Emacs 22 on April 23

[Posted April 10, 2007 by corbet]

From:		Chong Yidong <cyd-AT-stupidchicken.com>
To:		emacs-devel-AT-gnu.org
Subject:		Pretest schedule
Date:		Sun, 08 Apr 2007 10:21:49 -0400
Archive-link:		Article, Thread

I would like to propose the following plan for the 22.1 release.

I will roll the 22.0.98 pretest tarball on Monday, April 16 (8 days
from now).  I believe this should be the final pretest, unless a major
problem arises.

If no major problems with the April 16 pretest come to light, we
should cut the Emacs 22 CVS branch on Friday, April 20, and release
Emacs 22.1 on Monday, April 23.

(Log in to post comments)

Posted Apr 10, 2007 17:03 UTC (Tue) by smitty_one_each (subscriber, #28989) [Link]

Offhand quetion: what applications still in development are older than emacs?

Emacs 22 on April 23

Posted Apr 10, 2007 17:35 UTC (Tue) by flewellyn (subscriber, #5047) [Link]

Some government-owned databases still use COBOL-based storage systems written in the 1960s. These are kept running and updated because the agencies that run them either can't afford to change, or don't care enough to bother changing.

But, as far as end-user programs that are used on desktops, I think Emacs may qualify as the oldest still in use.

Emacs 22 on April 23

Posted Apr 10, 2007 18:19 UTC (Tue) by daney (subscriber, #24551) [Link]

AIUI there have been several versions of Emacs. I am not sure when GNU Emacs was created, but I believe that it was sometime in the '80s.

Surely things like FORTRAN could be considered to be much older.

Emacs 22 on April 23

Posted Apr 11, 2007 3:01 UTC (Wed) by flewellyn (subscriber, #5047) [Link]

Well, yes, Emacs was originally created as a set of TECO macros, but it was by the same person
who created GNU Emacs, RMS.

As for FORTRAN...yes, the language is much, much older. But implementations of it are probably
not.

Emacs 22 on April 23

Posted Apr 10, 2007 17:48 UTC (Tue) by dskoll (subscriber, #1630) [Link]

I can't think of any major apps older than Emacs that are still in development, but if you consider Unix to be a kernel plus a set of little applications (ls, cp, etc...) then those Unix apps are older.

Emacs 22 on April 23

Posted Apr 10, 2007 18:51 UTC (Tue) by tnoo (subscriber, #20427) [Link]

vi? It seems that the flame wars are as old as computing.

(funny cartoon here http://www.io.com/~dierdorf/emacsvi.html)

Emacs 22 on April 23

Posted Apr 10, 2007 20:14 UTC (Tue) by josh (subscriber, #17465) [Link]

That might hold true for the original vi, and some people do still use that, but most vi users now use one of the improved variants of vi, such as vim. One of the two main emacs editors in common use (GNU Emacs) comes from the original emacs. (I don't know whether GNU Emacs or XEmacs has more users.)

Emacs 22 on April 23

Posted Apr 10, 2007 20:50 UTC (Tue) by atai (subscriber, #10977) [Link]

Modern Emacs might conceptually come from the old Emacs (due to the inventor being Richard Stallman) but there was no continuity in the code (the current GNU Emacs was a rewrite by RMS in C)

Emacs 22 on April 23

Posted Apr 10, 2007 22:51 UTC (Tue) by JoeBuck (guest, #2330) [Link]

XEmacs is just a fork of GNU Emacs, that happened back in the version 18 days. It has plenty of RMS's code in it.

viper-mode

Posted Apr 10, 2007 23:47 UTC (Tue) by ncm (subscriber, #165) [Link]

This might be an appropriate place to mention viper-mode, which is a major mode (note, "mode"; emacs is very far from "modeless", by any definition) that implements Vi within Emacs. Since the command sets are almost disjoint, viper-mode is Vi and Emacs at the same time. That is, in viper-mode, almost all Vi and Emacs commands just work, with no mode switching needed. In practice that means that in addition to Emacs's regular command mode, entered with Alt-X, you get a Vi command mode, entered with the usual ESC.

Viper-mode in Emacs 22 works fine. I've been using it for 6 months in the beta releases.

viper-mode

Posted Apr 11, 2007 0:51 UTC (Wed) by bronson (subscriber, #4806) [Link]

In addition, there's a collaborative effort to create a vim-like mode for Emacs: http://www.emacswiki.org/cgi-bin/wiki/vimpulse.el

It's basically vi plus a few of the features that make vim really nice to use.

viper-mode

Posted Apr 11, 2007 1:43 UTC (Wed) by jasonspiro (guest, #38047) [Link]

Scott, thanks for the plug and for pointing me to the it. :)

All: I maintain vimpulse. Have you tried it? How do you like its visual text selection mode (press the letter v to enable) and its other features? What could be improved?

