting DocBook would
further this aim, allow inclusion of a manpage (hey, it's another
reference section within the document).
Anyhow, I suspect I've made my views fairly clear ;-)
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--4Ckj6UjgE2iN1+kY
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KHgxOEeIn1XyubARAghJAJ9sLCa5030POKgG4wMk4/L63OnlBACfT7c8
PTG+ZlYLCvLrdurYCOJDNu8=
=rx+A
-----END PGP SIGNATURE-----
--4Ckj6UjgE2iN1+kY--
Colin Watson
> on Tue, Dec 25, 2001 at 03:54:36AM -0200, Christoph Simon (cic...@kiosknet.com.br) wrote:
> > On Mon, 24 Dec 2001 20:38:49 -0800
> > "Karsten M. Self" <kms...@ix.netcom.com> wrote:
> > > ...followed by dire warnings that the manpage may not be updated,
> > > etc., etc. At which point the pitiless reader turns to the info
> > > document...which in many cases is a copy of the same manpage (now
> > > presented in an unfamiliar document viewer). Houston, we've got a
> > > problem.
> >
>
> Looks to me more like someone dropped the ball. "Gee...GNU doesn't do
> man pages, so we'll note that in the man page, but I don't feel like
> writing an info page....".
No, usually what's happened in these cases is that the man page is in
the main package while the info pages are in a separate -doc package
which you haven't got installed. When you type info, it can't find the
real info page, so it reverts to its neutered "try to view man pages"
mode instead.
(Usual disclaimer: I'm not a fan of info either. Maybe I should turn
man-db into a GNU package, just for the sheer irony value? :))
--
Colin Watson [cjwa...@flatline.org.uk]
Brian Nelson
> - It's (largely) bound to a specific viewer. Which, if you don't use
> emacs, isn't particularly usable, and is about as intuitive
> as...well, emacs. This has changed as additional viewers are
> avilable (e.g.: pinfo -- based on lynx...but, of course, I don't
> care for lynx's keybindings, and use w3m instead....)
Funny, I prefer w3m over links and lynx because of its emacs-like key
bindings. :)
--
Brian Nelson <nel...@bignachos.com>
BigN...@jabber.org
http://bignachos.com
John Hasler
> What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch, and ENTER?
Karsten M. Self writes:
> How about the fact that NPU have no relationship to your _own_ path through
> the documentation tree...
What does that have to do with my question?
> ...as they would in, say, a web browser, which is, along with 'less', the
> most common text-reading environment most of us know.
I thought you were a man page enthusiast. Now you want html documentation?
IMHO html is a lousy choice.
Michael Mauch writes:
> What's wrong with the (L)ast key? And then, of course, you have the
> (S)earch key and most of the times an (I)ndex.
And, of course, there is 'info info' for those who actually want to learn
to use info.
--
John Hasler
jo...@dhh.gt.org (John Hasler)
Dancing Horse Hill
Elmwood, WI
David Teague
Karsten
I LIKE emacs. We were using vi as our only text editor with System V
machines in the late 80s. I found and installed Emacs, within one
week everyone on my faculty was using emacs.
That said, every other point you make here is RIGHT ON. I find info
to be arcane, inspite of its keystrokes being emacs like.
Html information browsed with a decent TEXT mode browser that is
intutive (OK I know one man's intuitive is another's nightmare)
browser inteface. OK put the keystrokes at the bottom of the page
like Pine or Pico.
Karsten's response to Carel is omitted.
On Mon, 24 Dec 2001, Karsten M. Self wrote:
> on Tue, Dec 25, 2001 at 02:44:17AM +0100, Carel Fellinger (cfel...@iae.nl) wrote:
> > On Mon, Dec 24, 2001 at 03:07:41PM -0800, Karsten M. Self wrote:
> > ...
> > > I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
> > > basic concept still sucks. It should be scrapped for a searchable
... YEA!
--David
David Teague, d...@cs.wcu.edu
Debian GNU/Linux Because software support is free, timely,
useful, technically accurate, and friendly.
(I hope this is all of the above.)
Karsten M. Self
--3MMMIZFJzhAsRj/+
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 10:06:38AM +0000, Anthony Campbell (a.campbell@doct=
ors.org.uk) wrote:
> On 24 Dec 2001, Gary Turner wrote:
> > On Mon, 24 Dec 2001 15:07:41 -0800, Karsten M. Self wrote:
> >=20
> > >on Mon, Dec 24, 2001 at 06:37:25PM +0100, Martin Emrich (debianml@emme=
s-world.de) wrote:
> > >> Hi All !
> > >>=20
> > >> When newbies ask something, they are often asked to RTFM...
> > >
> > >I consider this acceptable only IITTNTRFMTFR [1]
> > >
> > Thanks, Karsten. When I query the list, it's because I couldn't find
> > the answer in the manual nor in the 8 or 10 books on my desk. Worse, I
> > found the answer when I didn't need it and can't find it when I do.
> >=20
> [snip]
>=20
> I've often had this problem. I therefore keep a file called tips, in
> which I store useful-looking hints and answers posted here and elsewhere
> in reply to other people's queries. Even if they are not immediately
> relevant to me, they quite often become so later.
