“Beryl 0.2.0 is a complete overhaul of Beryl. The last stable release (0.1) featured a very fun, and eye-candy based compositing window manager. However, since it’s release, many parts of Beryl have been rewritten, replaced, or simply dropped. It’s filled with eye-candy, better user support, new features, and stability.”
They just took the latest code from compiz to get more stability? Although I believe they did the extra eye-candy and features themselves.
wow, why don’t you write thieving scum.
I’m pretty certain that there is enough changes in beryl to actually call it a fork. They do seem to have all these plugins that compiz doesn’t have, they probably uncommented a few lines in compiz.
Seriously, the different goals and drive between these two projects are interesting, as well as their chosen licenses, but why would anyone want to sully their achievements.
I personally want thumbnails on my taskbar…when compiz does that start saying its stolen. This is open-source *under* a license to do just that. Thats the point of the license.
I may be wrong, and the could be labled as FUD or some other such term, however this is just an impression that I got (at least how the Compiz folk feel).
It would appear, that at least some of the Compiz folk feel that.
A) Beryl has chosen a licence that pretty much prevents Compiz from benefitting from Beryls development, however allows Beryl to continue obtaining code from Compiz.
B) Beryl developers “regularly” take code from compiz, and include it in Beryl with out any actual acknowledge ment to the compiz folk that this is being done.
I personally can’t vouch for any of the above, it’s just an impression that I’ve got from numerous posts.
If the above is the case, on either count, then I personally would have little respect for the Beryl developers.
One of the points of Open Source, is to share, and the strengths gained from people gaining from each others work. It is not intended to allow one group to gain solely off of the back of another group.
The reality is very different
http://lists.freedesktop.org/archives/compiz/2007-February/001413.h…
explains the beryl situation
How can you claim that the reality is very different when the exact same post you link contradicts what you say?
This is incredible, to put it mildly.
“How can you claim that the reality is very different when the exact same post you link contradicts what you say?
This is incredible, to put it mildly.”
Whats incredible is you must be reading a different article. The fact is I read the post *NOT* your interpretation. I’m happy for you to pluck quotes from that that make your points.
You want quotes?
“Quinn, obviously don’t share these opinions and using a core that is MIT licensed seem impossible to him. I’m a bit skeptical to to this though as they haven’t made any significant improvements to the core since they forked it (at least nothing near what’s been done in compiz since then) and they are OK with using MIT licensed X server, X libraries, loading proprietary binary blobs that some call illegal into the kernel and putting themselves in a situation where they feed on a project which they can’t easily contribute back to.“
No thats not the same is it. That actually shows two sides. That shows a disagreement. It actually says that that compiz has caught up “at least nothing near what’s been done in compiz since then”, and nowhere does it say 95%
This I expect, from the very different Development approaches. Compiz goes for “Correct”; Beryl “quick and Dirty”.
“They are OK with using MIT licensed X server, X libraries, loading proprietary binary blobs that some call illegal into the kernel and putting themselves in a situation where they feed on a project which they can’t easily contribute back to”
is not the same as.
“Should I not be allowed to point out that changing the lisence so that the project you rely upon to do all the hard work can’t incorporate your changes while at the same time you can and still do rely on the other projects code is quite disgusting”
What it actually describes is that Beryl team have deliberately chosen the License *not* so they don’t want to give back to compiz but becuase they don’t like the license for *their* code. He says they are hypocrites for this…this is very different from the picture you paint and a different point.
Read the post try and understand it. What you have written is offensive and lies, and if you had any decency you would retract it.
Edited 2007-03-15 20:27
From reading:
http://lists.freedesktop.org/archives/compiz/2007-February/001413.h…..
I get this:
‘compiz core is MIT licensed. I used the MIT license in compiz for two
reasons. When moving to a composited desktop, we’re moving more and more
of the X server (which is MIT licensed) functionality into the
compositing manager and I want to make sure that this new composited
desktop that we’re creating can be deployed anywhere a regular
non-composited X desktop is used today. I also believe that we’re
currently at a stage where a lot of experimenting is being done and it’s
impossible to know exactly where we’re going to end up so using a less
restrictive license leaves more doors open and we’re better prepared for
the unknown. Quinn, obviously don’t share these opinions and using a
core that is MIT licensed seem impossible to him. I’m a bit skeptical to
to this though as they haven’t made any significant improvements to the
core since they forked it (at least nothing near what’s been done in
compiz since then) and they are OK with using MIT licensed X server, X
libraries, loading proprietary binary blobs that some call illegal into
the kernel and putting themselves in a situation where they feed on a
project which they can’t easily contribute back to. But yet again, I’ve
only spoken to Quinn and I doubt that all people involved with beryl
share his opinions.’
This pretty much states exactly the point of view put forward by ‘wibbit’ … How this can be read otherwise is beyond me…
A) Beryl has chosen a licence that pretty much prevents Compiz from benefitting from Beryls development, however allows Beryl to continue obtaining code from Compiz.
B) Beryl developers “regularly” take code from compiz, and include it in Beryl with out any actual acknowledge ment to the compiz folk that this is being done.
That sums it up pretty well.
In regard to A:
The compiz devs have chosen a license that allows for this. Therefore there is nothing immoral or illegal about it. It stays within the spirit of the license.
In regard to B:
Besides that, the Beryl devs are giving the compiz people a lot of credit, even though the cores deviate more and more. It’s even right there on the front page. And it is in the source code etc.
So A is wrong and B is wrong.
Let me first say, i do not want to get in to a slanging match, so hopefully this can stay posative.
In regard to A:
The compiz devs have chosen a license that allows for this. Therefore there is nothing immoral or illegal about
it. It stays within the spirit of the license.
Yes, the compiz folk did choose a particular licence, the licence chosen was one that allowed xorg (I believe) to benefit from the code being written, and to allow, again free movement of idea’s and code between two very closely related projects.
I may be wrong, and this is just a guess (or a feeling), but I think i am safe in saying, that the reason why Compiz chose the licence that they did, was not due to some idealistic belief, other than the desire to be able to intercode with a very closely related project.
I would be curiuos to hear the reason behind choosing the alternative licence for Beryl, and how they felt about the fact that, any thing they do, can not be contributed back. I’m sorry, but for me, leaving aside the letter of the law it certainly does not seem to abide by the Spirit of the law. And I know which one is more important to me.
In regard to B:
Besides that, the Beryl devs are giving the compiz people a lot of credit, even though the cores deviate more and more. It’s even right there on the front page. And it is in the source code etc.
As i said before, what I said, was based on impresions that I got from posts. Yes, they NOW have it on their home page, but in the past when i’ve checked, there was very little indication that it was taking code from Compiz. And I’m sorry, but stating that “it’s in the code” is of little interest to me.
