What is actually in the way of the merge??

[RYX]

This is a question delfick brought up in the other thread. I think it is worth to have a separate topic for this.

Let's sum up everybody's reasons why he/she thinks we should or we shouldn't re-unite the communities.

RULES FOR THIS TOPIC:
(Why has this topic its own rules? To ensure that everyone can speak freely here, without being interrupted, blamed, criticized, corrected or insulted by others. There is no discussion wanted here, instead we want your honest personal opinion.)

- No discussion. (quoting/disagreeing is considered discussion, too :))
- No commenting on other people's comments (agreeing is allowed, if it contains no "but" ... :)).
- No offense.
- No "beryl/compiz is better".
- No complaining.
- Duplicate answers ALLOWED and wanted.

Everyone who violates these rules from now on will be put on the "black-sheep-list".

Black sheeps:
- ...


:)

[delfick]

thnx ryx :D

Reasons for merge (in my eyes)
*because it means no more wasted effort
*better window manager at the end of the day
*no more community vs community hatred
*a chance to put all this behind us, learn from the mistakes and work together, sorting out differences rather than running away from them

[wfarr]

I think what's interfering with the merger is the pride among several Beryl members.

They ask for things like name changes and such when they seem to forget they're the people merging into Compiz, not the other way around.

I understand they don't want to lose face or act like they "lost", but this constant bickering because they don't want to suffer a little bruised pride is only making this merger far more difficult.

Not to mention, a lot of unnecessary bureaucracy is being added in layers upon layers, which is only going to result in us shooting ourselves in the proverbial foot.

[jackkerouac]

I think the biggest hurdle to implementation so far is the egos involved. The Compiz devs seem to have their hackles up and it's interfering with the merge. People seem to want to frame this merge as someone 'winning' and someone 'losing' which is at best, counter-productive.

From what I've seen (and admittedly, I haven't seen it all), the Compiz devs are still po'd over the fork and seem to see the merge as an opportunity to 'get back some of their own' rather than a precious opportunity to work for the users.

On the other hand, reasons for a merge:

1. Duplication of effort stopped.
2. Stable code base.
3. Beryl can bring proper installation procedures, plugins, settings manager, emerald-like interface and a sane repository structure.
4. Compiz can bring rock-solid code, great community and a cool, new icon. :)

EDITED: to frame my comments more as a statement than a response. Sorry about that; my bad entirely.

[wfarr]

- No commenting on other people's comments.

Ahem.

[Alejandro Nova]

As I see the situation.

Reasons for the merge: Duplication of effort stopped, stable code base, and, for the first time, a good structuring of what to do and how to do it.

Reasons against the merge: "I don't want to lose, I want to win, I want to impose you MY NAME and you'll have to clean my boots with your tongue, 'cos I rock and you suck, HAHAHA" (both sides).

My stance: Certain sacrifices will have to be made, by both sides. I think changing the name of Compiz-extra is a good thing, not only because I don't like Compiz-extra as a name, but because, in the original plan revealed to us, Compiz-base will become a base package, and parts of it would be even integrated into X core (yes, MIT license HELPS here). So, in X11R7.4, we may see how Compiz disappears, fagocited by X, and Compiz-extra will be without its mother package, Compiz-core. And the Compiz-extra name won't have any reason to be anymore.

Coral and Blitz are both good names. Leadership? Who codes, rules. And yes, I'm sticking with some original names for specific things (notably Emerald... trying to change its name to cgwd is sad... say "emerald" and say "cgwd", which one is easier to remember?)

[Tommy]

This is my first post to these forums, but it's on a subject that clearly matters greatly.

I've never used Compiz as it now stands. I've been using Beryl for the last few months, and the apparent desire to experiment and surprise has kept me using it. Every update brings in new surprises, and this is good. Of course, there are also inherent stability issues, which is bad. This is obviously a result of how Beryl is developed, and while I personally quite enjoy finding bugs and poking around with it, it's not what's needed for a genuine end-user production. Compiz, on the other hand, appears to be based entirely around a work ethic that ensures stability, but with less user input. This is also good for what it aims to achieve. Each has its own caveats, and this is to be expected for two projects that have been developed in very different directions since the fork.