Emacs modes

Posted Apr 11, 2007 8:08 UTC (Wed) by ldo (guest, #40946) [Link]

... emacs is very far from "modeless" ...

Which is why my .emacs contains the following item in its custom-set-variables list:

'(auto-mode-alist (quote (("" . fundamental-mode))) t)

That disables all the modal cruft, and lets me use it as a nice, simple editor.

Emacs modes

Posted Apr 11, 2007 8:52 UTC (Wed) by ncm (subscriber, #165) [Link]

What, with no way to do search-and-replace? I guess that's simple. But how do you open a file? Ctrl-X switches to an unnamed mode, and Ctrl-F after that switches to "minibuffer-mode", and then Enter switches back to edit mode.

"Modeless emacs" was always a transparent fiction, from back when "modeless" was equated to "good": "Emacs is obviously good, therefore it must be modeless", facts notwithstanding. ("Object-oriented" took over that linguistic niche a little later.)

But viper-mode is a pretty good mode, as modes go.

viper-mode

Posted Apr 23, 2007 19:51 UTC (Mon) by jibal (guest, #44844) [Link]

"emacs is very far from "modeless", by any definition"

There is no comparison between emacs' modes, which are explicitly specified by switching between buffers or turning modes on and off, and vi's bimodal editing where every keystroke has two radically different meanings that one is constantly switching between (if they're actually editing, and not just typing a lot of new text). emacs's modes are more akin to different programs having different menus, or different user interfaces -- operations are context-sensitive, which is perfectly reasonable. But vi's modes both operate in exactly the same context. As Bill Joy said (http://web.cecs.pdx.edu/~kirkenda/joy84.html):
"One of the good things about EMACS, though, is its programmability and the modelessness. Those are two ideas which never occurred to me."

viper-mode

Posted Apr 23, 2007 20:00 UTC (Mon) by jibal (guest, #44844) [Link]

Here's another quote from Bill Joy (http://www.theregister.co.uk/2003/09/11/bill_joys_greates...) that is very revealing about its design, and the appropriateness of that design to the tasks at hand:

"It was really hard to do because you've got to remember that I was trying to make it usable over a 300 baud modem. That's also the reason you have all these funny commands. It just barely worked to use a screen editor over a modem. It was just barely fast enough. A 1200 baud modem was an upgrade. 1200 baud now is pretty slow.

9600 baud is faster than you can read. 1200 baud is way slower. So the editor was optimized so that you could edit and feel productive when it was painting slower than you could think. Now that computers are so much faster than you can think, nobody understands this anymore.

The people doing Emacs were sitting in labs at MIT with what were essentially fibre-channel links to the host, in contemporary terms. They were working on a PDP-10, which was a huge machine by comparison, with infinitely fast screens.

So they could have funny commands with the screen shimmering and all that, and meanwhile, I'm sitting at home in sort of World War II surplus housing at Berkeley with a modem and a terminal that can just barely get the cursor off the bottom line.

It was a world that is now extinct. People don't know that vi was written for a world that doesn't exist anymore - unless you decide to get a satellite phone and use it to connect to the Net at 2400 baud, in which case you'll realize that the Net is not usable at 2400 baud. It used to be perfectly usable at 1200 baud. But these days you can't use the Web at 2400 baud because the ads are 24KB."

Emacs 22 on April 23

Posted Apr 10, 2007 19:41 UTC (Tue) by job (guest, #670) [Link]

Lots. TeX is from the early eighties (or even seventies, depending on how you count), and it's still what most scientific articles are written in.

Emacs 22 on April 23

Posted Apr 11, 2007 2:28 UTC (Wed) by bfields (subscriber, #19510) [Link]

TeX is from the early eighties (or even seventies, depending on how you count), and it's still what most scientific articles are written in.

TeX is certainly still in use, but it would be a stretch to call it still "in development". (That may be because it's pretty much perfect, not because it's abandoned, but the point still stands.)

Emacs 22 on April 23

Posted Apr 11, 2007 15:49 UTC (Wed) by eklitzke (subscriber, #36426) [Link]

As you pointed out, TeX isn't really in development any longer. After TeX 3.0 came out (in 1989) Knuth announced that there would be no major changes to the language, and since that time the only changes have been bug fixes. Additionally, Knuth has stated that after his death there would be no more bug fixes -- any remaining bugs would be language "features."