TFA says that Beryl is moving apart from Compiz, but I’m skeptical. My interpretation of the fork has always been that the Beryl folks think that the compositing framework should go in the compositing window manager (i.e. Beryl) wherever possible, whereas the Compiz folks think that this stuff should be pushed down into the X server wherever it makes sense (i.e. for supporting alternative compositing window managers like Beryl).
The Beryl folks can establish a more aggressive roadmap with their development philosophy, but as Compiz moves forward with their roadmap, it will increasingly make sense for Beryl to adopt the compositing framework integrated with Xorg. Therefore one would expect the two projects to converge–rather than diverge–over time, which calls into question the long-term rationale for the fork.
To matters more complicated, Beryl has chosen the GPL, which makes their code incompatible with the Xorg codebase. This makes sense for the window manager and plugin code if the goal is to prevent (and/or monetize) proprietary flavors of Beryl. But the philosophical difference over what goes where means that Beryl code that could have otherwise been useful to other X11 compositing window managers is off-limits unless the project’s code is GPL and they don’t mind continually syncing with Beryl (as opposed to merely depending on Xorg).
While we can disagree as to the merits of the Beryl project and its relationship with Compiz, one thing is clear: before Beryl, many of us knew of seasoned X hackers like David Reveman who work on Compiz, but nobody knew the first thing about this Quinn Storm guy (yes he’s a guy).
Edited 2007-03-15 16:41
“””
but nobody knew the first thing about this Quinn Storm guy (yes he’s a guy).
“””
I thought he was currently female?
Quinn Storm, in person, on Beryl – http://www.lulu.tv/?p=4346
I don’t do Flash. Could you humor me and hand me the Cliff’s Notes version?
I know transgender is involved. I’m just not sure which direction it goes in this case.
At any rate, I hope that he or she is happy.
It takes a lot of courage, and respect for one’s self, to make that decision.
Quinn Storm is a woman. There was an interview with her on the Linux Link Tech Show a few weeks ago.
heshe actually, and a preop one at that. fwiw.
He’s a transsexual. [1]
I believe she prefers to be called a girl.
[1] http://en.wikipedia.org/wiki/Transsexual
“He’s a transsexual. [1]
I believe she prefers to be called a girl. ”
Well, that’s all the relgious nutjob users of Beryl switching back to Compiz then!
EDIT: It just dawned on me that the above posters aren’t joking about Quinn Storm being transsexual. OK. Well, I guess that explains why some people insist Quinn is male and other insist that Quinn is female. For once, we can agree to disagree and both be correct. Or can we? I’m not up on the political correctness implications in this context.
Edited 2007-03-15 21:06
He is a guy … but I would say g(ay)eek?
“He is a guy … but I would say g(ay)eek?”
“OMG that man has long hair! He must be gay.. or transvestite.. or even worse some damn geek! Somebody bring me my shotgun, quick!”
Damn hillbillies…
Edited 2007-03-16 00:07
I thought she was a she that wanted to be a he. I’m so confused.
http://www.lulu.tv/?p=4346 i’m pretty sure thats a guy. posted on nov 12, 2006…
I am getting really sick of people constantly beating the tired drum of “oh, Beryl stole it from Compiz.” It’s called OPEN SOURCE morons! They’re allowed to take any code from Compiz if they wish. Beryl has done a lot of hard work rewriting tons of code, adding new code and reimplementing various aspects of their compositing manager. It’s not fair the the hard working Beryl developers (and supporting community) to say that Beryl is simply a clone of Compiz. Especially when it’s not true. Beryl has forked now get over it already!!!
I am getting really sick of people constantly beating the tired drum of “oh, Beryl stole it from Compiz.”
Now are you…
It’s called OPEN SOURCE morons! They’re allowed to take any code from Compiz if they wish.
Oh, but nobody doubted that they can.
– However, should I not be allowed to point out that they don’t give enough credit to the project that provides them with 95% of their code?
– Should I not be allowed to point out that changing the lisence so that the project you rely upon to do all the hard work can’t incorporate your changes while at the same time you can and still do rely on the other projects code is quite disgusting?
– Should I not be allowed that relying on an other projects code for nearly your entire project but then not giving back bugfixes and on the contrary using the bugfixes to bash the other project is quite disgusting?
Beryl has done a lot of hard work rewriting tons of code
You know, that’s the thing, they haven’t rewritten anything.
adding new code
They added nothing to the core parts of the software. What they did was add some nice plugins, that’s all.
reimplementing various aspects of their compositing manager.
Again, they didn’t reimplement anything.
t’s not fair the the hard working Beryl developers (and supporting community) to say that Beryl is simply a clone of Compiz.
But it simply is a clone of compiz. Even some of the levelheaded beryl devs say this.
Proof of %95, Its do lines of code its relatively easy.
I thought they relicensed it so their work was under GPL and the rights within. They are not doing it out of spite. Making baseless accusations now that is disgusting.
How can it have addition plug-ins and still be a clone. I’m confused.
Proof of %95, Its do lines of code its relatively easy.
Well, if it’s so easy, do it. After all you made the claim that they made great changes.
Anyway, if beryl were so differnt from compiz, then why should quinn suggest to merge again and make beryl the first compiz++ distribution as she calls it?
http://lists.beryl-project.org/pipermail/beryl-dev/2007-March/00030…
A great suggestin, btw.
They are not doing it out of spite. Making baseless accusations now that is disgusting.
That wasn’t a baseless accusation. That they would still be able to take code from compiz while compiz wouldn’t be able to take code from them was clear to the beryl devs when they changed the license, wasn’t it?
How can it have addition plug-ins and still be a clone.
Well, aren’t you a clone if nearly all of your code and especially the nitty gritty hard parts come from an other project, even if you then add some nice icing on the cake?
so no proof, still baseless, and no understanding of what a clone it.
shame.
As you said yourself, the proof is out there, look it up.
so no proof, still baseless, and no understanding of what a clone it.
again
Edited 2007-03-15 19:14
so no proof
God dammit, look at the commits honey.
Look at the beryl hompage “Beryl is a fork of the Compiz project, started by David Reveman of Novell. We continue to port new changes from compiz, and consider them essentially our upstream. Beryl could not have existed were it not for the heavy lifting done both server side by David and in creating compiz, which is the base on which all of our code is built, and which still comprises a large portion of our code…”
Look at the discussions on the beryl mailing list about a merger.
still baseless
As pointed out already, not baseless at all. That you simply chose to ignore this doesn’t change the fact, I’m afraid.
no understanding of what a clone it
I already pointed out several times why it’s justified to call it a clone. That you again chose to ignore this is a shame, but doesn’t change the fact.
Again show the 95%. The reality is you plucked the figure from the air. There is nothing to be ashamed of for making up figures to prove a point.
So Beryl on there homepage of all places *acknowledge* the code they used.You imply some spite or nastiness or foul deed on behalf of the beryl team. Its a lie and thats disgusting. They just want *their* code under a different license, and fortunately the chosen compiz license allows them to do so.