After finally bothering to read and join the forums for both projects, I have to say I'm deeply alarmed by some of the attitudes amongst both users and developers on both sides. Here's the situation as I (a user who is interested in development) sees it:

I'm using software that I chose to try and found fun to use. However, after coming to these forums, it's clear that in the eyes of certain Compiz developers and users, this makes me an idiot. I know nothing of the fork. I wasn't using any form of Compiz when the political disputes kicked off. All I know is software which I've used and enjoyed is being described as "pointless" and apparently has "no reason to exist". After reading several threads on these forums, and several more links to various messages and mailing lists, it's clear where this animosity has come from; however, it is far from productive. I want to be involved in the development of whatever emerges from this fire, because I enjoy having a fully compositing window manager and everything that the concept could become. I don't enjoy being in the middle of a war because of it.

Let me say this with as much clarity as I can muster: there is no winning or losing. I don't care if I know next to nothing about the history; this is not war, this is free software development. Anyone (Beryl dev, Compiz dev or otherwise) who tries to suggest that the idea of a merger has anything to do with winning or losing needs to seriously examine why they are developing software. I was never aware that open-source development was about winning anything. This goes out to Beryl devs who complain that Compiz doesn't want to change the compiz-extra name because of misplaced pride, and to Compiz devs who complain that Beryl wants to merge under a whole new name because they don't want to feel like they lost the fight. This goes out to users, moderators, anyone who feels pride is more important than creating the best window manager the world has seen. The goals are the same: create a flexible, stable and configurable compositing window manager that can do as much or as little as the end user wants it to. If everyone works towards that, nobody can lose.

In the end, names mean nothing. You can have the best name in the world, but if the product sucks, that's all you've got. Any argument justifying a particular name is completely missing what really matters: the combination of the powerful core of Compiz and the pluggable flexibility of Beryl into a mature and world-beating application, and the galvanization of two different but active groups of people into a productive community of developers and users. A very short time from now, nobody is going to care about the name. After all, Windows is possibly the most uninspired and boring name ever devised, but look at the power it wields. It's that power that Compiz and Beryl have to join together to bring down, regardless of their roots. If, a year from now, the idea of an open-source compositing window manager is dead because of such bitterness, then there are a lot of people who are going to be sorely disappointed, and an incredible opportunity is going to have been missed. Don't let it happen.

[gnumdk]

Deleted message (by me)

[Tommy]

It doesn't matter to most of the Beryl devs, no. I didn't say it did; rather I was simply saying that the message was aimed at those few that it does matter to.

And the exact naming scheme depends on how Compiz is structured in the future. That debate could go on forever, but it's not what matters.

[jsage]

Background: I've used compiz, compiz-quinn and beryl. I'm currently running compiz.

Reasons for the merge:

Stable core to build on, consistent roadmap for core improvements
Robust beryl plugin developer & user community to revitalize compiz
Wider variety of plugins without the duplicated porting effort



Obstacles to the merge:

Some of the developers. Every user who's posted has said "let's do it". So let's do it.

To the beryl devs: Ignore the insults and bad-mouthing - they're just sour grapes. Ignore the ones who insist on peeing in the punch bowl. Cooperate with David. Make your plugins compatible and well-behaved. Be grateful for the stable compiz core.

To the compiz devs: This is not about the beryl community "coming back home". This is about the compiz community uniting with the much larger and stronger beryl community to carry the platform forward. Bring your fast and stable core and be grateful for the strength of the beryl plugin community.

To the devs who've been acting as adults: Thank you. Hopefully the rest of the devs will learn from your mature example.


thanks for asking my opinion.

[stalynx]

I think you are wrong here... Compiz-extra have to stay compiz-extra because it's a bundle of plugin for compiz...