I do something similar, it's my linux/linux mail folder (dunno why I
call it that). Particularly insightful responses get stuffed there.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--3MMMIZFJzhAsRj/+
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KPpbOEeIn1XyubARAuzRAJ0XH4hwgvwowI+NJ8+zZtctOuHF+gCfbwLX
dnPssHsyAE5ZiopBlG/vN18=
=ryqK
-----END PGP SIGNATURE-----
--3MMMIZFJzhAsRj/+--
Karsten M. Self
--GmiNL4+5WUWrod5m
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 07:18:08PM +0100, Michael Mauch (michael.mauch@gmx.=
de) wrote:
> Karsten M. Self schrieb:
>=20
> > on Mon, Dec 24, 2001 at 08:57:27PM -0600, John Hasler (jo...@dhh.gt.org)=
wrote:
> > > dman writes:
> > > > Personally it is the emacs-centric interface.
> > >=20
> > > What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch, and E=
NTER?
> >=20
> > How about the fact that NPU have no relationship to your _own_ path
> > through the documentation tree (as they would in, say, a web browser,
> > which is, along with 'less', the most common text-reading environment
> > most of us know.
>=20
> I don't see how a web browser could change the order of the pages ;-)
It changes the expectation of browsing. Plus, the browser's *own*
forward / back / up buttons follow the path of your browsing session,
not of the document hierarchy.
> > I constantly find myself surprised with where info wants to take me, and
> > perplexed at how to get back to where I wanted to be.
>=20
> What's wrong with the (L)ast key?=20
Discovering it.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--GmiNL4+5WUWrod5m
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KPqsOEeIn1XyubARArzqAKCAfcc1A/pfmytaYCdHGa5kpA//GwCeKQ+V
4cXmqrDEwO/ia6/AcAuoKRk=
=2Ejx
-----END PGP SIGNATURE-----
--GmiNL4+5WUWrod5m--
Karsten M. Self
--zH41lVBEV8cLJnCl
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 01:07:23PM -0600, John Hasler (jo...@dhh.gt.org) wro=
te:
> I wrote:
> > What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch, and ENT=
ER?
>=20
> Karsten M. Self writes:
> > How about the fact that NPU have no relationship to your _own_ path
> > through the documentation tree...
>=20
> What does that have to do with my question?
14 and green ducks.
> > ...as they would in, say, a web browser, which is, along with
> > 'less', the most common text-reading environment most of us know.
>=20
> I thought you were a man page enthusiast. Now you want html
> documentation? IMHO html is a lousy choice.
It's a well known standard. I know a lot of people (including many
nontechnical ones) who spend hours in a web browser. I don't know many
people (including many technical ones) who spend comperable time in the
info browser. It's a familiarity issue. Sometimes the familiar is
superior to the "good". Say what you will about the Web, it abstracts
content from the reading tool. I can read with Galeon, Mozilla, Konq,
MSIE, w3m, lynx, links, or dumped to a textfile and paged with less [1].
> Michael Mauch writes:
> > What's wrong with the (L)ast key? And then, of course, you have the
> > (S)earch key and most of the times an (I)ndex.
>=20
> And, of course, there is 'info info' for those who actually want to learn
> to use info.
As I noted: the 'man' man page is transitive between man and info --
you can get the full man page from within info. The 'info'
documentation is assymetric: you can't get useful information from
within man, which, if it's your preferred or known environment, is where
you know how to operate. This is a Bad Thing=AE.
Having spent a half hour or so browsing info pages via Web through dwww,
I have to say that info makes worse web pages than either man or
DocBook, though the DocBook document structure resembles the info
structure largely.
Note too: with DocBook, you've got the option of splitting a document
at major section breaks, or dumping it as One Big File=AE, depending on
your SGML parsing arguments. Anyone know if Info's got a similar
functionality?
Peace.
----------------------------------------
Notes:
1. Not uncommon for me when snarfing content with lynx -- I *really*
don't care for default lynx colors and navkey bindings, and haven't
been able to grok its config file to change this. W3M wins heavily
over lynx for the former's ease of configuration.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--zH41lVBEV8cLJnCl
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KPyuOEeIn1XyubARAr8eAJ9nEKb71VpySviXQwwpHppZ0blVLQCeK4ZU
eZ74S2OQjoFMUzCqLB3BGwo=
=mmwf
-----END PGP SIGNATURE-----
--zH41lVBEV8cLJnCl--
Karsten M. Self
--iUV/lbBrmPtUT9dM
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 08:07:40AM -0600, Colin Watson (cjwa...@debian.org=
) wrote:
> On Tue, Dec 25, 2001 at 02:21:55AM -0800, Karsten M. Self wrote:
> > on Tue, Dec 25, 2001 at 03:54:36AM -0200, Christoph Simon (ciccio@kiosk=
net.com.br) wrote:
> > > On Mon, 24 Dec 2001 20:38:49 -0800
> > > "Karsten M. Self" <kms...@ix.netcom.com> wrote:
> > > > ...followed by dire warnings that the manpage may not be update=
d,
> > > > etc., etc. At which point the pitiless reader turns to the info
> > > > document...which in many cases is a copy of the same manpage (n=
ow
> > > > presented in an unfamiliar document viewer). Houston, we've go=
t a
> > > > problem.
> > >=20
> > > This is the choice of debian maintainers.
> >=20
> > Looks to me more like someone dropped the ball. "Gee...GNU doesn't do
> > man pages, so we'll note that in the man page, but I don't feel like
> > writing an info page....".
