lkml.org 
[lkml]   [1998]   [Mar]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: GGI Project Unhappy On Linux
On Wed, 25 Mar 1998, Boris Tobotras wrote:

> GGI Project Unhappy On Linux!
I'm not crying about that...

> build a graphics subsystem that will have to compete with the likes of
> DirectX. IMHO, their software is getting bashed before its even run by
> others.
Hmm.... I'm sorry, but I have faced that kind of attitude from the GGI people
that would scare off even the most offence-proof people. When I suggested (in
a private talk - no names) that I write a driver for a) the Trident cards and
b) a VESA 2.x interface I heard what follows:

a) Tridents are SHIT! Nobody uses them and who does, either simply cannot
afford a better card or doesn't know what are graphics cards for. Well, no
comments...
b) VESA is a DOS-only thing and WE (i.e. THEM) don't need it.
c) They have their own programmers and the DON'T NEED anyone else

I'm sorry, but given such a response I won't support these people. And every
project is in the most part PEOPLE, only then SOFTWARE - look at Linux itself.
It's people who created it and it's people who make it the most impressive
phenomenon of the last years! GGI guys say they DON'T need us (ok, they don't
need me - but read the docs in the more recent SVGAlib dists...) - well, why
would WE care?
If I was paranoid I'd say it's racism behind them - because they're Germans
(for the most part) and I'm a Pole... But that would be ridiculous, of course
;-)

> Risks
>
> GGI has worked hard with the Linux OS to bring a graphics subsystem into
> being. SVGAlib, X servers that run suid, and risky hardware detection are
The latest X servers don't run suid root anymore and I'm personally working on
a library that won't run suid root as well.

> not good for the Linux OS. We need these features to compete with other
> OS's, and to converge with them. Moreover, GGI is needed if Linux is to
> truly run on multiple architectures. It is the only solution in development
> that allows one graphical non-X Linux program to run on an Amiga, and then
> an x86 machine. X is nice, but its slow for GAMES. I want to be able to take
Hmm... I suppose that games are not on the list of the top 10 priorities of
the Linux people in the world. Sure, they make software more popular with the
masses, but hey!, Linux is still considered to be TOO difficult for Joe Blow
to use it! Let's first show it isn't and they think about games and stuff like
that - things like KDE, state-of-the-art and good looking software like
StarOffice 4 - these are the top priorites that will attract PROFESSIONALS and
they will be followed by the average man-in-the-crowd.

> advantage of fast "no X required" apps so I can whup my friend's asses.
Well, haven't you got these around? Lincity, Doom and probably several more I
don't know of (but then, I'm not a gamer...)

> SVGAlib sucks. It was great while it lasted, but its card support and model
> does not move well across platform.
That's just because there's no real maintaier for it.

> What to do?
>
> Firstly, GGI faces a unique chicken-and-egg problem. Without people trying
> the thing out (even if it blows up) then there is little understanding. So
> try it out, if you can. And complement these guys on the work so that they
A friend of mine has been testing GGI for a while and he found a serious bug -
after submitting the detailed description and recipe on how to reproduce it,
all he received in answer was... silence. And the bug is still there (I don't
remember WHAT exactly it is, but I trust his words) - even though he sent the
message THREE times, so there's no excuse "they're overloaded"

> insult. I have seen the blanket statement "GGI will just bloat things" many
It will. GGI tries to be everything for everyone - and that IS a bloat. I like
the general idea of GGI, but that's it. The philosophy behind it is sick. I
don't want to sound hostile, but isn't such a monopolistic approach what makes
Winblowz such a piece of crap? All we need in the kernel is an interface to a)
set video modes, b) manage the memory mapping issues and c) handle VC
management on behalf of the applications - all that in the hw-indep manner, of
course. And GGI goes WAY beyond it - and that's a bloat.

> If there is a legitimate technical reason why GGI should not go in the
> kernel, then it should be proven thoroughly. And we should debate openly,
> with discrete reference, and objectively, with the understanding that at the
> end of the day we are all working for the good of humanity.
That's the ideal. But to make it happen, there have to be TWO parties that
talk - and I'm sorry, but I'm not ever gonna be a GGI advocate. Sure, I'll use
it if it becomes a standard, but I won't be happy about that. And, if time
permits, I'll try to start a project of my own that implements a simple and
efficient interface - just the elements we need, nothing else.

> GGI could bring 3d graphics libraries into every linux distribution as
> standard, and move many critical things into userspace daemons and libs.
There's already an OpenGL implementation for Linux (even with a Doom OGL
version - http://threeware.planetquake.com/linux/glquake.html) and dozens of
3D libs of all sorts (http://cg.cs.tu-berlin.de/engines.html)

> Without GGI, the promise of games doesn't look good. Without GGI, graphical
> console programs will remain messily bound to the hardware, without GGI, I'm
> worried that Linux will lose its competitive edge. But I could be totally
> wrong about this! And who would know unless we all started talking about it?
Is graphics all that counts in the world? Linux wins on all the other fields -
speed, security, efficiency, hardware support, software available, support,
price - all that is much more important than graphics, IMHO (but that's just
it - MHO)

> Give these developers a chance to show you their stuff Its too early now to
> tell how good GGI will be. Just open up the issue
I tried to do it - I failed, they showed me they don't give a f*k (sorry...
I'm getting a little nervous about that ;-))) about someone else than a
closely knit bunch of beer buddies of their own...

> BSD Getting There First Is Not The Issue
>
> The GGI guys are porting to BSD. But IMHO, Linux would bring the technology
> more to the front. The BSD developers have shows more interest than the
> Linux developers, and, it seems to me that this is a strange reversal of
> behaviour. I remember a time when Linux was considered more "open" to core
> development than BSD. It seems that the reverse is true for GGI on Linux. If
> GGI is adopted by both BSD and Linux, both win.
GGI is an exception to the rule here, IMO. It isn't a proof of anything - nor
the GGI value or Linux community blindness...

> Apologies to everybody I have offended. Just getting people to think about
> this is good. I have faith that Linus, as arbiter of what goes in the tree
> and what doesn't, will make the right choice GIVEN THAT PEOPLE GIVE HIM THE
> RIGHT INFORMATION Both sides of the argument need to do this, and approach
> the problem of graphics support together.
You're surely right - TOGETHER...

---
If God intended Men to Smoke, He would have put Chimneys
in their Heads


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:41    [W:0.071 / U:1.248 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site