A clone by its nature is a replica. This is simply not true. Clone implies they are the same; They are not; They are different. fork is not the same as clone. to say its a clone is a lie.
so no proof, still baseless, and no understanding of what a clone it.
again
Edited 2007-03-15 19:32
Again show the 95%. The reality is you plucked the figure from the air.
Oh Jesus Christ…
Of course my point wasn’t that exactly 95% of their code is from compiz, but that the vast majority of the code in beryl and the important parts at that are from compiz. As I repeatedly pointed out to you this can be easily verified by actually looking at the commits, looking at the beryl homepage and by following the discussions on the beryl mailing list.
So Beryl on there homepage of all places *acknowledge* the code they used.
After month and month of being reminded that they should do so, after a lot of discussions, they finally acknowledged it, yes. However, they obviously forgot it again in the release anouncement.
You imply some spite or nastiness or foul deed on behalf of the beryl team.
I don’t imply it, I say it openly.
Its a lie and thats disgusting.
It isn’t, as I repeatedly showed.
Do you want to imply that the beryl devs didn’t know the consequences of what the did?
Do you want to imply that they are not aware of the fact that they are taking from compiz (a lot) and don’t give back?
They just want *their* code under a different license, and fortunately the chosen compiz license allows them to do so.
Well, if that’s what they wanted, they wouldn’t have had to fork and relicense compiz, as you would know had you read the post by Quinn I linked to earlier.
A clone by its nature is a replica. This is simply not true. Clone implies they are the same; They are not; They are different. fork is not the same as clone. to say its a clone is a lie.
You simply don’t seem to know what the word clone means:
http://en.wikipedia.org/wiki/Clone_%28computer_and_video_games~…
http://en.wikipedia.org/wiki/Clone_%28computer_science%29
You don’t know what percentage…then don’t quote one. It could be 5% you haven’t a clue.
I feel nothing but sadness when reading your comments. You have convinced me that GPL is the *right* license for compiz.
Again its not a clone…by any stretch of the imagination, you describe it as a bug fixed compiz with plug-ins so I’ll except that.
From what I understand, it seems like beryl will try to use the compiz-core (with possibly some minor patching to hook on Beryl-specific functionality) and diverge on everything outside of compiz-core. Of course, Quinn hasn’t confirmed this yet (
http://lists.beryl-project.org/pipermail/beryl-dev/2007-March/00030… ) so I could be wrong. But if I’m right, then you still don’t know how much has changed unless you do a diff between each code base.
What can be said, however, is that Compiz *does* benefit from Beryl despite your claims that nothing is given back. Unless this post is incorrect,
http://www.johnstowers.co.nz/blog/index.php/2006/10/02/beryl-the-co…
Compiz benefits from several plugins that have been ported from Beryl and placed in the compiz-extras package.
Anyway, why be so hostile? Forks are a fact of life in open source and one if it’s key evolutionary strengths. If the fork doesn’t produce any real value, it dies (e.g. the old GoneME fork of GNOME). If it produces real value, it either kills the parent (e.g. EGCS forked GCC because it disagreed with the development process and then won over the GCC group when it succeeded) or competes with the parent (e.g. X.org versus XFree86).
So if you’re right, Beryl will die. If you’re wrong, it’ll live. Let time sort this whole thing out.
“That wasn’t a baseless accusation.”
If I understood intent of the original comment correcly, what you said is a baseless accusation. The argument wasn’t that they take but not give back. The argument was that your suggestion that they do so out of spite of compiz was baseless. Do you honestly beleive they changed the what licence they use in order to harm compiz?
“Well, aren’t you a clone if nearly all of your code and especially the nitty gritty hard parts come from an other project, even if you then add some nice icing on the cake?”
No you’re not a clone. You’re a derivative work, an offspring or mabye a fork.
Are you a Beryl developer? Do you work closely with the Beryl project? Do you even use Beryl? If Compiz is so freaking wonderful and Beryl is merely a cheap imitation, then why are so many top distributions choosing Beryl over Compiz? (And if you’re going to throw out the “Beryl gets all the hype” argument, then don’t bother.)
If Compiz is so freaking wonderful and Beryl is merely a cheap imitation, then why are so many top distributions choosing Beryl over Compiz?
Well, I’m not aware of any major distribuiton choosing beryl over compiz.
I also didn’t say that Beryl was a cheap imitation but basicly compiz with some experimental plugins add and nicer packageing.
Also, because of the packaging and because of the plugins compiz-quinn was already very popular and that of course carried over to beryl, especially as compiz.org (or whatever the exact domain of compiz-quinn was) one day all of a sudden announced that from now on beryl would be the thing.
So, I am lost, I keep Compiz in my almost-daily-visited bookmark, but use Beryl as my daily desktop.
The community/end user is not compiz devs.
I love how feeling from either side’s end users morph into the respective developers view on each other.
Would love to see official slackware packages of this:) and the 7.x series of x.org when I think about it
Hey, me again! Your resident 3D desktop nay-sayer!
I tried Beryl a few weeks ago. I was actually starting to see that there were some mild benefits to a composited desktop, once I turned off the really stupid stuff, like wobbly windows, and that silly, time consuming, crumpling animation as a window is minimized.
And then I realized that setting windows to be “On Top” was not supported.
I found this lack to outweigh the meager advantages, so I’m back to a nice, usable, 2D metacity desktop, which is notably faster on my Nvidia 6800 card with the latest Nvidia drivers.
Does the new Beryl have “On Top”?
The Expose’ clone thing (can’t remember the actual name in Beryl terms) was kinda sorta useful.
On top has been working fine for me for much longer than a few weeks.
I did have some issues with stickiness of windows when using emerald and gtk-window-decorator, but aquamarine worked just fine and, even stickiness works fine with the former two now.
Adam
“””
On top has been working fine for me for much longer than a few weeks.
“””
The On Top option appeared in the right-click drop down, but was always grayed out. I found a thread that explained that it simply was not supported.
This was on FC6 and Ubuntu Edgy. Both fully updated. And both just a few weeks ago.
I was using Emerald. Didn’t think to try Aquamarine.
Edited 2007-03-15 16:20
I still can’t believe “always on top” is missing either. I tried out compiz when SLED 10 was first released and I vowed not to use it until that basic usability was incorporated.
Do you have a link to the thread you mentioned, BTW?
I must confess I am confused on this. Using emerald on top is available and not grayed out for me. It has always worked when I used it, though I do not use it often.
Is it maybe a compile time option? For those who do not have on top, what window decorator are you using and what package?
I am running the official Ubunutu Edgy 1.9999 release since I have not updated to 2.0 yet. It is available on my Feisty 2.0 release as well.
oh well, that’s great news then, I haven’t tried out beryl/compiz since so this might motivate me
. Thanks for responding.
im with you, i really havent found this 3d desktop stuff to be anything other then ‘nifty.’ i find myself reverting back to my custom blackbox desktop all the time. the 3d effects look awsome but arent anything more then that too me. and we all no looks isnt everything.