>=20
> No, usually what's happened in these cases is that the man page is in
> the main package while the info pages are in a separate -doc package
> which you haven't got installed. When you type info, it can't find the
> real info page, so it reverts to its neutered "try to view man pages"
> mode instead.
Hmm...
If this is the case, I'd suggest it's a bug in Debian, and the man page
should indicate that, not only is the Info page preferred, but that, if
it's not found on the system, the appropriate Debian -doc package should
be installed.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--iUV/lbBrmPtUT9dM
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KP0wOEeIn1XyubARAmMeAJ4vEhFzOCSZSjKVIzjfo4BpMxs5BACfYoMS
0LjwOs4EeDOWGQHy1LwaoS8=
=fNRX
-----END PGP SIGNATURE-----
--iUV/lbBrmPtUT9dM--
Karsten M. Self
--aSnC4ZPPfhCvD8sN
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 11:38:43AM -0500, Brian Nelson (nel...@bignachos.co=
m) wrote:
> "Karsten M. Self" <kms...@ix.netcom.com> writes:
>=20
> > - It's (largely) bound to a specific viewer. Which, if you don't use
> > emacs, isn't particularly usable, and is about as intuitive
> > as...well, emacs. This has changed as additional viewers are
> > avilable (e.g.: pinfo -- based on lynx...but, of course, I don't
> > care for lynx's keybindings, and use w3m instead....)
>=20
> Funny, I prefer w3m over links and lynx because of its emacs-like key
> bindings. :)
W3M has, to my mind, fewer surprises. One of the things about Info is
that an awful lot of key combinations lead you places, from whence
returning is difficult. It's also easier IMO to find and read the help
reference in W3M. Riddle me this, but I find a full-screen help page
much more useful than a half-height one such as you get with emacs, vim,
or info. I've come to use 'screen' extensively, and find that I'm more
comfortable rapidly cycling between multiple full buffers than trying to
read two half-height windows.
I don't dislike emacs keybindings per se (I find most legacy MS Windows-
centric word processors unusable because I expect to find C-a, C-e, C-k,
C-p, C-n, C-s, C-r, etc., on them. Abiword's supposed to have an emacs
mode though I haven't got it to work). That said, I find vi a more
comfortable editor generally.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--aSnC4ZPPfhCvD8sN
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KQQSOEeIn1XyubARAsMfAJ9RUqKClOlc4REdyF83mECqLIYZYwCfdQt6
ZVG2wHGC0Zfzn+6w0cmmmRE=
=AUxv
-----END PGP SIGNATURE-----
--aSnC4ZPPfhCvD8sN--
Henrik Enberg
> on Tue, Dec 25, 2001 at 01:07:23PM -0600, John Hasler (jo...@dhh.gt.org) wrote:
>> I thought you were a man page enthusiast. Now you want html
>> documentation? IMHO html is a lousy choice.
>
> It's a well known standard. I know a lot of people (including many
> nontechnical ones) who spend hours in a web browser. I don't know many
> people (including many technical ones) who spend comperable time in the
> info browser. It's a familiarity issue. Sometimes the familiar is
> superior to the "good". Say what you will about the Web, it abstracts
> content from the reading tool. I can read with Galeon, Mozilla, Konq,
> MSIE, w3m, lynx, links, or dumped to a textfile and paged with less [1].
But none of the current browsers I'm aware of has the index and
searching facilities that info has. When I'm stuck with html
documentation I'm always extremely annoyed about how hard it is to find
what I'm looking for.
[...]
> Having spent a half hour or so browsing info pages via Web through dwww,
> I have to say that info makes worse web pages than either man or
> DocBook, though the DocBook document structure resembles the info
> structure largely.
This probably has something to do with the conversion. I'm not
familiar with dwww, but I personally think that texi2html (you'll need
the texi sources) creates better html pages than anything you can get
out of a man page.
> Note too: with DocBook, you've got the option of splitting a document
> at major section breaks, or dumping it as One Big File®, depending on
> your SGML parsing arguments. Anyone know if Info's got a similar
> functionality?
texi2html does, if you have the texi sources.
Henrik
--
For every fatal shooting, there were roughly three non-fatal shootings.
And, folks, this is unacceptable in America. It's just unacceptable.
And we're going to do something about it.
-- George W. Bush
Carel Fellinger
> on Mon, Dec 24, 2001 at 08:57:27PM -0600, John Hasler (jo...@dhh.gt.org) wrote:
> > dman writes:
> > > Personally it is the emacs-centric interface.
> >
>
> How about the fact that NPU have no relationship to your _own_ path
> through the documentation tree (as they would in, say, a web browser,
> which is, along with 'less', the most common text-reading environment
> most of us know.
I see. So you're surprised by all those web pages that have next, and
previous buttons too:)
> I constantly find myself surprised with where info wants to take me, and
> perplexed at how to get back to where I wanted to be.
Ah, at last I can full heartedly agree with you:) Especially when
dealing with nested uses of the Last button, I get lost and often have
to go back to the top level of the document.
--
groetjes, carel
Karsten M. Self
--gQt10JDuGyDb0HQ5
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 11:52:45PM +0100, Henrik Enberg (hen...@enberg.org)=
wrote:
> "Karsten M. Self" <kms...@ix.netcom.com> writes:
>=20
> > on Tue, Dec 25, 2001 at 01:07:23PM -0600, John Hasler (jo...@dhh.gt.org)=
wrote:
>=20
> >> I thought you were a man page enthusiast. Now you want html
> >> documentation? IMHO html is a lousy choice.