Just look at xmms, you have xmms-alsa, xmms-jess, ...

We have 3 solutions:
- Give user one archive per plugin (like most other softs)
- Give user one tarball with all community plugins.
- Do both.

I prefer the last solution but i think it have to stay compiz-anything... We can discuss on the "extra" part of the name.
I agree users should have the right to pick which plugins they want to run without installing extra packages. This makes the most sense to me. If however there are people who want an "entire" package then they should use compiz-extra or some other package. If the Beryl devs still want to work some specific DE-like package then they should be allowed to do so. Now that we have a common core all these options are open and they should continue to stay open.

[roico]

they seem to forget they're the people merging into Compiz, not the other way around.

no.
it's neither compiz is merged into beryl nor beryl is merged into compiz.
compiz and beryl are merged together.
actually, this is exactly the point of disagreement between beryl devs and compiz devs, the name issue is just a disguise for this issue.

[iznogood]

they seem to forget they're the people merging into Compiz, not the other way around.

no.
it's neither compiz is merged into beryl nor beryl is merged into compiz.
compiz and beryl are merged together.
actually, this is exactly the point of disagreement between beryl devs and compiz devs, the name issue is just a disguise for this issue.

For where do you came to such a conclusion ?? Put a llink ...

[imnotpc]

OK, everyone needs to settle down. It's time to end all this finger pointing and name calling. The bottom line is that we need to start getting used to working together as one community. The entire Linux community is watching us and it's not a pretty picture. Either we start discussing ways to work together, or this thread will end up locked too. Enough is enough!

[RYX]

Gnumdk, iznogood, roico - please read the rules in the first post and edit your posts according to the rules. Otherwise I have to delete them.

This thread is not for discussing those things again, it is for everyone to express his own feelings without hurting somebody else's ...

Thanks :)

[iznogood]

I think you are wrong here... Compiz-extra have to stay compiz-extra because it's a bundle of plugin for compiz...

Just look at xmms, you have xmms-alsa, xmms-jess, ...

We have 3 solutions:
- Give user one archive per plugin (like most other softs)
- Give user one tarball with all community plugins.
- Do both.

I prefer the last solution but i think it have to stay compiz-anything... We can discuss on the "extra" part of the name.
I agree users should have the right to pick which plugins they want to run without installing extra packages. This makes the most sense to me. If however there are people who want an "entire" package then they should use compiz-extra or some other package. If the Beryl devs still want to work some specific DE-like package then they should be allowed to do so. Now that we have a common core all these options are open and they should continue to stay open.

I fully support that solution, its the most flexible and allows everyone to do what he wants, users, developers and packagers. Also its distro independent and reduces pressure on the developers to deliver versions on specific dates

Also it leaves the name - managing issue out of the loop alltogether which actually is the best thing to do, at least for now

Feel free to delete my previous post...

[jackkerouac]

I like the idea of a compiz-beryl package or a compiz-coral package or a compiz-specialeffectcs package, etc., etc.

[Spillaz]

thnx ryx :D

Reasons for merge (in my eyes)
*because it means no more wasted effort
*better window manager at the end of the day
*no more community vs community hatred
*a chance to put all this behind us, learn from the mistakes and work together, sorting out differences rather than running away from them

*agrees with delfick*

I also think that its better for us users not to have to keep on choosing which one is better. Yes, I know competition is good for a while but when two projects are SO similair, the only path is to A) Make it extremely different or B) Merge the two projects.


I really think that there has been a lot of misunderstanding with what exactely a merge will be.

Firstly, we DONT have to make a new name for the communities. If it is desired, we can call it the Beryl/Compiz community until we might want to do something else.

Secondly, we will probably have to pick a new name for the project "Compiz extra"
AFAIK, David doesnt want unofficiall stuff under the name Compziz. (Correct me if I'm wrong

Thirdly, we really need to stop saying each others code is bad. Respect everything.

Fourthly, the merge is not going to be a WIN or LOSE situation. It is not one community taking over the other. When we say merge the websites, we mean that we either A) Make a new one or B) Merge into yours under the name "Beryl-Compiz community."