“””
im with you, i really havent found this 3d desktop stuff to be anything other then ‘nifty.’
“””
Yes. It’s not what you’ve got, it’s how you use it.
Though I label myself as a naysayer, I believe that it is important not to polarize on the issue. Composited desktops are neither the cure for all the ills of man nor the spawn of Satan.
IMO, compositing is a tool that, if used wisely and conservatively, *can* have some usability benefits.
Expect the real benefits to show themselves when we *stop* seeing the phrase “eye candy” applied.
Of course, by then it won’t seem like a big deal and we probably won’t be talking about it anymore.
Edited 2007-03-15 16:32
“MO, compositing is a tool that, if used wisely and conservatively, *can* have some usability benefits.
Expect the real benefits to show themselves when we *stop* seeing the phrase “eye candy” applied.”
I saw half billion doller launch recently based around eye-candy, and I think beryl has done more to shake of the “Linux is only for the server” image than anything else.
Compositing does not have to be used “wisely and conservatively” it needs to be useful…and I want it to be pretty and useful.
“””
I saw half billion doller launch recently based around eye-candy,
“””
If it were really a usability revolution it wouldn’t take a half billion dollars to convince people, would it?
And the response to that product with the half billion dollar launch has been notably lack-luster.
I’m not sure I’d be using it as an example if I were arguing your side of the debate.
WOW, I just don’t know what to say.
“””WOW, I just don’t know what to say.”””
Well, I often have that effect on people. I’ve got you beat. Best to just admit it.
Seriously, though, my previous post did discount the marketing worth of composited desktops.
You make a good point that the bling has some intrinsic value in and of itself.
But I do think that marketing trends are transitory.
In the end, I still expect a more toned down desktop with those compositing features retained which are actually useful.
Can you agree with that? Or do we still disagree?
The best video I have seen of a compositing desktop *is* the original video of compiz with no *bling*.
I myself have no desire for flames or any other nonsense, but other that the flames, I can’t think of another feature that has no use on one level or another. Beryl has an awful lot bling from objects zooming fading rotating that are often Visual *indicators*. I wouldn’t know how you would even measure their value. I suspect buying a larger higher resolution monitor would in most cases be more *useful*.
The trouble is I hate the word useful, how about the desktop offers a smoother experience; greater accessibility.
This is without talking about control; ownership; of the desktop. Which come about from theming etc.
Well, I’m generally encouraged by what I have seen.
My real complaint is that the reality does not come anywhere close to matching the hype.
That’s what I meant when I said that it is best not to polarize on the issue.
I don’t care for the visual indicators, myself. It usually translates to time wasted. OK, maybe just a half second here or there. But when I’m working, a half second, uselessly wasted, seems like an eternity.
I bought this Athlon 64 4000+ and NVidia 6800 for a reason… and it wasn’t to wait while some animation or fade out or rotation completes, over and over again.
It gets old after the one hundredth time or so.
One OSAlert contributor even went so far as to say that those particularly aggravating wobbly windows provided him with “valuable feedback” when he moved a window. To which another OSAlert subscriber responded “Isn’t the fact that the window is moving across the screen enough?”. (Arguably, one of the best one liners that has ever appeared in this forum!)
I do welcome our new 3D desktop overlords. But not too warmly.
The composited features need to prove their worth.
But rest assured that I am not so arrogant as to presume that my needs are the only ones that matter.
I’m confident that time will sort this out.
“I don’t care for the visual indicators, myself. It usually translates to time wasted. OK, maybe just a half second here or there. But when I’m working, a half second, uselessly wasted, seems like an eternity.
I bought this Athlon 64 4000+ and NVidia 6800 for a reason… and it wasn’t to wait while some animation or fade out or rotation completes, over and over again.”
Ignoring the nonsense. Visual Indicators are a *small* part of the useful check out Beryls accessibility section which I think is the *most* important section For me the ability to not look at a light bulb I find a great help. This is followed by the *bigger* desktop without having a bigger desktop section, the Visual Indicators are often about finding stuff quickly, and remembering where stuff it. Followed by responsiveness, the being able to move your mouse smoothly, windows smoothly. I’m not sure where windows management fits into this, but with things like tile ans tasks.
Now the *only* and I mean only example of transition actually taking place where your have to *wait* for it it drop down menu’s and the start button, both of which can be switched of, are the animations speeded up, but these are tiny sections.
BTW buying a monster PC to save a few seconds of a menu popping up is overkill. I hope thats a joke, as there alternatives to that, from I/O schedulers to alternative WM or Taskbars.
OK. I think we need to be careful to separate implementation details from innate design features.
I found Beryl to be irritatingly slow on my “monster pc”. BTW, I bought the NVidia card for Doom 3. And got a good deal on the Athlon 4000+.
I disagree that artificially slowing down the desktop with time-based animations is a good thing.
That’s an implementation detail.
The default WM and Taskbar really should be fast and efficient.
Also an implementation detail.
And I absolutely do not see what the I/O scheduler has to do with this.
You’re surely not implying that Beryl is trying to write a bunch of stuff out to disk?
But if it does, that is also an implementation detail.
On a related note, the Titanic had only transverse, not longitudinal, watertight bulkheads. And they only extended up to “E” deck.
Implementation detail…
Edited 2007-03-15 19:02
will you stop leaving such big gaps in your post.
we will play a little game, just to prove my point. You list all the time-based animations, that slow your working under beryl.
…and I’ll list all the plug-ins that provide all the things I talked about in my post that have little to do with the time-based animation.
I bet *my* list is longer.
I’m surprised your *default* WM taskbar isn’t efficient, mine is. I just said there are ways to speeding up your desktop *without* spending hundreds of pounds on a new system.
“””will you stop leaving such big gaps in your post.”””
What big gaps?I found the “crumpled up” window minimizations to be a problem.They tended to freeze for maybe a half a second to a second,midway,every time I minimized a window.The wobbly windows were irritating and took longer to be able to work in than my 2D metacity managed ones.Window maximization also required extra time as the windows exhibited their rubbery bling.Most everything that was an expression of 3D came with an irritating price tag.Just how many modifications and plugins do I have to install and configure before I start seeing the wonders of usability that are promised?
Edited 2007-03-15 19:44
window minimizations
Window maximization
wobbly windows
So three. which you can turn off or speed up animations
I give you
Highlighted focus
Zoom
Negative<–If you work at night essential, my top tip
Opacify
Desktop Plane
Windows Grouper
Application Switcher
etc.
and I’m not really trying.
“””
I give you
Highlighted focus
Zoom
Negative<–If you work at night essential, my top tip
Opacify
Desktop Plane
Windows Grouper
Application Switcher
etc.
“””
Cool.
Explain what these features are and what they are good for.
I’m listening.
You really might convince me if they are truly useful.
Edited 2007-03-15 19:57
at a suggestion. Try turning off the animations and wobbly windows plugins (under Visual Effects). Then turn on Snapping Windows under Windows Management.