> >
> > It's a well known standard. I know a lot of people (including many
> > nontechnical ones) who spend hours in a web browser. I don't know many
> > people (including many technical ones) who spend comperable time in the
> > info browser. It's a familiarity issue. Sometimes the familiar is
> > superior to the "good". Say what you will about the Web, it abstracts
> > content from the reading tool. I can read with Galeon, Mozilla, Konq,
> > MSIE, w3m, lynx, links, or dumped to a textfile and paged with less [1].
>=20
> But none of the current browsers I'm aware of has the index and
> searching facilities that info has. When I'm stuck with html
> documentation I'm always extremely annoyed about how hard it is to find
> what I'm looking for.
This is where the Unix philosophy takes over: simple tools, with
well-defined tasks. Browsing and navigating content is one task.
Searching and indexing it another. So you create a second tool to do
the indexing. The search/index functionality of info should be
extractable as a CGI or similar utility. A good browser (or
command-line tool) will allow you to access that CGI readily, including
by keystroke, if you wish.
> [...]
>=20
> > Having spent a half hour or so browsing info pages via Web through
> > dwww, I have to say that info makes worse web pages than either man
> > or DocBook, though the DocBook document structure resembles the info
> > structure largely.
>=20
> This probably has something to do with the conversion. I'm not
> familiar with dwww, but I personally think that texi2html (you'll need
> the texi sources) creates better html pages than anything you can get
> out of a man page.
AFAICT, dwww uses info2html.
> > Note too: with DocBook, you've got the option of splitting a document
> > at major section breaks, or dumping it as One Big File=AE, depending on
> > your SGML parsing arguments. Anyone know if Info's got a similar
> > functionality?
>=20
> texi2html does, if you have the texi sources.
Thanks.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--gQt10JDuGyDb0HQ5
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KQhbOEeIn1XyubARAqoVAJ9st1QXsZSOcJnzlPI4nyRCiub9GgCfUBM5
fc2GTxxqnQKt/7toJqcxo8A=
=XRz1
-----END PGP SIGNATURE-----
--gQt10JDuGyDb0HQ5--
Karsten M. Self
--b6Y3a1uK/7FdRsLm
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Wed, Dec 26, 2001 at 12:01:43AM +0100, Carel Fellinger (cfel...@iae.nl)=
wrote:
> On Mon, Dec 24, 2001 at 08:41:19PM -0800, Karsten M. Self wrote:
> > on Mon, Dec 24, 2001 at 08:57:27PM -0600, John Hasler (jo...@dhh.gt.org)=
wrote:
> > > dman writes:
> > > > Personally it is the emacs-centric interface.
> > >=20
> > > What is emacs-centric about (N)ext, (P)revious, (U)p, (S)earch,
> > > and ENTER?
> >=20
> > How about the fact that NPU have no relationship to your _own_ path
> > through the documentation tree (as they would in, say, a web browser,
> > which is, along with 'less', the most common text-reading environment
> > most of us know.
>=20
> I see. So you're surprised by all those web pages that have next, and
> previous buttons too:)
Previously addressed: there is a distinction between navigation native
to the _browser_ (following the browser history), and to that native to
the _content_ (following the document hierarchy). Info blends these, to
its loss IMVAO.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--b6Y3a1uK/7FdRsLm
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KQi+OEeIn1XyubARAvOtAKCL+k/70PDPRZpcaBXhQuZeb77/nQCeIqRb
lYjOrXRIhUdetiV8G8KQq4g=
=IAh7
-----END PGP SIGNATURE-----
--b6Y3a1uK/7FdRsLm--
Bud Rogers
> But none of the current browsers I'm aware of has the index and
> searching facilities that info has. When I'm stuck with html
> documentation I'm always extremely annoyed about how hard it is to find
> what I'm looking for.
Me too. And when I'm stuck with info documentation I am often
extremely annoyed about how hard it is to find what I'm looking for. I
don't think that is an info vs html issue. I think it is a problem not of
the document format or protocol, but of the structure of the document
itself. The problem is not the tool used to produce the document
but the person producing the document.
In defense of info I would say this: it predates html. AFAIK it was the
first widely known or used hypertext documentation protocol. In criticism
of info I would say this: it predates html. AFAICT it hasn't changed a
bit. We have learned a quite a bit about hypertext since info was
developed. Info was a marvel in its day, but it is IMHO simply obsolete.
Now I'm not trying to defend html in particular, although well written
html documentation can be very nice to read and quite intuitive to
navigate. So too can info, for that matter. I would much prefer well
written, well structured documentation in some more universal format, like
docbook, which can produce output to suit the reader's preference. Those
who prefer html or postscript or pdf or plain text or even info for that
matter, can read the docs in the format they prefer. That's what I'd like
to see.
--
Bud Rogers <bu...@sirinet.net>
All things in moderation. And not too much moderation either.
Carel Fellinger
> on Tue, Dec 25, 2001 at 02:44:17AM +0100, Carel Fellinger (cfel...@iae.nl) wrote:
> > On Mon, Dec 24, 2001 at 03:07:41PM -0800, Karsten M. Self wrote:
> > ...