So what will the benefits be?

For Compiz-
You get more plugins, some of which are really interesting. Beryl Settings Manager, which IMO is better than editing gconf keys and Emerald.
You also get a bucket load of users and that means users>youtube videos>hype>potentially more developers

For Beryl
You get a MUCH MORE STABE (and faster) core. Perfomance increases by around 20-40%. You can also be included in a number of distributions be default.

For both communities-
You can finnaly not have to waste time porting stuff around!

For users-
No need to decide which one is better at the loss of xyz

There, I think that clears things up. If you dont understand be (and I dont blame you) then just ask me what I mean by something and I will explain in more detail.

Also, If you see a flame war brewing, try not to contribute and refer back to this post. I think its good to have definitions ;P

[lowfi]

people, whats with all the thread locking? Is that really the best way to go? (and the rules about what one's allowed to speak about). As Noam Chomsky said "There's really nothing to argue about freedom of speech. Some are in favor of it and some are not." Maybe you think about that too.

[imnotpc]

Good point lowfi. Everyone needed a chance to cool off but there's no reason to keep them locked.

[Kevin]

I think in order to get this merge under way, we need to start by doing a couple things.

First, the core merge is already under way. Its called compiz, either way Beryl will be using it and improving it. There's no more discussion needed here.

Second, the Beryl community and devs need to outline exactly what their goals with the Beryl Project are. Is it to provide a compositing manager, or is it to provide a total solution complete with fun plugins, settings managers, things like docks etc.

Third, the Compiz community and devs need to to the same. When this discussion started, the compiz-extra "project" idea was still new. It had not been discussed what exactly we wanted it to be, and there seemed to be much confusion over whether it was simply community plugins, or more.

Now we need to unite the communities, but the issue of what beryl and compiz-extra "want" to be is irrelevant. The community will be dealing with extensions of compiz. Right now we have this Beryl Project, and we have the beginnings of this compiz-extra project, but in the future there could be more. This is why we should unite the communities under the name of the "Compiz Community". This community forum can have subforums, administered by their own leaders, for these projects. The important thing though, is that anything developed within these projects is under the same community, and anything developed within them can be made available outside of them.

That's my opinion anyways.

[delfick]

The important thing though, is that anything developed within these projects is under the same community, and anything developed within them can be made available outside of them.

That's my opinion anyways.

agreed......as long as the two community and projects become one, everything else is irrelevant. (especially names)

[Kevin]

I think also, there is a lot of trepidation within the compiz community towards the merge. I don't think this is because we don't want a merge, but because we like the community we've created here.

Personally, I like the smaller community here, with less traffic. Most of our community here knows the history of the project, and at least some technical things about compiz. The beryl community is very large, with many users who are uninformed about linux and compiz in general. This isn't a bad thing, its just that its pleasant for people who know what they are talking about to hang out with each other. I think we have a lot of that here, and its hard to give that up in order to have a large community with lots happening.

That's something we're going to have to get used to if this merge is going to happen.

[Kevin]

To elaborate on my first post in this thread, I thought I'd outline how I'd like to see this community merge.

The Compiz Community would be at forum.compiz.org. This community would be a place to develop plugins, settings managers, and anything else that developers would like to start here.

This community would have a subforum for the "Beryl Project" or whatever they want if they choose a new name. This project would be a package of the plugins they want, including the plugin for copy rendering, and plugins for libbs, and their own settings manager. This would enable them to create a package that anyone could install, and essentially have what beryl is now, but at the same time allow users to get pieces of the beryl project that they want, without having the entire thing.

For example, I like the plugins the beryl community has created, however, I think it is silly to pointlessly get rid of gconf, and I don't think that copy rendering should be used, I think drivers should be improved. With this setup, I can use the plugins the beryl project creates, without having to give up gconf etc.