In theory, LaTeX (which is just a really extensive set of TeX macros) is still under development, and at some point LaTeX 3 will be released. However, the latest version (LaTeX 2e) was released something like ten years ago, and the initial development of LaTeX 3 actually predates that release. AFAICT there isn't a clear set of release goals for the project, and I think there just isn't enough interest in such a radical change. Most people have a hard enough time with the current version without having to deal with a major language change, and since the release of LaTeX the number of people who understand TeX internals well enough to contribute to such a project is in decline. In practice, the only development that goes on for LaTeX is new macro packages (or more often, maintenance to existing packages).

Emacs 22 on April 23

Posted Apr 11, 2007 21:48 UTC (Wed) by wjhenney (guest, #11768) [Link]

As you pointed out, TeX isn't really in development any longer.

Well, yes and no... Knuth's original license required any derivative work to be called something other than TeX, but when you run, say, "latex" in any modern TeX distribution you will probably be invoking pdfeTeX under the hood, which is still under active development. As far as I know, it still incorporates much of Knuth's code.

TeX is still actively developed

Posted May 3, 2007 8:59 UTC (Thu) by jschrod (subscriber, #1646) [Link]

TeX *is* still in development. Owing to DEK's license, it is called pdfeTeX now, with regular new releases (including new features). There is also a separate development tier, Omega and Aleph, for better typesetting of international material; and also XeTeX that includes completely different font handling. And since Emacs 22 does not have much code in common with the original Emacs, we can also add research projects like exTeX to the mix.

In addition, there is work ongoing to merge many of these development tiers.

And almost all this work builds upon DEK's code base, so it is a continuation of the original program.

For reference: I'm one of the TeX guys; a CTAN maintainer, and a member of the LaTeX core team. I don't do core development, though; but I meet the guys regularly who do so.

Joachim

PS: What I don't know: How old is the FSF Emacs codebase, actually? TeX has seen two major implementations: The first in 1978 (in SAIL Pascal), and a complete rewrite in 1982 (in Web). Since then, the same codebase is used. How does this compare to Emacs?

old apps

Posted Apr 10, 2007 21:52 UTC (Tue) by rfunk (subscriber, #4054) [Link]

It might depend on how you count the origins of Emacs, and the definition
of application. Do you count from the TECO days? Does Sendmail count as
an application?

Emacs 22 on April 23

Posted Apr 11, 2007 21:38 UTC (Wed) by oak (guest, #2786) [Link]

X window system is almost as old as Emacs and still/again in active development.

CCM

Posted Apr 12, 2007 1:07 UTC (Thu) by sanxiyn (guest, #44599) [Link]

Wikipedia gives 1984 for starting date of GNU Emacs. CCM, the Community Climate Model, is an elaborate simulation of global atmosphere written in Fortran. It was first created in 1983, and is still in development with pretty much all codes inherited.

See http://www.ccsm.ucar.edu/overview/ for their history. Their last release is CCSM 3 in June 2004. Changelog http://www.ccsm.ucar.edu/models/ccsm3.0/notable_improveme... is available. CCSM 4 is planned to release by 2008.

Maxima

Posted Apr 12, 2007 1:17 UTC (Thu) by sanxiyn (guest, #44599) [Link]

Ah, there's also Maxima. How could I forget that! It's a genuine descendent of 1960s code, and still in active development.

http://maxima.sourceforge.net/

You can read its source and still find conditional compilation directives for 1970s hardwares long discontinued. Well, they are pretty much useless now, so people are working on cleanup.

Emacs 22 on April 23

Posted Apr 23, 2007 19:14 UTC (Mon) by jibal (guest, #44844) [Link]

April 23 of what year?

Emacs 22 on April 23

Posted Apr 23, 2007 19:19 UTC (Mon) by corbet (editor, #1) [Link]

This year. They won't hit that date, but it does not look like they will miss it by more than a few days.

Emacs 22 on April 23

Posted Apr 23, 2007 21:37 UTC (Mon) by jibal (guest, #44844) [Link]

Silly me, I should have clicked on the archive link (it dawned on me that it was there when I started thinking about what you might be "looking" at).

Emacs 22 on April 23

Posted Apr 26, 2007 20:18 UTC (Thu) by jibal (guest, #44844) [Link]

Having read http://news.gmane.org/gmane.emacs.devel for a while, I think "a few weeks" is more accurate.