Masking Gnome 3

[Anon-E-moose]

Edit: whatever

[asturm]

The work has already been done re: gnome 2 ebuilds and packages.
There will be no new ones. No updates. I've already mentioned that
there will be none from gnome itself. By the very nature of the word
unmaintained that means no extra work for devs.

Have you ever heard of build and runtime dependencies? Unless you think that Gnome-2 emerges and runs all by itself, you can't be serious with the above. Are you actually a Gentoo user?

[eccerr0r]

The problem is that there are a lot of shared libraries for Gnome2. It is a LOT of work to make sure all the revs work each time a new rev comes out. So do we need to slot every single library for Gnome2 just to make sure we don't accidentally upgrade? In fact, it would have to be a "Gnome 2" slot instead of just a particular version... some software could use a slightly different version where Gnome may or may not be able to tolerate... And if a security hole is found in a particular version of a library that hits Gnome2? What if a libc version someday breaks Gnome 2? Not like that's likely but it's possible. Needs a maintainer and it's work.

In fact I worry about gcc issues, I just had udev/systemd complain about my gcc 4.5.4 being too old. While compile time issues are different than runtime... gcc should be stable?

It's a lot of work. I agree, I'm disappointed that Gnome2 had to go away, but it was inevitable with the Gnome devs dropping support - maintaining versions specially for it is a project on its own.

[Anon-E-moose]

Edit: whatever

[EvaSDK]

Guys, before going about how the Gnome team wants to push Gnome 3 the hard way and undermine any attempt to use Gnome 2, just please read https://wiki.gentoo.org/wiki/Gnome_Team_Policies

We did and still do our best effort to slot everything that can be slotted but that does not mean what some of you have written here.

We perfectly understand that you are not happy about our decision but we cannot realistically support Gnome 2 libraries with the constant stream of bug reports about it due to gcc, glib, gnutls, etc changes while none of us are still using it.

If anybody wants to keep Gnome 2, please step up and become a Gentoo developer or a Proxy maintainer or create an overlay like has been suggested already. If you cannot do that, I am sorry for you but you cannot force people into doing the work you do not want or cannot do.

[saellaven]

Guys, before going about how the Gnome team wants to push Gnome 3 the hard way and undermine any attempt to use Gnome 2, just please read https://wiki.gentoo.org/wiki/Gnome_Team_Policies

We did and still do our best effort to slot everything that can be slotted but that does not mean what some of you have written here.

You guys couldn't even bother to leave a list of what to block if people didn't want to upgrade to GNOME 3, leaving it up to us to take care of for you. You certainly didn't give your "best effort" to slot things to keep GNOME 2 and 3 separated and your fellow devs have admitted that they didn't do what could have been done and can't be bothered now.

That's to say nothing of the headaches caused by the GNOME team automatically upgrading people to GNOME 3 with systemd even if they didn't want it, the systemd fanbois harming the system while proclaiming technical superiority (when in reality it was the limitations caused by LP's arrogance that people should only use their systems the way he wants to). I can't blame the GNOME herd for what williamh and friends did with regards to intentionally withholding information from the Council, but the GNOME herd could have handled things MUCH better, particularly because you knew there were going to be people upset.

At a minimum, a list of blockers should have been crafted... in reality, slotting should have happened from the very beginning, even if you had every intention of dropping GNOME 2 sometime after GNOME 3 went stable. But it's never about the end user, it's always about what the devs want at any given time and the end users are, at best, an afterthought.

We perfectly understand that you are not happy about our decision but we cannot realistically support Gnome 2 libraries with the constant stream of bug reports about it due to gcc, glib, gnutls, etc changes while none of us are still using it.

Nobody is asking you to support GNOME 2, just to not actively hamper people that don't want GNOME 3. Let them use GNOME 2 out of an unsupported overlay if that's what they want. Like Anon, I abandoned GNOME entirely after GNOME 3 went stable, so it doesn't affect me... unlike you guys, we are just trying to help our fellow users rather than blow smoke.

