Xwayland/Xmir is not a magical separated magic but is just a translator from X11 api to Wayland/Mir API, so if in Mir the operation took 100 cycles and in wayland took 5 cycles you will notice it in the xorg apps right away for example, it just a translation layer so the speed on the actual backend regardless of how the layer convert between the two
I thought XWayland and XMir were different implementations of the X api that used their respective backends. But it seems that was too much work to put on so instead they just created another layer .
It would have been of much performance if the x.so library was replaced with a a different implemtation xwayland/mir.so
Applications running under XWayland will also have a performance hit. It is impossible to have a compatibility layer without a performance hit.
Not necessarily. if you actually look at the architecture of XWayland and XMir you actually avoid a few expensive steps at the end vs real X. In theory, its possible for us to have slightly better performance via XMir and XWayland than Real X. But obviously an alpha-quality release won't have the optimizations and design to actually make that claim true just yet.
We're on page 7 right now. How many pages would we have, if we left out any BO$$ post and related reply, while increasing overall quality?
I must admit that BO$$ has the most impressive tackle box of lures here. Not that there is any sort of difficulty getting a bite around these parts, but he's got a certain finesse and effectiveness that has not gone unnoticed.
So in order to make sure they get it right we will probably have Wayland ready somewhere around 2050. Or 2100 to get it REAALLY right.
Then make your f-ing Project Leaders and their ass-licker developers to contribute the Wayland Project in order to make it work properly. (as they desire - since it's actually working.)
Originally Posted by BO$$
One word: Freedom. They can do it. They can fork the entire linux kernel and create their own OS and create the greatest marketing campaign on Earth and sell it and become more successful than all the distros combined.
And if they fork the hell out of our kernel, Linux Ecosystem and Linux Community achieves nothing. Canonical forks everything from Linux and becomes what Apple became after BSD. - Like a BO$$.
Everyone has their freedom of doing whatever the f! they want to. But what would our community gets from this? Not all of us are blind Canonical fanboys and in need of a new "Steve Jobs"-a-like. We need an overall progression and Distribution Managements are getting more retarded everyday and not trying to let their community to test some certain part of this ecosystem. They just "move" - How the f! is this logical?
Originally Posted by BO$$
It's been 95% done for a couple of years now. That Weston genius from what I heard had trouble with the minimize button. As in they didn't have it. Brilliant reference implementation. Everything in Wayland is a sad joke and free software hipsters won't see it because they think Canonical == Evil while Wayland == Good. Mir is already the future. And it's already here. Wake up people!
Mir is not the future. It's yet to prove itself. When it's ready for a showdown, we're going to test it and like or dislike the way it's walking. But this objectivity (which fanboys like you cannot understand) is for better future for our ecosystem. We're not some company's fanatics or just a bunch of hippies; we're people with different choices, different ideals. The one thing binds us here is Linux Ecosystem and this ecosystem is more than just our hobbies but our jobs.
Before we decide something, we analyze the every possible aspect of such work. Then we decide if that's for us or not. But raising our voice cannot hurt any f-ing homo sapiens sapiens on this pathetic planet but only some fanatically blind people like you.
Originally Posted by BO$$
One word: freedom. Because they can. They can fork the whole kernel and create their own modified version and then create the coolest marketing campaign on Earth and wipe the floor with all the other distros.
Freedom. Because they can consume the hell out of their market. Freedom to consume and to be consumed. Freedom to stay blind but saying otherwise. Freedom to decline every logical path and just lick the path some Rich Hillbilly Boy decided.
But, yes, they have this freedom. It's just, we're sick of Apple wannabees and Apple Fanatics.
If you are interested the highest profile case I can think of this type of thing happening is with Open Office (originally under GPL). Sun used to have a similair agreement for Open Office contributions. Sun changed the licence so that it could licence the code to IBM under a closed source arrangement and when Oracle took over this is how they were able to relicence it to an apache licence and donate it to the apache project. Anyway thats why LibreOffice is GPL and OpenOffice is now Apache.
Apache actually changes the APL from Oracle OpenOffice (that still retained the LGPL3+) year after official LibreOffice launch due Apache own policies that suits IBM interest under Rob Weir who often keeps smear The Document Foundation effort.
They can fork the whole kernel and create their own modified version and then create the coolest marketing campaign on Earth and wipe the floor with all the other distros.
If a day comes when Ubuntu = Linux (or = GNU/Linux, if one is so inclined to refer to it that way), or when Ubuntu has even displaced Linux in the minds of users, a time when Ubuntu has become essentially the only distribution, then who would pick up the slack that would be created in the funding and development of open source? Canonical? One interesting thing to come out of the rolling release discussion on the ubuntu-dev mailing list is that (currently) about 75% of Ubuntu is unmodified Debian. If Debian was laid low because Ubuntu had wiped the floor with it, is Canonical (or the Ubuntu community, to the extent that their contributions are welcomed/accepted by a Canonical which operates as gatekeeper) going to take on that burden? This of course does not even take into account upstream projects, many of which to varying degrees rely on contributions and funding from both individuals and companies behind other distributions. Are many survivors of some Canonical + its partisans perpetrated floor-wiping crusade of the non-Ubuntu Linux world going to want, or be willing or able to move to Ubuntu as the be-all and end-all platform once their respective distro communities have been brought down by Canonical? A bazaar isn't really a bazaar if there is only one shopkeeper.
Granted, the aforementioned hypotheticals are highly improbable.
I really have no problem with the existence of Ubuntu the distribution or the fervent devotees of the Ubuntu community. I have, however, become less enthusiastic about Canonical as time has passed. I also don't care for either-or mentalities which often seem to crop up on any side of any given topic at Phoronix. I do hope that the success of Wayland and Mir is related to technological and performance merit and not popularity, which don't always go hand in hand. While there is enough feuding to fuel various biased camp fires merits be damned, it's too early to say definitively how well these technologies will perform in the wild.
Originally Posted by BO$$
Fedora is basically and experimental branch where they test new things in the wild for the first time, probably to improve next version of RHEL.
Apparently this can now similarly be said of non-LTS Ubuntu releases. There was a time when, perhaps while somewhat still being the case, it was less so. Perhaps this change has to do with the uptick in Ubuntu-focused technologies pushed by Canonical, as these will not be first tested by the Debian Project before they reach Ubuntu.
Originally Posted by alexThunder
How many pages would we have, if we left out any BO$$ post and related reply...?