Politics [was: Re: Possible WM Documentation Project]

Phillip Steinbachs wrote:
> 
> 
> On Thu, 10 Mar 2005, Alex Perez wrote:
> 
>> Whaaa! *hands you a tissue* Excuse me, but when you write a patch and 
>> it doesn't get applied for what you highly suspect are purely stupid 
>> political hatreds, I think I have every right to be somewhat ticked 
>> about it. You have no right to tell me to "get over" it when you have 
>> no real understanding of what's actually going on. The WM team has 
>> revoked CVS access from people who actually had cvs commit privelages 
>> to the wm2 branch who were pro-gnustep. They have a very clearly 
>> established track record of stonewalling and also of completely 
>> disappearing for months on end and then getting all defensive when 
>> people threaten to fork. Since they're back to their same old 
>> shenanigans, I suspect you can expect to see a fork by some other 
>> folks of some form here in the next six months.
>>
> 
> Alex,
[snip]
> 
> For the record, your comments and attitude towards the Window Maker 
> developers have already gotten you into trouble within the GNUstep 
> development team for misrepresenting the GNUstep project.

That was not intentional, and the site was updated within 24 hours, so 
please, do not blow that aspect of it out of proportion.

>  Your 
> unilateral decision to frame Window Maker as not having had any real 
> ties with GNUstep was largely unsuccessful.

This is a gross misinterpretation of what we said. First of all, it was 
not unilateral. It was agreed upon between several individuals, but not 
OK'd by Adam. From your perspective, it might appear as unilateral, 
however, from mine, it was anything but. All of the wording was a 
collaborative effort, and it was only done because there was a lot of 
misunderstanding as to what GNUstep was. I /never/ attempted to frame 
Window Maker as not having had any real ties with GNUstep. That is a 
patently false accusation. I myself am fully aware of the origins of 
Window Maker as well as the relationship that historically existed 
between the two projects. You seem to assume I do not, which is 
completely incorrect.

>  Window Maker has a lot of 
> history in supporting and promoting the GNUstep project.

While this may have been historically true, it's not been in the recent 
past. Actions speak louder than words in this case.

>  The Window 
> Maker developers have not always agreed with some of the decisions the 
> GNUstep project has made and vice-versa, but that is healthy discourse 
> between two open source projects with complementary goals.   Why you
> continue to paint the Window Maker project and developers as being 
> anti-GNUstep or having "stupid political hatreds" toward GNUstep is 
> beyond any of us.

Is the patch applied? No. Why (this is not a rhetorical question)? I 
even sent it twice, once months ago and one much more recently. There's 
been no explanation, no "we will examine this later", nothing. Actions 
speak louder than words, and there has been zero action on this front. 
Does this mean that the WM devs are harboring something against me 
personally? Not necessary, but it's entirely possible. Clearly, there is 
a problem, and clearly, that problem is not being addressed. There need 
to be a few more people who have review and commit privs for trivial 
stuff like my patch which currently are lost on the cutting-room floor. 
This is the underlying issue, and it's not even being recognized as an 
issue by Alfredo or Dan. This is a very common problem with open source 
projects whose core developers decide to enjoy real life instead, and 
instead of responsibly delegating to others who are interested in 
continuing to actively contribute to and manage the project, just 
disappear and come back to "their baby" whenever they feel like it, 
choosing to let it languish the rest of the time. This is not responsible.

> 
> As for revoking the cvs access to the wm2 branch, yes, we did do that.  
> It was not without explanation to those involved, and it was generally 
> agreed upon by all parties.  We encouraged the creation of a wm2 branch 
> instead of a fork, because forks are not always the best way to resolve 
> differences.
I agree completely. It's best to not do so, but clearly the core 
developers did not learn anything from that debacle, because they've 
reverted right back to their same old behavior. Sometimes forks are 
unavoidable, in order to force political changes. (see egcs for an 
example of an ultimately positive outcome if such a process)

>  Fragmenting a user base over small things like patch 
> submissions, code usefulness, and eye-candy is counter productive. 
> Alfredo and Dan set standards for the quality of work that goes into the
> project.  The majority of patches that get submitted don't live up to 
> that because they're not well thought out and tend to break other parts 
> of the code.  Ultimately it's their project, and their perogative to 
> decide what goes in or out.  Some perceive that as arrogant, and others 
> take it for what it is.