That might be about the compromise you are looking for.
Alternatively you can adjust animation speed and choces to give you the responsiveness level you like. This is what I did, but it did take me a couple of weeks of occasional tweaking till one day I noticed I was not tweaking anymore. (I did not spend a lot of time, maybe 30-60 min total. But i would do a minor adjustment and see if I liked it, then a bit later another, etc…)
“I think beryl has done more to shake of the “Linux is only for the server” image than anything else.”
I seriously don’t think anyone other than Linux users and some OS enthusiasts have ever heard (or care) about Beryl so it’s impact on the Linux image would be pretty marginal (so far).
Edited 2007-03-16 03:53
I seriously don’t think anyone other than Linux users and some OS enthusiasts have ever heard (or care) about Beryl so it’s impact on the Linux image would be pretty marginal (so far).
You should take a look at the beryl forums. It is full of new linux converts… The downside is that the vast majority of them are unable to get beryl working out of the, and noticeably frustrated as a result.
Adam
“You should take a look at the beryl forums. It is full of new linux converts… The downside is that the vast majority of them are unable to get beryl working out of the, and noticeably frustrated as a result.”
I hope this is not the case…but it would not surprise me. The move to a compositing window manager is painful, and has not been helped by proprietary drivers; experimental this and that; and a whole host of other things. The truth is getting a compositing window manager on Linux has been a hit and miss affair. For me it only went stable 3 weeks ago…and I’m not even sure what fixed my various problems, from white cubes, to restoring having a “delay”, video playback is still not what I would like.
Beryl *is* one of the reasons to switch to Linux especially if you have an Intel laptop, but like much of the cutting edge stuff in Linux it takes time to be stable. Whether this will come about now after this release and Xorgs7.2 release, I don’t know. I suspect it will for many. I suspect some with have to wait for Xorg7.3 in about 5 months or the release of open source NVidia drivers or AMD R500 and above drivers.
The reality is though. This is the best thing that has happened to Linux for a long time. Graphics Card Support has been a weak point in Linux for a long time, and suddenly there is a great deal of enthusiasm in this area. I think this will help Linux adoption in Long term.
Beryl is not just about fluff. It’s also about usability enhancements. See here for a least of great usability features in the latest release:
http://liquidweather.net/howto/index.php?id=92
http://liquidweather.net/howto/index.php?id=101
http://liquidweather.net/howto/index.php?id=102
http://liquidweather.net/howto/index.php?id=104
http://liquidweather.net/howto/index.php?id=105
http://liquidweather.net/howto/index.php?id=107
http://liquidweather.net/howto/index.php?id=106
Now that 0.2.0 is out all the features “featured” above should be available to all.
I also was very skeptic about Beryl, but I’ve tried it and I won’t go back easily. I have disabled cube etc., (I have kept wobbly windows just because it’s unobtrusive eye candy), but I can assure that the Expose-clone and the Wall plugin are real improvements.
You can read about Beryl and usability http://liquidweather.net/howto/index.php?id=92“>at , it’s enligthening.
I think ive had 0.2.x installed in ubuntu for several months now..
nothing like having a command line driven operating system look more like vista!!
Here’s a recent desktop. (note: i use both ubuntu and vista)
http://www.sn0n.com/wp-content/uploads/2007/01/vistagnome.png
I don’t see why Beryl etc. shouldn’t implement a blur filter. Then you really could emulate Vista
.
They do
Oh. I didn’t see any when I tried it and I haven’t yet seen a screenie showing it. I assumed that it hadn’t been implemented.
my original post shows a screenshot with trans.
But not with blur.
Obviously a bit much for everyday usage but:
http://img252.imageshack.us/my.php?image=screenshotvy5.png
Wow. Thanks. When I get round to installing Beryl, I’ll have a play.
Please don’t use the words Beryl and Vista in the same sentence. It is insulting to the Beryl developers.
This theme http://themes.beryl-project.org/repo/images/26-Blueray.emerald.png looks pretty cool. Anyone got other screenshots of this?
Thing that annoys me about beryl: terrible bug handling. I’ve filed four bugs, all with extremely detailed explanations of what should be fairly trivial issues to resolve (three of them being cases where heliodor simply needs to be modified to behave the same way metacity does in a given situation). not a single one has been touched in any way by _anyone_ since I filed it.
gneesh.
tickets 1265 and 1671 through 1673, for anyone interested.
tickets 1265 and 1671 through 1673, for anyone interested.
Ok I took a look….
First off, for a user, your bug reports are ok, but they are missing a lot of information that would have made researching the problem easier. I am sure you consider them extremely detailed, but from a developers standpoint they were missing quite a bit of information that would have been useful. At a suggestion try and remember to include the version of the effected program(s), the architecture compiled for, the distro used, and the compile options if you compiled it. All of these elements take but a moment in time to type and help narrow down a test environment to try and reproduce the problem.
1265 is interesting, but lower level into X than I am capable of. Have you tried in Compiz? In this case it might be a bug in Beryl but the Beryl coders rely on Compiz for much of the core X code (and contrary to popular opinion do note it as such in the commit logs) and tend to concentrate on the plugin architecture and plugins. Also see my notes above about listing distro and version info (in this case of X.org).
1671 through 1673 are a slightly different matter though. Suspecting quirks in window hints I did a quick search. Assuming you are running .20 or .21 of Gimmie have you tried .24? In the change log for Gimmie are notes for patches that appear to be attempts to correct this. The original patch was in 0.22 and additional corrections in 0.24. If you had listed the version of gimmie I might have been able to discount this (hint, hint). If you are using 0.24 I will try and compile gimmie and see if I can reproduce the problem.
NOTE: I am not a Beryl Dev (or a Gimmie dev for that matter) but the problem seems to be misdiagnosed so I hope to offer some assistance in getting the correct project patched. I do not make any promises though.
running gimmie 0.24:
[adamw@lenovo usb-rndis-lite]$ rpm -q gimmie
gimmie-0.2.4-1mdv2007.1
you’re right that I should have mentioned that, I tend to always run latest versions of things and forget that others may not assume that I am.
no, I haven’t tested the first bug with compiz. I will, once compiz is working again in mdv cooker.
x.org version, well, should be pretty obvious from the fact that the bug report is about xrandr 1.2: running the server-1.3 branch from git.
x.org version, well, should be pretty obvious from the fact that the bug report is about xrandr 1.2: running the server-1.3 branch from git.
So you are running an unstable branch straight off the repository using new features and you are filing a bug with Beryl? Sorry, but that is not reasonable. An RC is not a release and it is not in use by any major distro. Not only is the bug unreasonable, it is detracting from bugs that should be addressed. (regardless I already admitted this one was beyond my ability)
I will see what I can find on gimmie (also an unstable program, but at least an officially released one) over the weekend.