> > > I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
> > > basic concept still sucks. It should be scrapped for a searchable
> > > format based on HTML, XHTML, or preferably something like DocBook
> > > capable of creating multiple output formats.
> >
> > What is it in the program that is so horrible to ***DESPISE*** it?
>
Help, I didn't mean to start a war,
I was only surprised by the vehement nature of your dislikes, but
I think I see. The main problem you and others have with info are:
1) you don't no the keys to use
2) the FSF put a ban on manpages
Let me start with 2. I personally don't like the deprication of man
pages either. It is a good place to lookup syntax, options and the
like. For deeper insight a suplementairy format like info is better
suited, but nothing beats a manpage if you only have to lookup the
name of an option [okee program --help]. But man bash is a nightmare.
Best would be if the man pages could be derived from the info pages,
skipping much of the deeper stuff.
The anwser to 1 is simple: learn it;) (<--- note there is a smily there)
For search you might try the vi-like / (be aware that ? is bound to
help) The nice thing of the search is that it's a multi page and even
a multi file search. So unlike web pages, the search is not restricted
to the currently viewed page, but covers all of the info document.
Or use i (and ,) to search the index, almost like using a real book:)
--
groetjes, carel
Carel Fellinger
> on Wed, Dec 26, 2001 at 12:01:43AM +0100, Carel Fellinger (cfel...@iae.nl) wrote:
...
> > I see. So you're surprised by all those web pages that have next, and
> > previous buttons too:)
>
> Previously addressed: there is a distinction between navigation native
> to the _browser_ (following the browser history), and to that native to
> the _content_ (following the document hierarchy). Info blends these, to
> its loss IMVAO.
right and wrong. The distinction is sound, and *kept* by info. Up, next and
previous are document content related, and last is browse history related.
But using last you get easily lost, because a reverse of last is missing:(
--
groetjes, carel
Karsten M. Self
--qMm9M+Fa2AknHoGS
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Wed, Dec 26, 2001 at 12:40:46AM +0100, Carel Fellinger (cfel...@iae.nl)=
wrote:
> On Mon, Dec 24, 2001 at 08:38:49PM -0800, Karsten M. Self wrote:
> > on Tue, Dec 25, 2001 at 02:44:17AM +0100, Carel Fellinger (cfelling@iae=
.nl) wrote:
> > > On Mon, Dec 24, 2001 at 03:07:41PM -0800, Karsten M. Self wrote:
> > > ...
> > > > I ***DESPISE*** info. The pinfo alternative helps somewhat, but the
> > > > basic concept still sucks. It should be scrapped for a searchable
> > > > format based on HTML, XHTML, or preferably something like DocBook
> > > > capable of creating multiple output formats.
> > >=20
> > > I don't get this, just as I don't understand people bashing Stallman.
> > > What is it in the program that is so horrible to ***DESPISE*** it?
> >=20
> ...all of Karsten's objections snipped
>=20
> Help, I didn't mean to start a war,
Whaddya expect if you fly your hijacked civillian email plane into the
spiritually symbolic twin towers of man and info ;-)
> I was only surprised by the vehement nature of your dislikes, but
> I think I see. The main problem you and others have with info are:
>=20
> 1) you don't know the keys to use
No. For the nth time: it's the binding of content to a specific
browsing tool.
> 2) the FSF put a ban on manpages
Yes.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--qMm9M+Fa2AknHoGS
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KRzrOEeIn1XyubARAjFlAJ485euA6r1D/f5l47cpHQLNi3c3VgCeJVTe
pGQOoxESWrJaJM9AEs3yr2s=
=GwB5
-----END PGP SIGNATURE-----
--qMm9M+Fa2AknHoGS--
Carel Fellinger
> on Wed, Dec 26, 2001 at 12:40:46AM +0100, Carel Fellinger (cfel...@iae.nl) wrote:
...
> > I was only surprised by the vehement nature of your dislikes, but
> > I think I see. The main problem you and others have with info are:
> >
>
> No. For the nth time: it's the binding of content to a specific
> browsing tool.
Okee, the keys are no problem then, though from most of your other
reactions it sure seemed like a really hot item. But the nth time
mentioned thing seems strange to me, as there is info2www proving the
content isn't bound to info. There are other info replacing browsers
AND I stronly believe it to be possible to derive info pages formatted
in many a ways as to serve the browser of your liking given that the
info pages are described in TeX (texinfo to be precise). Okee you
have to write a whatever-suits-you driver in TeX, but then you get to
browse them in whatever way you like. The most widely used texinfo
converter in use produces the normal info pages geared to be browsed
with an info like browser, but I doubt there is a fundamental problem
with writing your own converter. Much like nroff files can be printed
on a laser printer or viewed on a text only screen.
> > 2) the FSF put a ban on manpages
>
> Yes.
Given that my number 1 was wrong and your number 1 is mute I take it
that this is the real reason for your vehement dislike of info.
If so, it's no info you dislike, it's loosing manpages that you hate.
Let me restate then a heartedly me too.
--
groetjes, carel
Karsten M. Self
--/NkBOFFp2J2Af1nK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 05:28:56PM -0600, Bud Rogers (bu...@sirinet.net) wro=
te:
> On Tuesday 25 December 2001 16:52 pm, Henrik Enberg wrote:
>=20
> > But none of the current browsers I'm aware of has the index and
> > searching facilities that info has. When I'm stuck with html
> > documentation I'm always extremely annoyed about how hard it is to find
> > what I'm looking for.