If anybody wants to keep Gnome 2, please step up and become a Gentoo developer or a Proxy maintainer or create an overlay like has been suggested already. If you cannot do that, I am sorry for you but you cannot force people into doing the work you do not want or cannot do.

Again, too many (not all, but too many, and too many in key positions) are cliqueish and hostile to anyone that doesn't think just like them. I constantly see devs say that Gentoo as a project needs more help everywhere. Ask yourselves why you don't get it... it all starts with the attitude from the devs that devs are somehow superior to users and some devs are more special than everyone else.

[Anon-E-moose]

Edit: whatever

[TomWij]

Guys, before going about how the Gnome team wants to push Gnome 3 the hard way and undermine any attempt to use Gnome 2, just please read https://wiki.gentoo.org/wiki/Gnome_Team_Policies

We did and still do our best effort to slot everything that can be slotted but that does not mean what some of you have written here.

You guys couldn't even bother to leave a list of what to block if people didn't want to upgrade to GNOME 3, leaving it up to us to take care of for you.

This information can be obtained from upstream's dependency specifications.

You certainly didn't give your "best effort" to slot things to keep GNOME 2 and 3 separated and your fellow devs have admitted that they didn't do what could have been done and can't be bothered now.

There are limited resources to maintain GNOME2.

That's to say nothing of the headaches caused by the GNOME team automatically upgrading people to GNOME 3 with systemd even if they didn't want it,

They want it by default, as that's what happens when you upgrade your system; anything that deviates from that is up to the user to specify and maintain, given that the distribution has limited resources to accomplish this for them.

the systemd fanbois harming the system while proclaiming technical superiority (when in reality it was the limitations caused by LP's arrogance that people should only use their systems the way he wants to).

That is an implication of running GNOME 3; the only way to change this implication, is to take action upstream.

I can't blame the GNOME herd for what williamh and friends did with regards to intentionally withholding information from the Council,

Which information?

but the GNOME herd could have handled things MUCH better, particularly because you knew there were going to be people upset.

When expectations are high, but resources are low; people could upset themselves, which is motivation for them to make it happen by stepping up and maintaining it. This is how a lot of us became developers; other reasons to become one, are for example giving back to the community for the great things they gave you.

At a minimum, a list of blockers should have been crafted...

Given its removal from the Portage tree, as well as the lack of resources; that's nothing more than a waste of time, as it misleads people to use something which will get removed. This could instead be spend on making users happy by spending it on what will stay and become more relevant in the future; in this case I'm not talking about GNOME 3 / systemd in specific, but also about minimal efforts with helping MATE enter the Portage tree (as a small bit of collaboration with the GNOME team might be necessary).

in reality, slotting should have happened from the very beginning, even if you had every intention of dropping GNOME 2 sometime after GNOME 3 went stable.

Slotting is meant for compatibility purposes; in the case of GNOME 2 or GNOME 3 it has no benefit afaik, unless one wants to run both DEs which I think it would be a rather unique goal.

But it's never about the end user, it's always about what the devs want at any given time and the end users are, at best, an afterthought.

We work toward the reason that the end users have chosen the distribution for, under the resources we have; maintaining GNOME 2 is outside of that border, which pertains to a small number of end users with special needs. There are valuable alternatives available instead of using an unmaintained DE; maintaining an unmaintained DE is something that falls outside of the scope of Gentoo, of which the poster from the about page gives you an idea...

We perfectly understand that you are not happy about our decision but we cannot realistically support Gnome 2 libraries with the constant stream of bug reports about it due to gcc, glib, gnutls, etc changes while none of us are still using it.

Nobody is asking you to support GNOME 2, just to not actively hamper people that don't want GNOME 3. Let them use GNOME 2 out of an unsupported overlay if that's what they want. Like Anon, I abandoned GNOME entirely after GNOME 3 went stable, so it doesn't affect me... unlike you guys, we are just trying to help our fellow users rather than blow smoke.