Edited 2007-03-16 03:19
it’s entirely reasonable. the behaviour of randr 1.2 is already clearly defined; it won’t change before the release of the final server. and metacity is already entirely capable of handling it. if the problem were that beryl had already been adjusted to cope with randr 1.2 but there was a bug in X.org causing it not to behave as it’s supposed to, that would be a different matter. but that is not the case.
to put it another way: if nothing changes in beryl before the release of the final X server 1.3, then this bug will still happen with that. I absolutely guarantee that.
No, it is not reasonable because it is not a released product. What your are asking the devs to do is install and configure a moving target for one user using a beta of X.org. Yes it is likely going to be a problem on release (for compiz as well) but it is not something that can or should be supported yet.
If you think it reasonable for the devs to have to compile X server 1.3, look into randr 1.2, then attempt to fix an interaction issue while the X server is still under development, then I have nothing I can say to sway you. The team have enough to do supporting environments that are officially supported by released products without going off and downloading head from git.
For that matter you can not guarantee that something won’t change in X server 1.3 to correct this. (Highly unlikely, but possible.)
That’s not what I’m asking them to do at all. I’m asking them to prepare in advance support for an extremely important future feature of X.org which will be used by a huge amount of people, in order to save themselves the trouble of having zillions of duplicate bug reports from pissed off Intel users the day the randr 1.2 stuff goes final (or the day it shows up in an Ubuntu release, or whatever). It’s not at all unusual to code support for a planned future feature of a product your product depends on in advance, nor is it at all unreasonable for me to expect the Beryl guys to do this.
Or, more to the point, for me to expect them to respond to the bug by saying ‘we don’t want to bother doing this yet’. That would *also* have been fine. Complete neglect is not.
those who complain it is slow etc (i read some guy posted his athlon 4000 could not handle)
Check this beryl videohttp://www.youtube.com/watch?v=dJO6CLln-B0
Remember , its just a centrino laptop 1.6ghz and 512 ram with Nvidia geforce2go 7400 . and i was natively recording this video. but still all actions in the window manager are smooth like butter.
I am only imagine things must have gone tad faster in the latest stable release.
(right now i have 1.5gb ram just upgraded)
Well, let me understand this:
Take code from a project w/o giving anything in return isn’t against the GPL spirit?
Because that’s what I’ve told before.
It is against the spirit of the GPL to modify and distribute GPL’ed software w/o giving back.
However, this does not apply fully for LGPL or BSD/MIT licenses, because the spirit of those licenses are different.
OTOH, by releasing the code under the GPL the code is given back to the community, so the spirit is not violated.
Im sorry but I don’t buy your excuse, because “The community” is not exclusive of the GPL license, MIT license HAS also a community that doesn’t get anything back from beryl, to me at least is not moralty correct.
No licenses have a community.
However, the FLOSS community has several licenses. Whether source code comes back with one or another license is irrelevant to me, as long as the license allows for free use, free modification and free redistribution.
The funny thing here is that you wouldn’t bat an eye if Microsoft took parts of the compiz code, modified it and released as a proprietary package. You’d be happy to see that, and so would the compiz devs. But when it is released to the community under the GPL it’s suddenly very very bad. But not releasing it to the community is perfectly okay with you.
Talk about double standards
While I agree it is not against the letter of the MIT License and LGPL, I would not conclude that it is not without some fault.
The Compiz devs have expressed consternation and disagreement over the need or merit of shifting to GPL at this time. While there is no debate that it is perfectly legal, I would say that it is not being helpful to the Compiz community. While some beryl devs have worked with the compiz team to send patches upstream it would be hard to argue that changing the license to GPL has made the process more difficult. And as has been pointed out the framework which they depend and run on (X) is IT licensed as well, yet they have no problems with that.
If there came a reason or example of misuse of the code being contributed to beryl, then a license shift might have merit. But as it stands they gain little and cause hurt feelings as well as not reciprocating in due kind to the project they admittedly are based upon and use. I would not classify this leeching off of another project to be a stellar example of how the open source community should act.
It is the relicensing of Compiz to GPL that I am most critical of when it comes to Beryl. The coding styles and quality I have discussed in the past and see no need to repeat.
Edited 2007-03-16 01:41
The license of compiz hasn’t changed. It’s still MIT licensed. Beryl is a fork and under a different license. So what?
The license of the X-server is irrelevant of several reasons. One of them being the Beryl devs not contributing to X.org (or at least I sincerely doubt they do).
What the Beryl devs have done is perfectly within the spirit of the MIT license. And it doesn’t break the spirit of the GPL license. The compiz devs lost any right to complaints on ethical grounds when they chose the MIT license.
EOF.
That right there says it all imo. The fact that the beryl project DOESN’T contribute to the xorg effort. Compiz does and all the changes that are made by Dave can benefit any other project not just Beryl. That is where the issue is. The issue is that beryl can leech off of compiz with no imperative to give back to the community that started it. The license used shouldn’t matter, its about etiquette and the beryl team has shown little to none. Yes they give acknowledgment to the compiz team, but not before they scream from the rooftops about this incredible feature they just implemented in beyrl. The user may find out about the origins of said feature whenever it shows up in the changelog. There are quite a few examples where this has happened.
BTW, someone said that most major distros are going with beryl. As far as I know compiz has been the winner when it comes to what is supported by the major distros. Let’s see, Novell(duh), Redhat(Fedora), and Ubuntu(desktop-effects installs compiz by default). Doesn’t get more major than that.
“The issue is that beryl can leech off of compiz with no imperative to give back to the community that started it”
If that bothers them so much then they should have picked another license. It’s as simple as that.
“The license used shouldn’t matter, its about etiquette and the beryl team has shown little to none. ”
The license doesn’t require them to show none.
“As far as I know compiz has been the winner”
So what the hell are they complaining about then?
They can’t pick another license if they want to include compiz technologies into X so that all composite managers can benefit. I thought that was obvious.
I think it bothers the compiz users more than anything, not the compiz developers.
Well ironically I am a Beryl user but I do not find them without fault. I just find the Beryl Project more usable.
I think it bothers the compiz users more than anything, not the compiz developers.
You do not seem to have read the developer discussions on the subject.
Read this:
http://lists.beryl-project.org/pipermail/beryl-dev/2007-March/00030…
This is a suggestion from Quinn to use a MIT-licensed core, and submit MIT-licensed patches back to compiz, while extensions are under the GPL.
I’m not going to spend time right now on your post, because it is so factually incorrect and unstructured, I can spend hours trying to put some structure into it before replying.
The issue is that beryl can leech off of compiz with no imperative to give back to the community that started it.
This is not an issue. This is a deliberate decision from the compiz devs. If they don’t like it, they should’ve chosen another license. Besides that, the statement is wrong. The community that started it is the FLOSS community. And it still gets the changes back. Different license – same shit
However – this debate can turn out to be pretty moot if the Beryl core goes MIT (and it looks like it).
This is a suggestion from Quinn to use a MIT-licensed core, and submit MIT-licensed patches back to compiz, while extensions are under the GPL.