>=20
> Me too. And when I'm stuck with info documentation I am often=20
> extremely annoyed about how hard it is to find what I'm looking for. I=
=20
> don't think that is an info vs html issue. I think it is a problem not o=
f=20
> the document format or protocol, but of the structure of the document=20
> itself. The problem is not the tool used to produce the document=20
> but the person producing the document.
>=20
> In defense of info I would say this: it predates html. =20
Actually, they're very nearly coincident.
The info changelog starts with a June 26, 1988 entry by RMS. Tim
Berners-Lee's work on HTML and the World Wide Web started at CERN in
1988:
In 1980 I played with programs to store information with random
links, and in 1989, while working at the European Particle Physics
Laboratory, I proposed that a global hypertext space be created in
which any network-accessible information could be refered to by a
single "Universal Document Identifier". Given the go-ahead to
experiment by my boss, Mike Sendall, I wrote in 1990 a program
called "WorlDwidEweb", a point and click hypertext editor which ran
on the "NeXT" machine.
http://www.w3.org/People/Berners-Lee/ShortHistory.html
Various concepts concerning hyperlinked texts have been kicked around
since, depending on your perspective and definitions, the 1980s, 70s,
60s, or 50s, with the work of Marshal McLuhan and Vannevar Bush. By the
mid-1980s, there was already a hypertext conference...and Jakob Nielsen
was there:
http://www.useit.com/papers/tripreports/ht87.html
By that time, we'd already seen Ted Nelson's Xanadu proposal, the Apple
Hypercard stack, work by Xerox (another blown PARC chance...),=20
The Nielsen report makes for IMO interesting reading, it's a good
historical referrent from just before the emergence of a number of
systems we're currently discussing. Interesting is footnote 10, which
refers to the "getting lost" problem. There are additional "early /
precursor days of the Web" reports at:
http://www.useit.com/papers/tripreports/
> AFAIK it was the first widely known or used hypertext documentation
> protocol. =20
Not quite, by 10-20 years depending on your reckoning. But one of the
earlier implementations.
> In criticism of info I would say this: it predates html.
Heh!
> AFAICT it hasn't changed a bit. We have learned a quite a bit about
> hypertext since info was developed. Info was a marvel in its day, but
> it is IMHO simply obsolete.
>=20
> Now I'm not trying to defend html in particular, although well written
> html documentation can be very nice to read and quite intuitive to
> navigate. So too can info, for that matter. I would much prefer well
> written, well structured documentation in some more universal format,
> like docbook, which can produce output to suit the reader's
> preference. Those who prefer html or postscript or pdf or plain text
> or even info for that matter, can read the docs in the format they
> prefer. That's what I'd like to see.
Agreement.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--/NkBOFFp2J2Af1nK
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KSymOEeIn1XyubARAvXmAJoCBsSlnBnzwzXj/Dw5fO1Rpu8ewQCfbdUG
U5KneqVJMWnZX61MagN+F80=
=OZ9V
-----END PGP SIGNATURE-----
--/NkBOFFp2J2Af1nK--
Brian Nelson
> on Tue, Dec 25, 2001 at 11:38:43AM -0500, Brian Nelson (nel...@bignachos.com) wrote:
> > "Karsten M. Self" <kms...@ix.netcom.com> writes:
> >
> > > - It's (largely) bound to a specific viewer. Which, if you don't use
> > > emacs, isn't particularly usable, and is about as intuitive
> > > as...well, emacs. This has changed as additional viewers are
> > > avilable (e.g.: pinfo -- based on lynx...but, of course, I don't
> > > care for lynx's keybindings, and use w3m instead....)
> >
> > Funny, I prefer w3m over links and lynx because of its emacs-like key
> > bindings. :)
>
> W3M has, to my mind, fewer surprises. One of the things about Info is
> that an awful lot of key combinations lead you places, from whence
> returning is difficult. It's also easier IMO to find and read the help
> reference in W3M. Riddle me this, but I find a full-screen help page
> much more useful than a half-height one such as you get with emacs, vim,
> or info. I've come to use 'screen' extensively, and find that I'm more
> comfortable rapidly cycling between multiple full buffers than trying to
> read two half-height windows.
Aw, come on, this is emacs. I'm sure there's a way to customize it so
that it won't split the window when showing a temporary help buffer. Or
C-x 0 (that's a zero) will get the job done too.
> I don't dislike emacs keybindings per se (I find most legacy MS Windows-
> centric word processors unusable because I expect to find C-a, C-e, C-k,
> C-p, C-n, C-s, C-r, etc., on them. Abiword's supposed to have an emacs
> mode though I haven't got it to work). That said, I find vi a more
> comfortable editor generally.
The thing that really bothers me about vi is that it's very
QWERTY-centric and feels very awkward with another layout, like Dvorak.
The most often-used keys are chosen due to their position in the home
row, and are completely meaningless in any other context (for example, j
moves down, k moves up???).
Emacs' key bindings, however, are not aimed toward any particular layout
but instead are often chosen as abbreviations of English words (C-n =
next line, C-p = previous line, C-s = search, etc.). Since the Dvorak
layout has the most commonly used letters in the home position, these
key bindings tend to feel more natural on Dvorak (at least to me).
