Re: [dwm] still simplicity or featureitis?

From: Anselm R. Garbe <>
Date: Thu, 13 Mar 2008 17:46:33 +0100

On Thu, Mar 13, 2008 at 01:07:01PM +0100, markus schnalke wrote:
> I remember a time (probably between version 4.1 or 4.4 or so), when
> the development of dwm became more and more stable. dwm began to reach
> a point near "finished" (or "perfect"). Changes became smaller, and
> mostly it was optimizing the existent features and simplyfing.
> With the multihead thing comming up, I realize a lot of new ideas in
> different directions. Changes became big again and everything is a
> little bit experimental.
> Also complexity increased and the code base as well.

I don't think your observation is true.

; wc -l dwm-4.7/dwm.c
1914 dwm.c
; wc -l dwm-4.7/config.def.h
93 config.def.h
; wc -l dwm-4.7/dwm.1
156 dwm.1

Binary size of dwm-4.7:
-rwxr-xr-x 1 anselm anselm 28048 2008-03-13 16:40 dwm

; wc -l dwm-4.8/dwm.c
1911 dwm.c
; wc -l dwm-4.8/config.def.h
93 config.def.h
; wc -l dwm-4.8/dwm.1
157 dwm.1

Binary size of dwm-4.8:
-rwxr-xr-x 1 anselm anselm 28048 2008-03-13 16:38 dwm

So dwm-4.8 is not bigger or more complex than 4.7.

The differences in both are -- dwm-4.8 allows for a more flexible
setup regarding master area, bar position, and tile area.
dwm-4.8 also allows for easier integration with new layouts and
it comes packed already with 4 different layouts. It does not
contain setmwfact(), and togglebar(), because these should be
changed in a user-defined function, similiar to config.anselm.h.

It also does not include togglemax() anymore in favor for

There have been also a lot of code cleanups and polishing, so
imho the stability should be similiar as in dwm-4.7 again.

> I think special stuff that only few people use should not be
> included in core (or mainstream) dwm.
> These functionality should be provided as patches instead.

Well, if you'd consider the current code base more carefully,
you would have noticed that dwm-4.8 is exactly about this.
Compare config.def.h to config.anselm.h to see how your remark
is realized.

> (Patching core dwm should be seen as central characteristic of dwm
> usage, in my eyes.)


> I would like to see dwm coming back to this nearly finished state
> again, because it was nearly finished ... and now it's again in this
> "featureitis" trap.

I don't think you are right in this regard. There were some
experiments during the last months, but most of them disappeared
already again.

Kind regards,

 Anselm R. Garbe >< >< GPG key: 0D73F361
Received on Thu Mar 13 2008 - 17:46:33 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 15:25:30 UTC