[RYX]

I think we already have some important points here ... Is'nt it nice to see that everyone can freely express his thoughts without being attacked by someone else? If anyone doesn't feel like this topic is a good thing, I understand that. Please, just don't comment here to tell that you dislike this thread.

@lowfi: I absolutely dislike locking down threads and I didn't do it myself. I tried to create this thread under the rules defined above to ensure that everyone expresses what he/she thinks - not people start attacking others for what they think. Discussion is great but sometimes it is just better if everyone can talk the anger off his soul ... (sorry - bad translation) - without being interrupted by others with a "better" opinion.
If you closely read the rules you'll see that I absolutely not say what anyone is "allowed" to talk about ... instead the rules only say what should _not_ be done ... I guess we need some "kindergarten-tactics" here since many people acted like little children in the other threads ... (yeah, maybe including myself :))

:)

[maniac]

For example, I like the plugins the beryl community has created, however, I think it is silly to pointlessly get rid of gconf, and I don't think that copy rendering should be used, I think drivers should be improved. With this setup, I can use the plugins the beryl project creates, without having to give up gconf etc.
Well, IMO it's better to have less packages (not only one, but lets say 5-10) instead of having one package per plugin (which would mean 40+ packages if one wants to install all compiz stuff).
You don't need to actually use all installed stuff ;) If you dislike e.g. copy rendering, just leave it disabled. Same applies to libbs vs. gconf. But all plugins are equal in nature, that's why I think it's a bad thing to have separated "Plugin discussion" and "Beryl discussion" sections.

[jackkerouac]

Personally, I like the smaller community here, with less traffic. Most of our community here knows the history of the project, and at least some technical things about compiz. The beryl community is very large, with many users who are uninformed about linux and compiz in general. This isn't a bad thing, its just that its pleasant for people who know what they are talking about to hang out with each other. I think we have a lot of that here, and its hard to give that up in order to have a large community with lots happening.

This is a very good point. Sometimes I forget that this merge is basically going to drop literally hundreds of users into the 'Compiz lap' so to speak, with their attendant questions, problems, etc.

For example, I like the plugins the beryl community has created, however, I think it is silly to pointlessly get rid of gconf, and I don't think that copy rendering should be used, I think drivers should be improved. With this setup, I can use the plugins the beryl project creates, without having to give up gconf etc.

Another good point. While I hate gconf, I am all for some masochist willing to use it themselves ;) Choice is the name of the game.

I guess we need some "kindergarten-tactics" here since many people acted like little children in the other threads ... (yeah, maybe including myself )

Yeah, me too and for that I apologize.

Well, IMO it's better to have less packages (not only one, but lets say 5-10) instead of having one package per plugin (which would mean 40+ packages if one wants to install all compiz stuff).

As long as there is a 'full smash' package, i.e. a package that contains ALL the plugins, so that someone who wants the 'full experience' can get it easily, that would be great.

[delfick]

i think the domain of the forum should be decided purely on which domain is faster and crashes less (including domains other than compiz and beryl already have)

[Spillaz]

i think the domain of the forum should be decided purely on which domain is faster and crashes less (including domains other than compiz and beryl already have)

*agrees and thinks that it should not be the beryl server because of the stupid "Too many connections thing" that keeps on popping up

[Spillaz]

i think the domain of the forum should be decided purely on which domain is faster and crashes less (including domains other than compiz and beryl already have)

*agrees and thinks that it should not be the beryl server because of the stupid "Too many connections thing" that keeps on popping up

[Spillaz]

oops, double posted.

Stupid windows

[nesnomis]

From a users perspektive there is absolutely nothing i the way of a merge!!! As i see it there are only benefits! .. Greater community, no fighting and the nicest desktop any os have!!! ...

My humble opinion about names .. hmm... Beryl seems to be rejoining more than merging so offcourse Compiz is the name... The merge problem seems more like the community trying to get along (as i understand developers are allready agreeing!?)... so dump compiz-extras and call it compiz-beryl... :) ... (i believe someone else said it here somewhere?!)...