Users can use an unsupported overlay and/or switch to MATE; we're helping them by making these suggestions, where is the fire?

If anybody wants to keep Gnome 2, please step up and become a Gentoo developer or a Proxy maintainer or create an overlay like has been suggested already. If you cannot do that, I am sorry for you but you cannot force people into doing the work you do not want or cannot do.

Again, too many (not all, but too many, and too many in key positions) are cliqueish and hostile to anyone that doesn't think just like them. I constantly see devs say that Gentoo as a project needs more help everywhere. Ask yourselves why you don't get it... it all starts with the attitude from the devs that devs are somehow superior to users and some devs are more special than everyone else.

Where do you get this impression?

[asturm]

Again, ...

You don't need to be a dev at all to start and publish your own overlay.

[EvaSDK]

saellaven, you obvsiously do not understand what slots are for.

It is not meant to allow two programs to be installed at once, like evolution-2 and evolution-3 simply because this is not possible without heavily patching software in question (and by that I mean 100kb patches at the minimum between the various helpers, the dbus interface, the includes, pkg-configs, etc). It is meant to allow parallel installation of non-conflicting libs (maybe over-simplification but that's mostly it).

Gnome team did slot everything that was possible. We did warn users in advance when Gnome 3 would become stable with a news item and a wiki explaining everything that was about to happen.

Indeed, we did not provide a package.mask, but that is not because of lazyness or because we love systemd (almost all of us don't), it is because such mask has no future. Packages come and go from the tree according to various conditions, but most notably here because they cease to work with newer compiler or underlying library. Yes this is still not the case for Gnome 2, but that is because you are ignoring the 3 years or continuous effort to keep it working despite those library changes until we were satisfied with Gnome 3 stability and features.

As for the cliqueish and hostile attitude, I have no idea who you think has this attitude, but please be in contact with comrel if you cannot have a reasonable discussion with any dev as it would certainly not be normal.

[saellaven]

Again, too many (not all, but too many, and too many in key positions) are cliqueish and hostile to anyone that doesn't think just like them. I constantly see devs say that Gentoo as a project needs more help everywhere. Ask yourselves why you don't get it... it all starts with the attitude from the devs that devs are somehow superior to users and some devs are more special than everyone else.

Where do you get this impression?

I didn't read the rest of your post, precisely because YOU are one of the worst offenders when it comes to this. You've repeatedly done it not only to me but at least a half dozen other people on the forums alone. I've asked you not to reply to my posts after you played games jerking me around for your amusement in the past, you continued to antagonize me afterward anyway, even trying to goad me into playing your games, and an admin asked you to adhere to my wishes to not reply to me since nothing good was going to come of it. You even have your own thread in the forums feedback section because of your desire to harass people here.

So frankly, to keep this polite as I can, look in the mirror.

[Anon-E-moose]

Edit: whatever

[TomWij]

Again, too many (not all, but too many, and too many in key positions) are cliqueish and hostile to anyone that doesn't think just like them. I constantly see devs say that Gentoo as a project needs more help everywhere. Ask yourselves why you don't get it... it all starts with the attitude from the devs that devs are somehow superior to users and some devs are more special than everyone else.

Where do you get this impression?

I didn't read the rest of your post, precisely because YOU are one of the worst offenders when it comes to this. You've repeatedly done it not only to me but at least a half dozen other people on the forums alone.

That appears to be a misunderstanding; it is natural that when people have a certain vision, that people trying to understand that vision and help are misunderstood because they come over as questioning it.

I've asked you not to reply to my posts after you played games jerking me around for your amusement in the past, you continued to antagonize me afterward anyway, even trying to goad me into playing your games, and an admin asked you to adhere to my wishes to not reply to me since nothing good was going to come of it.