You seem to have overlooked this in your holywar against those critical of compiz, but this was already posted in this discussion.
I’m not going to spend time right now on your post, because it is so factually incorrect and unstructured, I can spend hours trying to put some structure into it before replying.
That’s sad, as the post you are replying sums up the issues pretty well, though the author doesn’t seem to be a native speaker. I hope you can forgive him for that…
This is not an issue.
No, that’s exactly the issue. That in all your posts you try to pretend it isn’t doesn’t change this.
This is a deliberate decision from the compiz devs.
No, it’s a deliberate decision from the beryl devs to change the license. That the compiz devs chose the MIT license for some good reasons doesn’t change the fact it was the beryl dev’s decision to put beryl under the GPL thus making it impossible for compiz to incorporate their changes.
If they don’t like it, they should’ve chosen another license.
What kind of weird morale is this? Just because something is not illegal doesn’t mean it’s all right.
You seem to have overlooked this in your holywar against those critical of compiz, but this was already posted in this discussion.
I am not on a holy war against those critical of compiz (nor am I on a holy war against those critical of Beryl – that’s what you meant, wasn’t it?). And not – I’ve not overlooked it. I just saw it _after_ the discussion started.
That’s sad, as the post you are replying sums up the issues pretty well, though the author doesn’t seem to be a native speaker. I hope you can forgive him for that…
Nope, it doesn’t. The poster and you forget in your holy war against Beryl that compiz isn’t a X.org project, that compiz does not (at the moment) contribute anything to X.org – and even if it did using GPL’ed software on top of X.org does not pose a problem.
No, that’s exactly the issue. That in all your posts you try to pretend it isn’t doesn’t change this.
There are no technical issues – and nobody (of the compiz-devs) are complaining about technical issues. The only problem is compiz-devs whining about somebody using MIT-licensed code in a non-proprietary and non-MIT-licensed project. Had Microsoft taken the code and used it in a proprietary product you’d be celebrating it – and so would the rest of the anti-GPL zealots. The persons in here are solely complaining about “lack of credit” given to the compiz-devs despite the fact that more than due credit is given everywhere. In the sources, in the binaries (dialogues) and on the website. This is how the MIT-license is intended to be used. If this is against the intention of the compiz-devs then they chose the wrong license. They told everybody that forking the project was perfectly okay, by deliberately choosing the MIT-license.
No, it’s a deliberate decision from the beryl devs to change the license. That the compiz devs chose the MIT license for some good reasons doesn’t change the fact it was the beryl dev’s decision to put beryl under the GPL thus making it impossible for compiz to incorporate their changes.
Beryl-devs have submitted patches under the MIT-license. Besides that it’s not a problem. By deliberately choosing the MIT-license the compiz-devs gave up any legal and ethical right they had to derivatives.
If they don’t like it, they should’ve chosen another license.
If you say I can eat your food and I do – does that make it wrong? Nope, you said it was okay, and therefore it is legally and ethically okay.
The same goes for Beryl. By choosing the MIT-license the compiz-devs told everybody that they didn’t care what you did with the sources as long as you gave them credit. Therefore they have zero legal and ethical right to complain. Beryl-devs are merely doing what the compiz-devs are urging everybody to do (as implied by the choice of license).
If it’s not illegal it is all right. There is no such thing as ethical absolutes.
Nope, it doesn’t. The poster and you forget in your holy war against Beryl that compiz isn’t a X.org project
Who said it was?
that compiz does not (at the moment) contribute anything to X.org
Compiz devs are contributing to xorg.
and even if it did using GPL’ed software on top of X.org does not pose a problem.
And nobody claimed it did. Your point is?
There are no technical issues – and nobody (of the compiz-devs) are complaining about technical issues.
Who said there were?
Had Microsoft taken the code and used it in a proprietary product you’d be celebrating it – and so would the rest of the anti-GPL zealots.
Oh, Jesus…
One day I’m a GPL zealot here on osnews, now I’m an anti-GPL zealot and MS lover. Go figure…
The persons in here are solely complaining about “lack of credit” given to the compiz-devs despite the fact that more than due credit is given everywhere. In the sources, in the binaries (dialogues) and on the website.
Nope. Read the discussion agein.
This is how the MIT-license is intended to be used. If this is against the intention of the compiz-devs then they chose the wrong license.
1. As has been pointed out several times already, there were good reasons for choosing the MIT license.
2. As has also been pointed out several times already, just because what the beryl devs did was legal does not mean it was good for the free software desktop nor does it mean that it was moral.
If it’s not illegal it is all right.
If that’s your moral, then that’s pretty sad, but we’ll probably just have to agree upon having opposing moral values then. Btw., I’ll remind you of this when there’s talk about the Novel-MS deal, binary video drivers, etc.
> > This is how the MIT-license is intended to be used.
> > If this is against the intention of the compiz-devs
> > then they chose the wrong license.
>
> 1. As has been pointed out several times already, there
> were good reasons for choosing the MIT license.
> 2. As has also been pointed out several times already,
> just because what the beryl devs did was legal does not
> mean it was […] moral.
Your 1 implies that the compiz people want to use the MIT license, which is for any and all kinds of adoption.
Your 2 implies that the compiz people really wanted a license that prevented incompatible forking.
These two options are incompatible, you can’t have it both ways.
Let’s say I release something under the GPL (for some “good reasons”), but I don’t want others to release modified versions of it. Then when others do release modified versions of it I complain about it. Other say that I should have used a different license if I don’t want people to release modified versions. To this I reply: “But I had good reasons to choose the GPL!”
Do you see how illogical I would be in that situation? I can’t have my cake and eat it too.
Having good reasons for choosing a license does not mean that you don’t have to face the consequences of your choice.
Edited 2007-03-16 09:04
Your 1 implies that the compiz people want to use the MIT license, which is for any and all kinds of adoption.
It doesn’t imply it, it says so clearly. As to why they used MIT, look at the post by DavidR allready linked to in this discussion.
Your 2 implies that the compiz people really wanted a license that prevented incompatible forking.
No, it doesn’t. It doesn’t even mention what the compiz people want, nor is it relevant to the issue at hand what they want.
What’s relevant is what the beryl devs chose to do with the possibilities the MIT license gave them. That you guys seem so bend on ignoring this is annoying, but it’s also very telling.
First off nowhere did I talk about “the spirit” so drop it please. I said that to take code from the project that is the core of yours and not reciprocate in kind is wrong. This is an opinion that obviously you do not share, but “spirit” interpretations are also an opinion.
And no, “they should have picked another license” does not patch it up and make it right. If I patch a project to fix a problem, I consider it good manners to send a note to the original project with a patch and an explanation of what it does. This makes my upstream better and the overall product more stable. Some beryl devs (not Quinn) have begun offering patches to Compiz and have opened channels of communication with Compiz. And if they had done that from the beginning then none of this would matter. But this is a relatively recent turn of events (since about .2 beta 1). Before that upstream patches were less frequent and spotty.
