Re: What should we do with src:fteqcc (and Nexuiz) ?

Simon McVittie <[email protected]> Fri, 31 Jul 2026 10:04:30 +0100
Newsgroups gmane.linux.debian.devel.games
Message-ID <[email protected]>
On Thu, 30 Jul 2026 at 10:57:36 +0200, Sébastien Noel wrote:
>From what I understand, the almost 17 years old version [of fteqcc]
>in the archive 
>has only one purpose: build the nexuiz-data package.
>That old version doesn't build LibreQuake.
>A newer version doesn't build Nexuiz.
>
>Considered options:

>* remove Nexuiz ?

As one of the people who was keeping Nexuiz Classic on life-support, I 
have no objection to this. Back in 2011 I said:

     The facts are that it's a dead-upstream game, running on an
     upstream-maintained engine that we want in the archive anyway (for
     Xonotic and (non-free) Quake 1). As long as it isn't a burden and
     has users, we might as well keep it; but if it becomes a burden,
     then it needs either an active maintainer looking after it, or
     removal.
     — https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=646377#20

and I still think that's the case, except for the fact that I'd also 
hesitate to call darkplaces "upstream-maintained" at this point (its 
most recent beta release was in 2018).

Now that you want to use fteqcc for librequake, nexuiz-data has become a 
burden, so if we follow my logic 15 years ago, it's time to remove it.

If fans of multiplayer deathmatches want a game derived from Nexuiz 
Classic in Debian, its upstream replacement, Xonotic 
(https://bugs.debian.org/646377) would be a better choice. But that 
appears to be blocked by 
https://gitlab.com/xonotic/xonotic-data.pk3dir/-/work_items/3016 which 
is an upstream licensing issue in a file copied from elsewhere 
(*hopefully* easily resolved by replacing the offending file with a 
relicensed newer upstream version, if someone wants to work on that).

If Debian 14 users particularly want to play this unmaintained FPS from 
2009, there are alternatives to having it *in* Debian 14: they can grab 
the nexuiz and nexuiz-data packages from Debian 13 (which isn't going 
away), or they can install the Flatpak app 
<https://flathub.org/en-GB/apps/com.alientrap.nexuiz-classic> which has 
the advantage of some amount of sandboxing around an engine that I 
assume probably has unfixed security vulnerabilities. (Or, of course, 
they can volunteer to take responsibility for it, and re-introduce it in 
Debian.)

>* introduce another source package ('fteqcc-modern' ?)

I think if we were going this route, it would be better to vendor a 
sufficiently old copy of fteqcc into nexuiz-data and remove its B-D on 
fteqcc, allowing "the real fteqcc" to be updated. But fteqcc itself 
occasionally needs patching to compile correctly with modern gcc, so 
that would just mean that nexuiz-data gets kicked out of testing next 
time there's a bug like #1096662. I think removing nexuiz is likely to 
be better, particularly if nobody in the Games Team actually plays it.

>* fix the Nexuiz code to build with a recent version. I did it already.
>  But with QuakeC, "build successfully" != "no crash at runtime"
>  I got a first crash at startup & fixed it, but I fear I just opened a
>  gigantic can of worms :-/

I agree that seems too fragile unless there is someone genuinely 
maintaining nexuiz.

>  (I'm not going to do this, I don't use/play Nexuiz)

That's why I removed myself from its Uploaders, too.

     smcv