Sorry, but categorization of people is what I try to avoid at all costs, and my actions there were to help you in an useful manner; if you have previously asked so, then I'm sorry to let you know that I have forgotten that as this is unlisted.

You even have your own thread in the forums feedback section because of your desire to harass people here.

Or rather, to help people here; given my high participation on the forums, it's natural for there to be a small few individuals to perceive that differently.

So frankly, to keep this polite as I can, look in the mirror.

(Oh, what a nice view... )

[EvaSDK]

Packages come and go from the tree according to various conditions, but most notably here because they cease to work with newer compiler or underlying library. Yes this is still not the case for Gnome 2, but that is because you are ignoring the 3 years or continuous effort to keep it working despite those library changes

Then what is the harm in simply leaving it in portage until it is obvious that it won't work anymore.
By your own admission it works at this point in time (and thanks to you/others for keeping
it working even though I don't use it) so why not leave it. Put a large warning *no longer supported*
in the ebuilds. When/If someone posts a bug report against gnome2 pkgs, simply close it with "it's not supported".

It cannot be left alone, because nobody in the team can actually state that it works. It is a general feeling from flow of bug reports. It is roting at maximum speed due to no upstream and already gcc-4.8, latest glib and latest gnutls are expected to break some builds. I have not actively verified it but it happened sufficiently enough times in the 8 years I have been a maintainer to guarantee you it will happen and sooner rather than later. But of course, those packages can be masked too.

Leaving those packages to maintainer-needed would be a solution too, but see my next point.

Instead what we end users see, is what amounts to a flip of the middle finger
and told to either support it yourself or switch to something else, rather than
"when it quits working completely we'll remove it". That is how it comes off
to many, whether intended that way or not.

I understand this point of view, however, if we provided an overlay, it would give a false sense that Gnome 2 will keep on being supported and that it will receive patches. Of course, it may happen once in a while but leaving that to people that do want to keep using it has better chance to gather a community of active people that something put up by current Gentoo developers and left abandoned.

[SamuliSuominen]

SLOTing WAS possible and still is

As explained, untrue, the binaries and co. are named the same. Futhermore, Gnome 2.x is broken with many Gnome 3.x and other ~arch libraries.
Of course everything is possible to SLOT if you do massive renaming, but then you have also the reverse dependencies that call those binaries with their original (upstream) name.
Not to mention nobody in Gentoo's GNOME team is using Gnome 2.x anymore -- nearly impossible to maintain a desktop without more or less active maintainers.
MATE seems to be the clean solution to all these problems, and I'm still willing to help to package it, if someone is serious about it.

And what about sep. /usr? It has nothing to do with udev, or eudev. What udev-init-scripts patch? What about diminishing eudev? That I've told there is no point in using it, unless you want the old-style rule-generator back. Does that upset you? Why, because it's true?

[Tony0945]

MATE seems to be the clean solution to all these problems, and I'm still willing to help to package it, if someone is serious about it.

I'm serious about it. The package is in overlay. I'm using it on one box and about to transition to three others. I'm sure I don't qualify as a gentoo dev, but I'm able to test, but only on four generations of AMD64. I still have a k6 box in the basement (nostalgia, my grandson and I built it from discarded parts and eBay when I was out of work), but I doubt if anyone is interested in that. I may be acquiring a Pentium II laptop that I will be doing a 32 bit install on. The four AMD64's (k8 through fam10) are the serious boxes.

[eccerr0r]

What DE do most Gentoo devs use, to minimize the chances of getting version bumped again? :D

[saellaven]

And what about sep. /usr? It has nothing to do with udev, or eudev. What udev-init-scripts patch? What about diminishing eudev? That I've told there is no point in using it, unless you want the old-style rule-generator back. Does that upset you? Why, because it's true?

TODAY...

and when Lennart decides that udev will not be usable outside of systemd, as he's already stated he wants to do?