In the end Compiz is some great work and the Beryl community have contributed with some great innovative plugins... :) ... So.... the 'merge' is a great thing making everything even better... Especially for us users!... :D

[jackkerouac]

Now, I do NOT want to start a flame war


:) :)

[mikedee]

This was a private mail and not supposed to go to the compiz list.
Please don't don't read it unless you're on the compiz admin list. :)

Lets keep it that way shall we, he is sorry and he is optimistic for a merge. Lets leave it at that. I am not censoring, the post is available on the compiz dev mailing list for anyone to see.

We all know what post we are talking about.

Edit: More clarification in case people are upset. The post was an accident, I personally think its good that everyone is honest. If you want to reply to 'that post' please do, nobody will be offended.

[jackkerouac]

I am not censoring, the post is available on the compiz dev mailing list for anyone to see.

Actually, yes you are, but upon further reflection, I agree with the decision.

[iXce]

i think the domain of the forum should be decided purely on which domain is faster and crashes less (including domains other than compiz and beryl already have)

*agrees and thinks that it should not be the beryl server because of the stupid "Too many connections thing" that keeps on popping up
That's not berylbox ;) Forums & trac are hosted separately, berylbox runs everything else smoothly.

[Suzuran]

I think at this point there's really nothing in the way of a merge, but egos on both sides could get in the way of a smooth merge, as could the drive to constantly rehash the past and attack any sentence or part thereof that could be interpreted as a disagreement with one's own views. Also discussion of things that will only be relevant once the merge takes place (such as the names of the community and what is now called Compiz-Extra) seems to get out of hand sometimes. My personal view is that the community should have the same name as the plugins package; it should be something catchier than Compiz-Extra but probably not Beryl (even though I do like that name). However, that decision is best left for after the merge is complete code- and community-wise.

The merge will hopefully result in a better experience for all and no more duplicated efforts or needless fighting.

[scooper86]

this is indeed a good thing as most people realise i have been a long time beryl/quinnstorm user instead of the regular compiz purely because of the more wider range of plugins and a good gui but i still believe the obvious strength and stability comes mainly from the compiz developers. Everyone from both beryl and compiz do contribute an equal amount even if it isnt on the same things. I hope a name change does occur as it will put an end to some people squabbling and bring a fresh light to the community.

Anyway the main things i hope to come from this is just more integration, we already have a LOT more effects than vista and OSX combined all we need now is integration of the desktop and all its apps, i know this isnt purely compiz/beryls undertaking but it does need to stem from somewhere and a lot of app makers are still on the fence on what to go for when concentrating on a compositing manager.

I hope i havent tread on anyones toes this is just my opinion and not an opinion of the greater community just like everyone elses opinions do not represent the wider community the great thing about open source is that its governed by the community even the developers have no control over who uses what and in the end if its a great app then people will use it.

[delfick]

^^ speaking of which, i can't wait till apps can have true transparency backgrounds :D

[wfarr]

^^ speaking of which, i can't wait till apps can have true transparency backgrounds :D

They already can.


Gnome-Terminal 2.18.0's transparency feature is now smart: if a composite manager is running the background is now actually transparent rather than "transparent-to-your-desktop-background".

See this screenshot: http://img142.imageshack.us/img142/8993/gnometerminalec7.png

[delfick]

Gnome-Terminal 2.18.0's transparency feature is now smart: if a composite manager is running the background is now actually transparent rather than "transparent-to-your-desktop-background".

it's been like that for ages :P :D

i'm talking making this a feature of a gtk theme

like what RYX was trying to do a while back :D

[apoclypse]

Gnome-Terminal 2.18.0's transparency feature is now smart: if a composite manager is running the background is now actually transparent rather than "transparent-to-your-desktop-background".

it's been like that for ages :P :D

i'm talking making this a feature of a gtk theme

like what RYX was trying to do a while back :D

Thats exactly what I was thinking. I was kind of hoping I could find it online somewhere to see how far he got. Ah well.

