I've read the entirity of the locked thread in question and many more besides. I feel that a lot of the squabbling is based on several things:
* Poor communication - For many in both communities, English is a second language. As a native speaker, I'm thankful the discussion is taking place in "my" language, yet I can easily see how this could make it difficult for folks on both sides to express themselves clearly.
* Personality clashes - After reading so many posts by the key players on both sides, one can't help but form personal impressions of those involved. Several people come to mind as having particularly difficult personalities. To the outsider, it seems unfortunate that such people are so influential in this decision-making process. I think the entire community would be better served if the developers were selected based upon coding skill, but community representatives selected based on their ability to speak articulately and maintain a creative, open-minded approach to problem-solving while keeping a cool head.
* Lack of shared underlying goals - While I've seen a lot of lip service given to doing what best serves the community, I - like several others who've posted their feelings - perceive that many of the key players seem to have a very egocentric perspective. Again and again the sticking point seems ultimately to be a desire for recognition, at a project level as well as an individual level. Both projects want to keep their names. Contributors from both sides worry that suggested paths forward implicitly invalidate their work or disrespect them as people. These concerns are very understandable, yet I feel they are petty and a great detriment to progress.
The most insightful post on the subject of the proposed merge was made by (I believe) mcook. I can't find it right now, but here's what I got from it - He likened compiz to a library such as Cairo and Beryl to (some app) that relies on Cairo for display.
As I understand it, the compiz project is primarily focused on compiz-core. There is much less focus on ease of installation and configuration, and on user-space code in general. As important as the core is, this makes total sense to me. I think the core needs a dedicated development team. Working on the rest as well would be a distraction and would seem like too big a task.
Beryl seems like the opposite, focusing on friendly user-space apps, easy installation, and creative applications of Compiz's power. Where Compiz focuses on development, Beryl focuses on implementation. This is extremely valuable, too, as it brings the composite / 3D desktop experience to newer or more casual Linux users. It also serves to popularize desktop Linux by making it more productive, prettier, and more fun.
I'm glad Beryl is readopting compiz-core. This will clearly eliminate a lot of duplication of effort and allow a larger user base to debug the code; it benefits everyone. I'm on the fence as to whether the much-debated fork was necessary, however it was handled. I think it's a moot question now and one best left alone. There is no benefit to looking backward from here.
I think it would make sense to "merge" the plugins side of things, as well. There has been much porting of plugins between Compiz and Beryl, which again wastes time. It's been suggested several times that Beryl is a distribution of Compiz, and especially now that Beryl is once again utilizing compiz-core, I think that is accurate. Some in the Beryl community take issue with that, feeling that that diminishes their contributions somehow. I disagree with that sentiment. Ubuntu is "only" a distribution of GNU/Linux, yet Ubuntu as an organization and as a community is terribly important. The Ubuntu team has made desktop Linux viable in a way and to a magnitude never seen before, and at just the right time. I see Beryl as doing the same thing with Compiz, and by extension, furthering the spirit of Ubuntu by making geek toys available and useful to the general populace.
For what it's worth, I think both the Compiz name and the Beryl name are valid and important. The creators of Compiz have made a name for themselves and garnered well-deserved respect for their work. That they should name their own project and maintain their own identity is without question. Yet the contributions of the Beryl team have been very significant as well, and as Youtube attests, the Beryl name has tremendous recognition, respect and value. I think it, too, should be kept.
I feel the Compiz devs should continue to develop compiz-core aggressively, maintaining both their high coding standards and their name. I don't see value in their continuing to maintain and include plugins. To me, they are similar to the widget toolkits packaged with X / Xorg. Theoretically, they turn the support library / core app into something useful in and of itself. In reality, they are seldom if ever used because "third-party" addons are so much more sophisticated and well-developed. I imagine using compiz-core like I use Xorg and using Beryl plugins like I use GTK. I need, love, and respect both projects. They each provide useful, wonderful things in their own respective areas.
Finally, I'd like to point one thing out - I've seen it argued over and over that the vast majority of Beryl is Compiz, and that is supposed to diminish my perception of the value of the Beryl project or of the contributions of Beryl devs. All that does for me is diminish my respect for the poster of that opinion. Compiz, like Beryl and every other project on Linux, relies heavily on hundreds or thousands of other packages. That's how it's supposed to work. Unless you are Linus Torvalds or Richard Stallman, you're standing on the shoulders of giants. There's no shame in that; it's fabulous. It's one of the things that makes Linux and open source great. There's no reason to reinvent the wheel, or the chassis, or even the interior. You can design and implement a badass steering wheel and make everyone's life that much better.
iznogood wrote:Hi maniac,
i only want to add that i like the way you handle the entire thing, it clearly shows that you mean well. I also never doubted that from the start, for you and almost all beryl devs.
It is pretty obvious that most compiz people are extremely annoyed from certain comments made by a few beryl people, mostly users...This triggers immediate responses on the thread (i have done it a few times) and things go on in circles...
I do not want to justify anything but if you read some threads you will see that most responses from compiz where directly aimed at certain comments from some people(i know from mine, Mikedee's and RYX). Check the 24 page thread if you don't believe it.
What i propose is that you should post an announcement on Beryl forums describing your plan in detail, including how things stand now and also how you want to go on...
This will help people from posting stuff here that hurts the process...
I want to see the merge move on but at the same time i would not like to have people around with the wrong impression on things. I can not do anything about that but you can.
Personally i say that i do not want to post on that matter anymore, at least in a provokative manner, but at the same time i can not listen to people saying things that are not true. Words can have great power and influence on people, just to remind you that if certain people have not cried loud about how David is not paying attention to them and have stopped to press the matter (sometimes with truths, sometimes with lies) and instead been more calm, we would not have this problem now.
Users browsing this forum: No registered users and 1 guest