Rather than be silent, you criticize the very people, your fellow gentoo devs I might add, that sought out to protect us from that scenario, because it competes with your version of the official package. Again, ego driven. Your faith in systemd's future doesn't make it a fact, it is opinion, not truth.


As far as separate /usr goes, the patches have been here for 2.5 years. williamh knew about the patches, refused to add them to openrc to make openrc more robust and then deliberately withheld disclosure of them from the Council, as /usr was entirely a political decision based on the Lennart intended technical limitations of systemd which I can only conclude he favors.

But we're getting off topic here... in short, arrogance and ego is limiting Gentoo development and robustness and the whole thing with GNOME2 (and systemd, your anti-eudev hate, etc) is just another symptom of it.

[weedy]

In case this helps others. And can your guys stop whining to devs?
systemd is here to stay unfortunately, and we just have to deal with it (or in my case ignore it as best as possible).

Code: Select all

layman -o http://suigintou.weedy.ca/trac/gentoo-overlay/export/HEAD/repository.xml -a fuckyeah-overlay

[steveL]

What steveL is saying simply isn't true. It's not the Gentoo packagers that have "poisoned" some packages but rather the original upstream of them, such as dev-libs/glib, have changed and since Gnome 2.x doesn't have a upstream anymore, they haven't been patched to be compatible with the newer libraries.

..It was GNOME upstream who decided to call binaries/libraries/headers and so forth the same, like eg. gnome-terminal is still gnome-terminal and not gnome3-terminal so that
SLOTting was impossible. Those that were possible to SLOT, have been SLOTed.

I'm calling "bullshit" on this. SLOT conflicts are no excuse for deliberately keeping the 2.0 SLOT for 3.0 ebuilds. The conflicts go in the old ebuilds, as you move to the next generation.

The point, as you well know, is that the GNOME herd could have simply moved to SLOT 3.0 for the new ebuilds, just like KDE or any other setup. The only reason not to do so is to force people to upgrade, and to remove the the option of masking by SLOT.

If it's bit-rotting, so what? Drop it from the tree, in normal time, as other ebuilds are done. The point is you guys went against the normal methodology, and for all your posturing that simply broke the tree, and went against the proclaimed intent of allowing people to use a systemd-gnome profile.

It also completely screwed anyone else from doing any further work on the GNOME2 ebuilds, if they had wanted to, such as making it work with a later glib.

I should have guessed, you are in for the complaining, but not for the contributing part. No offense, but the distribution doesn't lose much with users like you. Don't slam the door on the way out.

Yeah that's the way to encourage people. Good show. How about you take your own advice? I'm sure exherbo will be glad for you to stop playing the charade, as will I for one. You've had a good run, and lasted longer than any of them thought you would.

No, I'm not being serious: but it's not nice, is it? And it carries an awful lot more weight when attached to a Gentoo badge.

As for you presenting eudev as an option.. LMAO. You were the one bad-mouthing its developers instead of helping them; you appear to have a knack for denigrating people instead of responding to the underlying problem, and taking the negative fork in the road instead of choosing to turn things into a positive. In this instance, you could have recruited a hard-worker, instead of making him feel worse. With eudev you could have fostered a new collaboration with other devs, instead of bad-mouthing them. You clearly have a lot of maturing to do as a person and developer both.

You're not going to do it if no-one tells you where you go wrong, or if you refuse to listen to any critical analysis. But good luck with it (the key is being able to critically evaluate your own behaviour, not others'.) When you get tired of being a bully, you'll look back at this sort of behaviour with profound regret. Hint: next time someone you know well is putting you down, instead of trying to think of clever rejoinders, try to ask yourself why you even hang out with someone who treats you like that. Perhaps it's a deep-rooted psychological issue, and not something technical at all: your humanity is far more complex than a computer.