--
Brian Nelson <nel...@bignachos.com>
BigN...@jabber.org
http://bignachos.com
Karsten M. Self
--pe+tqlI1iYzVj1X/
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
on Tue, Dec 25, 2001 at 10:55:10PM -0500, Brian Nelson (nel...@bignachos.co=
m) wrote:
> "Karsten M. Self" <kms...@ix.netcom.com> writes:
<...>
> > I don't dislike emacs keybindings per se (I find most legacy MS
> > Windows- centric word processors unusable because I expect to find
> > C-a, C-e, C-k, C-p, C-n, C-s, C-r, etc., on them. Abiword's
> > supposed to have an emacs mode though I haven't got it to work).
> > That said, I find vi a more comfortable editor generally.
>=20
> The thing that really bothers me about vi is that it's very
> QWERTY-centric and feels very awkward with another layout, like
> Dvorak. The most often-used keys are chosen due to their position in
> the home row, and are completely meaningless in any other context (for
> example, j moves down, k moves up???).
I have to agree. I taught myself Dvorak at one point (about ten days to
get over the hump), and found it worked relatively well for text
editing in, say, a standard word processor. But it absolutely blows
chunks with vi.
The problem isn't just vi, though. _Most_ Unix commands are based on
mnemonic, consonant-heavy, abbreviations: ls, cd, rm, mv, ll, who, vi,
ps, mutt, df.... Most of these are balanced between left and right
hands, leading to good natural rhythems, many are based on home-row
keys, etc. Two of the most annoying Dvorak keytrokes are 'ls' (both
right pinky) and 'cd' (right middle top row, right index home left
reach). It sounds trivial, but you end up typing these repeatedly, and
the motor memory is hard to break.
> Emacs' key bindings, however, are not aimed toward any particular layout
> but instead are often chosen as abbreviations of English words (C-n =3D
> next line, C-p =3D previous line, C-s =3D search, etc.). Since the Dvorak
> layout has the most commonly used letters in the home position, these
> key bindings tend to feel more natural on Dvorak (at least to me).
Interesting, hadn't considered that. Then again, who wants to
remember that escape-x-alt-control-left shift-b puts you into
super-edit-debug-compile mode? ;-)
I've always found the emacs keystrokes to be considerably
_anti_-ergonomic. Pessimal, really.
Peace.
--=20
Karsten M. Self <kms...@ix.netcom.com> http://kmself.home.netcom.com/
What part of "Gestalt" don't you understand? Home of the brave
http://gestalt-system.sourceforge.net/ Land of the free
We freed Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org
Geek for Hire http://kmself.home.netcom.com/resume.html
--pe+tqlI1iYzVj1X/
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE8KVPROEeIn1XyubARApe/AJ9b6jjADEnT/HipmfRtKFsF8IqiNgCfcHZu
VMNu2n6X1B5JYtpYjyvmRgU=
=VtIZ
-----END PGP SIGNATURE-----
--pe+tqlI1iYzVj1X/--
Erik Steffl
...
> mnemonic, consonant-heavy, abbreviations: ls, cd, rm, mv, ll, who, vi,
> ps, mutt, df.... Most of these are balanced between left and right
> hands, leading to good natural rhythems, many are based on home-row
> keys, etc. Two of the most annoying Dvorak keytrokes are 'ls' (both
> right pinky) and 'cd' (right middle top row, right index home left
> reach). It sounds trivial, but you end up typing these repeatedly, and
> the motor memory is hard to break.
alias the party spoilers! even without funny keyboard layout problem I
usually have l aliased to my favourite ls -??? considering there are not
many (none?) one letter common unix programs you have 26 shortest
possible command names for your ergonomic needs (YOU already know that
but I guess it might be helpful to some other members of the list,
there's a solution for almost any problem)
erik
Ted Harding
> The problem isn't just vi, though. _Most_ Unix commands are based on
> mnemonic, consonant-heavy, abbreviations: ls, cd, rm, mv, ll, who, vi,
> ps, mutt, df.... Most of these are balanced between left and right
> hands, leading to good natural rhythems, many are based on home-row
> keys, etc. Two of the most annoying Dvorak keytrokes are 'ls' (both
> right pinky) and 'cd' (right middle top row, right index home left
> reach). It sounds trivial, but you end up typing these repeatedly, and
> the motor memory is hard to break.
While that last sentence is very true, it's not an argument
for or against any particular keystroke combinations.
If you change keyboard layout, you have to re-learn. And if
you change back, you have to re-adapt.
For what it's worth, if I have a long session at the computer
keyboard (e.g. writing a report), then it takes a short while
before I can play the piano properly again. And vice versa.
It's Karsten's "motor memory" thing. I have to "flush the
motor buffers", so to speak. But this doesn't mean that
I should re-write the music, nor that I need a "piano locale"
on my computer ...
Best wishes to all,
Ted.
--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.H...@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 167 1972
Date: 26-Dec-01 Time: 09:41:17
------------------------------ XFMail ------------------------------
Colin Watson
> on Tue, Dec 25, 2001 at 08:07:40AM -0600, Colin Watson
> > No, usually what's happened in these cases is that the man page is in
> > the main package while the info pages are in a separate -doc package
> > which you haven't got installed. When you type info, it can't find the
> > real info page, so it reverts to its neutered "try to view man pages"
> > mode instead.