As far as the mailing list talks go, yes I am aware of it since I am subscribed. It is still an internal discussion though and not without contention. You hit the highlights that got mentioned in the Beryl forums but that is not the whole story. Since it is an internal discussion at this stage, I am not sure what it has to do with their past actions. Further given Quinn’s post on direction for Beryl it looks like she decided to not go the route discussed in the ML. It might happen still, there are certainly a significant number of devs who want it. Quinn just happens to be one dev who happens to be reluctant to share a common core under a common license with compiz.
Edited 2007-03-16 11:24
Where one can learn a few programming languages and paradigms and be heralded as a genius. No offense, but I’m more impressed with those that develop the mathematics, applied physics, engineering and whatnot of the world who also have to know 2 or more languages to get their work done. These folks are geniuses. This guy has a project. That is that.
Beryl seemed to have issues whenever I tried to get it to display in 1680×1050 (my monitors native res).
Anyone know if they’ve sorted this out?
beryl adopted GPL. As far as I know, it’s not any exotic kind of license. Looks even rather mainstream. So, why howling this way?
All this is plain pretext for hate, jealousy. Not a word from Compiz supporters in these posts about the sameful hacking of Beryl data last december by a prominent member of the Compiz community.
Beryl did nothing illegal. Right of wrong? Right. OK.
One Compiz leader did something illegal. Right or wrong? Right again.
So, stop all this nonsense, if Compiz is not able to use GPL…well, it will be going the Novell way, I guess.
a prominent member of the Compiz community?
Compiz leader?
It was some random member of the compiz and beryl forums. Stop trolling.
Compiz is dual MIT/GPL licensed to be able to share code with the xserver and beryl is GPL only. That means beryl can take any code they want from compiz and relincense it under the GPL while compiz can’t take code back from beryl. No, it’s not illegal.
“However, since it’s release, many parts of Beryl have been rewritten, replaced, or simply dropped. It’s filled with eye-candy, better user support, new features, and stability.”
Aka. They did a SVN update from Compiz, and then made minimal acknowledgment.
Aka. They did a SVN update from Compiz, and then made minimal acknowledgment.
Any documentation for that claim or are you just trolling and whining?
I just installed Beryl on Fedora 6 and it is interesting. I attempted to use the Beryl manager to switch to Beryl for a window manager and it crashed X. I used the Beryl manager to swich to Compiz and it worked.
At first I thought all this transparency and 3D stuff was kind of gimmicky but then I found the Expose like scaling feature (Ctrl+Alt+Up Arrow) and I was like “Woh, cool.” I turned off the annoying wobbling windows and then I loved it. The new scaling and Alt+TAB and Ctrl+Alt+Down Arrow features are actually very useful to see which window to switch to quickly. In general the system looks beautiful. It is amazing how all this runs on half a gig of memory on an old Athlon Thunderbird CPU, granted I did recently purchase a Nvidia 6500 video card so I could run 3D better.
It is also interesting how switching to Compiz has effected my behavior. I use a lot more workspaces now instead of minimizing windows. I then scoll between the workspaces to switch windows.
Edited 2007-03-16 10:54
I started reading some of the more recent responses in this discusison; what a waste of time. People whining about licenses and the like and no decent discussion on the topic at hand.
“I started reading some of the more recent responses in this discusison; what a waste of time. People whining about licenses and the like and no decent discussion on the topic at hand.”
Beryl getting to a 0.2.0 release, and releasing more features is a milestone in itself…and new features come out all the time.
but Beryl is interesting for other things than wobbling you windows. The License is interesting for two reasons. They have chosen it because the wanted it for their code, and its incompatible with that of the original project, and the license allows just that, and people are actually upset about it. I’m looking forward to quoting this is *every* BSD discussion from now on. Its also going to be quoted in every Solaris kernel Discussion from now on, becuase Beryl can cherry pick code from compiz, and carry on with their flashy headline grabbing plug-ins.
If I was going to get upset…I would have chosen a different License. I actually think this is getting to be about credit and control.
Now the reality is the projects *are* different with different philosophies and goals. The simple fact that one uses a mailing list…and the other a forum speaks volumes, and this is interesting too. Compiz is looked at as a closed off; company project; does things the right way; slow moving; unpopular. Where beryl is seen as an open; community based; hacky; rapid development; headline grabbing.
Because of the one-way License Beryl will always look cooler; ahead on features; overshadow Compiz. Even though a lot of the *real* work is done in Compiz, and this is becuase of the Licenses. Everyone *accepts* this. Most of the arguments are about whether malice was the intent of the chosen License. I have seen *NO* evidence of this. The harsh reality is I can’t see the problem as the License allows them to do just that…and the point is Beryl doesn’t want others to do that with *their* code.
Whats also interesting is both sides seem to be trying to make a connection. I personally would rather they didn’t partially becuase I agree with the Licensing of Beryl, and why the license was chosen, and the fact that it makes a damn good development model. The only thing wrong with it is peoples egos.
Now the reality is the projects *are* different with different philosophies and goals. The simple fact that one uses a mailing list…and the other a forum speaks volumes, and this is interesting too.
Which project does what?
Beryl forums:
http://forum.beryl-project.org/
Beryl mailing list:
http://lists.beryl-project.org/pipermail/beryl-dev/
Compiz forums:
http://forum.go-compiz.org/
Compiz mailing list:
http://lists.freedesktop.org/archives/compiz/
Compiz is looked at as a closed off; company project; does things the right way; slow moving; unpopular. Where beryl is seen as an open; community based; hacky; rapid development; headline grabbing.
This is exactly the “wrong assumptions” David wrote about.
A few weeks ago I actually went looking for the differences between Compitz and Beryl. I found no real differnce apart from philophosy.
I don’t want to do Beryl a disservice (I use Beryl not Compitz) but I think I can sum what makes Beryl different from Compitz.
Beryl adds hacks and workarounds to Compitz code to add eye candy type features which users are asking for. Does it really matter that the licence differs?? Not really because the changes are hacks and work arounds – the vast majority of which would not be welcome in Compitz since most Beryl hacks are exactly that – hacks.
When compitz fixes the underlying structure for certain operations Beryl then drops some of it’s kludges and comes out with even more eye candy hacks that have been requested by users.
Beryl benefits Compitz by bringing ideas to the table for the compitz folk to code properly and Compitz benefits Beryl by improving the framework upon which Beryl is based upon.
People seem to forget that kludging and hacking is easier and less time consuming than fleshing out a framework – move too fast on a framework and you can easily end up hitting a wall and have to start all over again – KDE is a prime example of this type of development.
thingi
Edited 2007-03-18 13:07
i use latest nvidia driver, suse 10.2 64 bits and beryl 2
i lost often my mouse cursor… i do a right click… and my mouse cursor is displayed…
any idea to resolve that?