[Anon-E-moose]

Edit: whatever

[SamuliSuominen]

What steveL is saying simply isn't true. It's not the Gentoo packagers that have "poisoned" some packages but rather the original upstream of them, such as dev-libs/glib, have changed and since Gnome 2.x doesn't have a upstream anymore, they haven't been patched to be compatible with the newer libraries.

..It was GNOME upstream who decided to call binaries/libraries/headers and so forth the same, like eg. gnome-terminal is still gnome-terminal and not gnome3-terminal so that
SLOTting was impossible. Those that were possible to SLOT, have been SLOTed.

I'm calling "bullshit" on this. SLOT conflicts are no excuse for deliberately keeping the 2.0 SLOT for 3.0 ebuilds. The conflicts go in the old ebuilds, as you move to the next generation.

The point, as you well know, is that the GNOME herd could have simply moved to SLOT 3.0 for the new ebuilds, just like KDE or any other setup. The only reason not to do so is to force people to upgrade, and to remove the the option of masking by SLOT.

If it's bit-rotting, so what? Drop it from the tree, in normal time, as other ebuilds are done. The point is you guys went against the normal methodology, and for all your posturing that simply broke the tree, and went against the proclaimed intent of allowing people to use a systemd-gnome profile.

You don't seem to understand how SLOTs work. Say, if package gnome-base/random-2.0 was SLOT="2.0" and this gnome-base/random-2.0 installed binary 'random', and gnome-base/random-3.0 was released and it still installed binary called 'random', then the SLOT shouldn't be changed, or otherwise you'd be hitting file collision.
The "2.0" is meaningless random number, it has no meaning, it's the same thing as "0" or "whatever". What's important is that it stays the same if there isn't anything that can be co-installed/parallel-installed for Portage to upgrade normally without anykind of blockers.
KDE is/was different, as KDE3 and KDE4 could entirely be parallel installed. Much better comparison would be MATE vs. GNOME3, with MATE renaming binaries and co. allowing the parallel installation.

As for you presenting eudev as an option.. LMAO. You were the one bad-mouthing its developers instead of helping them; you appear to have a knack for denigrating people instead of responding to the underlying problem, and taking the negative fork in the road instead of choosing to turn things into a positive. In this instance, you could have recruited a hard-worker, instead of making him feel worse. With eudev you could have fostered a new collaboration with other devs, instead of bad-mouthing them. You clearly have a lot of maturing to do as a person and developer both.

I don't want to bad mouth anyone, but you are wrong, again. It was the otherway around, people refused to help with sys-fs/udev and insisted in creating an entire fork. Since, to this day, the only real visible change
is the rule-generator, it's like fixing a minor bug with a sledge hammer.
It's still completely possible to package rule-generator separately as an 3rd party udev helper that could be used with sys-fs/udev, and heck, even with sys-apps/systemd, but instead it's now, unfortunately, bundled with
sys-fs/eudev.

Anyway, I apologize, if you have been offended by anything I've said. I'm known for being blunt when speaking the truth. I'm still open to helping anyone intrested in changing status-quo by contributing.

[GFCCAE6xF]

I gotta admit I get a little confused why you guys give ssuominen so much shit when all he does is point out what the eudev guys already admitted about there project.

That is of course unless since the FOSDEM talk the eudev guys have learnt udev inside-out suddenly and the technical reasons and FUD they acknowledged werent even real have suddenly materialised out of thin air?

[Anon-E-moose]

Edit: whatever

[SamuliSuominen]

I gotta admit I get a little confused why you guys give ssuominen so much shit when all he does is point out what the eudev guys already admitted about there project.

Propably because I'm the only one answering. Likely they have read up some anti-systemd propaganda from some blog post and assumed it to be true, and now they are angry because it turned out to be bs.

For The Record, I've just packaged udev-209.ebuild to Portage out from systemd-209.tar.xz *without anykind of issues*. I only wish someone would help me with porting the rule_generator to separate
package, and ultimately, ebuild.