[malwk-]

As an end user I have used compiz, then compiz-quinn, then again compiz, then beryl and now compiz on AIGLX.

I have a few things to say before summarizing my feelings and toughts about the merge:
- as a pure window manager both compiz and beryl lack things I liked pretty much in metacity, for example (and mostly) the intuitive focus stealing prevention, it was really good working, I tried the FSP in beryl - it sucks, most of the time it was doing annoying stuff.
- as animations and bling beryl took the leadership, now it has that many animation types it takes half an hour to test them all, still few of them are practical/usefull. Exception taken on the water + dbus end used for gaim new message signalization.
- as a window decorator beryl comunity developed some different engines most practical and used (here I just guess) being the pixmaps engine, some really interesting stuff as themes are shown, but unfortuantelly still no new exciting and suitable for a default slick funcky and aorable theme.
- as stability and operation it is hard to tell which one is doing the right for me, I face difficulties with both, beyl locking the screen too often on my laptop (i855GM with AIGLX) so I have to cold reboot (I cannot even switch to a VT) and compiz doing stuff to the Xvideo overlay, preventing normal playback of video (then the players use X11 wich is VERY SLOW).
- compiz loaded with the exactly same plugins and enabled effects use 30 to 40 less memory - this is a big plus for me as I dont have 2GB beast.
- beryl having the cute GTK configuration GUI might be more tempting to the average user but I really dislike it, there are TOO MANY options really just to setup really simple effects. I am not saying that using gconf is the best but... I will strongly disaglree that presenting the user with 10000 options is good as well.

What I expect from the merge?
Honestly - the wors;
I expect compiz to slow down its performance (On AIGLX with i855GM + beryl when I move a window i I see the window movement running atfer the pointer (for example the Gnome terminal which is translucent you know:) ) and on the exatcly same laptop with exactly same plugins and features enabled i can move in real time the same window ... this is pretty much enought for me to tell that compiz is doing better for me for now )
I expect compiz to inherit the bugs in beryl (I kow of few but it still annoyed me, for example the hotsopts stop working I dont know why and how but in a moment I find myself with hotspots (corners) not working, I think this is happening with compiz too sometimes so I dont know who from who inherited this bug but it is really really annoying as I have to restart the window manager to get it back to normal.)
I expect the plugin front to get even wors because of the entusiastic plugin writers in beryl comunity - everyday a new fresh idea, developed into code for few hours and that is it. No suport, no refinement afterwards (want example? I've been begging for extended DnD support for scale and show desktop for months, never got them. I now have it in compiz, so thanks but no bling bling when no functionality!)

Okay, it is not that bad.
I hope that some sane plugins will be developed, animated dacing, bouncing, collapsing and so on windows are good but... c'mon, there HAVE to be something more usefull we can do in order to make the desktop more user attractive and user friendly.
I hope the plugins will be able to compile separately and be installed per user (in ~/.compiz/plugins for example:) ) I know this is possible now with compiz, not sure about beryl... Why? because I can patch my own versions of the plugins - usually just one file and test it locally ( put in $HOME/.compiz/plugins) and it is loaded before the system wide file thus if I broke something it wont affect my mom's loggin later and work as usual.
I hope the plugins will have sane defaults. and the list of active plugins will also have sane default:)
I hope a new configuration panel will be developed and as per every config panel for a complex app it will have the tree structor for the options (gcnof-editor style) and not the crazy panel of beryl-settings as it is now, excuse me but buttons on top buttons on left, tabs on right and hidden elements? What am I - a detective?

And I would reall be happy if the new core becomes a really stable and sane development environment for all the plugin developers with a good and stable api so to make possible all types of plugin, sane and insane:)

At last I hope that the comunities will burry the hatchet and will work together. as well as the core developers. I hope if I now stop using this forum and come in a few months in the new forum I will not be able to tell that back in time there were two of them.