>
> Hmm...
>
> If this is the case, I'd suggest it's a bug in Debian, and the man page
> should indicate that, not only is the Info page preferred, but that, if
> it's not found on the system, the appropriate Debian -doc package should
> be installed.
I agree. Filing bugs like that would be worthwhile.
--
Colin Watson [cjwa...@flatline.org.uk]
Colin Watson
> I LIKE emacs. We were using vi as our only text editor with System V
> machines in the late 80s. I found and installed Emacs, within one
> week everyone on my faculty was using emacs.
Given a 1980s-era vi, I'd probably have gone for emacs too. Unbound
cursor keys, single-level undo, counter-intuitive screen updates with
'c', no backspace across line endings or even the point where you
started the current round of insertion, etc.
Fortunately vi implementations like vim have moved on considerably since
then. While they share vi's basic interface, its heritage of user
interface bugs is barely recognizable. I find traditional vi quite a
mental jolt now.
You can even have vim start up in insert mode, with graphical menus and
a completely different set of keybindings, and generally act like a less
buggy version of Windows Notepad now if you so choose, although it's not
a set-up I like myself.
--
Colin Watson [cjwa...@flatline.org.uk]
dman
| On Tue, Dec 25, 2001 at 03:25:09PM -0500, David Teague wrote:
| > I LIKE emacs. We were using vi as our only text editor with System V
| > machines in the late 80s. I found and installed Emacs, within one
| > week everyone on my faculty was using emacs.
|
| Given a 1980s-era vi, I'd probably have gone for emacs too. Unbound
| cursor keys, single-level undo, counter-intuitive screen updates with
| 'c', no backspace across line endings or even the point where you
| started the current round of insertion, etc.
|
| Fortunately vi implementations like vim have moved on considerably since
| then. While they share vi's basic interface, its heritage of user
| interface bugs is barely recognizable. I find traditional vi quite a
| mental jolt now.
I agree with this. My first encounter with 'vi' was /bin/vi on
Solaris systems. According to :version it is real vi, not a clone.
It is fine for tweaking your shell config, but not for writing code.
I tended to use nedit because it wasn't _too_ slow over a dialup and
many times better than effielbench (yeah, I learned eiffel my first
year at school, fall of '98).
Later I taught myself emacs (with help from Harley Hahn's book).
However the following quarter I had to use win95 systems on which we
were not "allowed" to install software. vim fit on a floppy so I used
it (better that DOS "edit"). I learned how to configure vim to be
very comfortable, and I found the vi-style commands easier to
remember.
Now my IDE consists of vim in combination with a Unix environment
(cygwin if it must be).
-D
--
What good is it for a man to gain the whole world, yet forfeit his
soul? Or what can a man give in exchange for his soul?
Mark 8:36-37
dman
...
I'm not sure where in the thread my comments belong so I'll put them
here.
| So what is it that makes you (and others) react so vehemently?
I've tried to use info a few times, and it is always difficult. The
keybindings are unnatural to me -- I use vim, less, and bash regularly
(readline is set to "input-mode vi"). The info keys seem much more
like emacs (which I've used, but rejected in favor of vim). Sometimes
I've managed to use it a bit, but other times it tells me there is no
link under the cursor (what, "Next" isn't a link!?). I use man a lot
since I know the layout of a manpage and I can view it easily ('less').
I think that info is ok for a book-like document. I know that many
GNU tools have nice HTML and PS/PDF documents available on gnu.org,
and the TOC resembles what I've seen in 'info'. I also like how the
reader has the choice (for HTML anyways) of one-massive-page,
one-page-per-chapter, or one-page-per-section. I like the second for
on-line viewer and the first for printing.
If someone made decent manpages (for reference), _and_ if they made a
info2vim converter then I could be happy :-). If you don't already
know : vim allows for hyper-links (start with :help) that can be
followed with ^] and ^T takes you back where you were before.
-D
--
He who belongs to God hears what God says. The reason you do not hear
is that you do not belong to God.
John 8:47
Erik Steffl
...
> info2vim converter then I could be happy :-). If you don't already
> know : vim allows for hyper-links (start with :help) that can be
> followed with ^] and ^T takes you back where you were before.
and for those who really didn't know: you can 'hyperlink' your code
(at least c, c++, perl, probably other languages as well) using *tags
programs (etags for c/c++ and maybe others, ptags for perl).
erik
John Hasler
> on gnu.org, and the TOC resembles what I've seen in 'info'.
With good reason. Try 'apt-cache show texinfo'.
--
John Hasler
jo...@dhh.gt.org (John Hasler)
Dancing Horse Hill
Elmwood, WI
dman
| dman wrote:
| ...
| > info2vim converter then I could be happy :-). If you don't already
| > know : vim allows for hyper-links (start with :help) that can be
| > followed with ^] and ^T takes you back where you were before.
|
| and for those who really didn't know: you can 'hyperlink' your code
| (at least c, c++, perl, probably other languages as well) using *tags
| programs (etags for c/c++ and maybe others, ptags for perl).
The command is 'ctags', the package is "exuberant-ctags". It supports
something like 15 languages now. It's really cool. See the project
site on Sourceforge if you want more details.
-D
--
If we claim to be without sin, we deceive ourselves and the truth is not
in us.
I John 1:8