If they bothered to even review my patch, which *I EXPECT*, and have no 
problem with, I could revise it and make the required changes which they 
feel should be necessary to apply the patch. They didn't even grant me a 
virtual audience, so to speak, and that's what irritates me more than 
anything else. If they want to do other things, that's perfectly fine, 
they are free to do as they wish with their own time, but as I stated 
earlier, it's irresponsible to not delegate. That is what has been 
happening for the last several years. (no form of delegation)
> 
> If a group wants to fork the code, it's GPL, and they can do so without 
> permission from anyone.  The only thing we're guilty of is asking those 
> who would do so to think about the consequences and the real reasons for 
> doing it.  Most of the time the real reason is that so and so 
> contributor wanted to see a particular feature that scratched their 
> itch, but didn't really help anyone else.
This is not one of those cases. This is a gnustep interoperability 
issue, and the projects apathy of patches which make WM compile under 
some other funky non-linux OS' as well as integrate better with GNUstep 
(which you folks /CLAIM/ to have some sort of special relationship with, 
which IMHO as of the last few years is false)

>  The other reason is usually 
> that they're angry their patch was rejected because the code was lousy 
> in general at solving the problem, or as mentioned above, broke other 
> portions of the code. Instead of taking the advice given by Alfredo and 
> Dan on how to improve the patch for submission, the first impulse is to 
> give them the proverbial finger and fork the code.

My patch was not rejected. It was not accepted. This argument completely 
misses the point. My patch was never reviwed by anyone. This is the 
problem. It's not some big huge complicated patch which requires days of 
thoughtful contemplation.

> Finally, as for "disappearing for months on end", Alfredo and Dan don't 
> get paid for this work and they have real jobs.

Of course they do. We all do. That's fine; As I said, I have no right to 
demand that they spend more time on the project, but I do have a right 
to expect that /someone/ might be able to review and apply a trivial 
patch in under four+ months, and I do have a right to expect that the 
core developers to whom we are all beholden act responsibly and delegate 
to others if they know they are too busy to work on this project on a 
regular basis. Nobody made them Gods.

>  To date, nobody else 
> has proven they have the ability to take on large portions of the code 
> and do so in a responsible manner without radically changing the goals 
> of the project.

It's a nice convenient excuse, however it rings pretty hollow.

> If such a person stepped forward and could devote time 
> to it, Alfredo and Dan would be happy to have them onboard.  The other 
> misconception is that if the project doesn't release every few weeks, it 
> must be dead or the developers must be ignoring everyone.  The fact that 
> we waited two years between a release shows the level of commitment by 
> the developers to make sure things worked well.  It also shows that 
> Window Maker is a mature, stable piece of software that doesn't need 
> every little doodad under the sun added to it.  Lots of people like it 
> the way it is.

Some of WM's current Net_WM support is very broken (it's new, so this is 
to be expected, but needs to be resolved, and new releases made which 
address the known NET_WM problems) and nothing is being done about it. 
More WM integration with GNUstep is not a "trivial doodad" to any 
GNUstep user. This is a continuous problem, where any Window Maker team 
member just dismisses relevant patches as "little doodads" which are not 
relevant to them personally, and as such are not worth their time.

Your e-mail has for the most part only served to reinforce the position 
of the Window Maker team that they don't need to adjust their behavior, 
that they've done nothing wrong in irresponsibily failing to delegate, 
and that patches they personally don't find important are "little 
doodads" which are unworthy of their time (even just to review them and 
provide feedback).

It's just all really disappointing, because I like WM and would like to 
help make it better, but basically it seems that this is impossible at 
this point, because of the laundry list of excuses I've heard which it's 
not necessary to re-state.

I really would like to help elicit positive change, but I feel that 
sentiment is not shared by the core WM devs. :( There's not much more I 
can do.

Respectfully,
Alex Perez

_______________________________________________
INFO: https://windowmaker.org/lists/listinfo/wm-user
ARCHIVE: https://windowmaker.org/lists/archive/wm-user/
FAQ: http://windowmaker.org/faq.html