I also would like to see better forum structure, one that i really will like is a separate places for questions about each plugin or at least separate places for plugin development, questions about plugins and plugin d/effects reports:)

This is it. Wow it got longer than I expected, sorry:)

[mikedee]

- as a pure window manager both compiz and beryl lack things I liked pretty much in metacity, for example (and mostly) the intuitive focus stealing prevention, it was really good working, I tried the FSP in beryl - it sucks, most of the time it was doing annoying stuff.

Compiz has focus stealing now, I think its the same as in metacity so you should not have any problems now with FSP. Adding things like extra placement modes to place and the window matching means that we are starting to get the features of 'real' window managers.

- as animations and bling beryl took the leadership, now it has that many animation types it takes half an hour to test them all, still few of them are practical/usefull. Exception taken on the water + dbus end used for gaim new message signalization.

Thats really cornelius' fault ;) - I know he plans to split up the animation plugin at some point.

- as a window decorator beryl comunity developed some different engines most practical and used (here I just guess) being the pixmaps engine, some really interesting stuff as themes are shown, but unfortuantelly still no new exciting and suitable for a default slick funcky and aorable theme.

Most people think the same, people are thinking about a much more extensible decorator so hopefully new things will come of it. I do not think there is much more that can be done with emerald (I dont know, its just what I hear)

- as stability and operation it is hard to tell which one is doing the right for me, I face difficulties with both, beyl locking the screen too often on my laptop (i855GM with AIGLX) so I have to cold reboot (I cannot even switch to a VT) and compiz doing stuff to the Xvideo overlay, preventing normal playback of video (then the players use X11 wich is VERY SLOW).

There is no chance of the slowness creeping into Compiz, David still keeps a tight reign on the core and people cannot just commit anything they like. He recently wrote the video plugin which should cure your video problems (when support is in the server and apps)

- compiz loaded with the exactly same plugins and enabled effects use 30 to 40 less memory - this is a big plus for me as I dont have 2GB beast.

I do :D - but I also agree that a computer should be used for your apps not just a window manager (which is a tiny part of a large machine).

- beryl having the cute GTK configuration GUI might be more tempting to the average user but I really dislike it, there are TOO MANY options really just to setup really simple effects. I am not saying that using gconf is the best but... I will strongly disaglree that presenting the user with 10000 options is good as well.

I am personally trying to make it easier for anyone to write whatever settings tools they like, the dbus plugin and the http server I just released should help this. There are a few minimilistic tools around gnome-compiz-manager looked nice to me last time I looked.

Most of the plugin bug reports / development are in the plugin development section (each plugin has a thread). We have just created a wiki template page which plugin writers can use as a home page for their plugin.

[malwk-]

Okay I KNOW this isnt the right place to do this and I am really really sorry but... what plugin again?!?!? OMG I am so excited, where should I go and read about this? Please let me know! God this is the last barrier for me to absolutely love my desktop (of course I would like to see the blug plugin advantage in motion with intel but i doubt this is possible) Thanks...

There is no chance of the slowness creeping into Compiz, David still keeps a tight reign on the core and people cannot just commit anything they like. He recently wrote the video plugin which should cure your video problems (when support is in the server and apps)

[mikedee]

Okay I KNOW this isnt the right place to do this and I am really really sorry but... what plugin again?!?!? OMG I am so excited, where should I go and read about this? Please let me know! God this is the last barrier for me to absolutely love my desktop (of course I would like to see the blug plugin advantage in motion with intel but i doubt this is possible) Thanks...

There is no chance of the slowness creeping into Compiz, David still keeps a tight reign on the core and people cannot just commit anything they like. He recently wrote the video plugin which should cure your video problems (when support is in the server and apps)


I wouldn't get too excited just yet, but there is light at the end of the tunnel :)

http://lists.freedesktop.org/archives/compiz/2007-March/001576.html

I do not know what is happening for the patched apps, but the video plugin is included in 0.5.0 (although it will not do anything without the patches for mplayer and xorg)

More info here

http://www.google.com/search?q=video&sitesearch=lists.freedesktop.org%2Farchives%2Fcomp iz%2F