Could GNOME 3 suffer the same problems as KDE 4.0/4.1 did?

[Mazza558]

If you look back at the release of KDE 4.0, which was intended for developers but somehow got confused as a working desktop, the transition was a shambles as distros hastily switched, despite the new KDE 4 still missing functionality. It's only now, approaching the release of 4.2, that it's usable as a productive DE.

From what I've read, GNOME 3 will follow a similarly radical departure from what users know and love. What's to stop users treating GNOME 3 as a ready-to-use release and causing the same problems?

[smartboyathome]

You read my mind. That is exactly what I said in another thread. XD

[SunnyRabbiera]

Well there is more time to prepare though :D

[smartboyathome]

Well there is more time to prepare though :D

Yeah, but I am guessing many will become mad anyway, since GNOME 3.0 wouldn't be perfect upon release, just like KDE4. ;)

[RichardLinx]

I just posted an answer in another thread that would have fitted right in here as well. :) I'll say again that I don't think GNOME 3.0 will be anywhere near as disasterous as the KDE4 release. I'm sure they'll know better to release GNOME 3.0 Beta before they decide to risk calling it a finished product.

EDIT: "anywhere near as disasterous" - What I mean is I don't think It will be a disaster at all.

[Giant Speck]

Maybe they will learn from KDE's mistakes? I mean, we all saw what happened to KDE and why it happened. GNOME developers will just have to ensure that the same screw-up won't happen a second time.

[Mazza558]

Maybe they will learn from KDE's mistakes? I mean, we all saw what happened to KDE and why it happened. GNOME developers will just have to ensure that the same screw-up won't happen a second time.

Surely KDE 3 -> 4 would have learnt from KDE 2 -> 3?

[RiceMonster]

I think no matter what they do, people are going to get mad because they didn't really want anything to change. Happens all the time.

[RichardLinx]

Maybe they'll give GNOME 3 a new look but keep a "classic view" for users that don't like the new interface. Kind of like how Windows XP had a classic view which made It look similar to Windows 2000.

[castrojo]

From what I've read, GNOME 3 will follow a similarly radical departure from what users know and love. What's to stop users treating GNOME 3 as a ready-to-use release and causing the same problems?

Where are you reading this?

[cardinals_fan]

Releasing something when it's actually done would be a novel concept.

[igknighted]

Releasing something when it's actually done would be a novel concept.

And would kill the release early, release often philosophy of open source software.

I thought the KDE4 release was handled (on the KDE end) absolutely perfectly. I put the blame for any "disaster" squarely on the shoulders of overzealous distributions that pushed out desktops that did not meet the needs of their users. If distributions make the same mistake as they did with KDE, it will be ugly. But if they (the distro's themselves) learn from their errors, it could go perfectly smooth.

Essentially, distro's who aim for cutting edge should ship Gnome 3 right away, but more conservative distro's shouldn't drink the kool-aid so quickly. The Fedora user in me was pleased with how soon KDE4 hit the repos, but the Ubuntu user in me was very disappointed with how soon kde4 was default. Users always drink the koolaid and will want kde4 or gnome3 or whatever package as soon as possible, but someone higher up in Ubuntu has to have the sense to not ship it (at least as default)

[cardinals_fan]

And would kill the release early, release often philosophy of open source software.

The "release early, release often" mantra refers to making development versions available immediately so that the community can get involved. In other words, the source code for development versions of GNOME 3 should be publicly available regardless of stability but the formal release shouldn't be held until it is finished. There is a difference between having a development release out there for the ambitious or daring to test and releasing your software when it isn't finished.

[igknighted]

The "release early, release often" mantra refers to making development versions available immediately so that the community can get involved. In other words, the source code for development versions of GNOME 3 should be publicly available regardless of stability but the formal release shouldn't be held until it is finished. There is a difference between having a development release out there for the ambitious or daring to test and releasing your software when it isn't finished.

I don't think the first release of a major re-write should ever be considered usable for an important component like a DE. We'll have to agree to disagree here.

[cardinals_fan]

I don't think the first release of a major re-write should ever be considered usable for an important component like a DE.
Why release it at all if it isn't usable?


We'll have to agree to disagree here.
I'm happy to :)

[igknighted]

Why release it at all if it isn't usable

Poor choice of words... I used kde 4.0 for a while and enjoyed it, despite its flaws. But it certainly wasn't "mainstream ready" for a distro like Ubuntu where users expect stable, feature-complete software.

[SunnyRabbiera]

well the transition from gnome 1 to 2 certainly was better, I did use linux just before gnome 2 came out.
Gnome 1 wasn't that great, gnom3e 2 was much better even when it was still new

[swoll1980]

If I remember correctly the monstrous kde 4 was released under the name RC1 or something to that effect. In my mind RC means that it is ready for release baring some kind of showstopper that no one noticed before. I don't think people were to blame for thinking this was a finished product. I have not tried it since then, but think I will give it a go later tonight. I hear it's very impressive now.

[RichardLinx]

If I remember correctly the monstrous kde 4 was released under the name RC1 or something to that effect. In my mind RC means that it is ready for release baring some kind of showstopper that no one noticed before. I don't think people were to blame for thinking this was a finished product. I have not tried it since then, but think I will give it a go later tonight. I hear it's very impressive now.

Your probably better of waiting a few more days, the current release date for 4.2 is January 27th I'm pretty sure.

[SunnyRabbiera]

Your probably better of waiting a few more days, the current release date for 4.2 is January 27th I'm pretty sure.

Pretty much, it goes out on tuesday.

[swoll1980]

Your probably better of waiting a few more days, the current release date for 4.2 is January 27th I'm pretty sure.

Will do thanks for the heads up.

[Skripka]

Pretty much, it goes out on tuesday.


HOPEFULLY-a MASSIVE bug regarding reading audio CDs has been fixed. I'm a KDE4 fan and love it-but this was an EPIC FAIL.

https://bugs.kde.org/show_bug.cgi?id=180014

EPIC FAIL and showstopper.


There's also a glaring problem with Konqueror and mouse gestures that needs fixed...or rather mouse gestures need implemented again period.

[SunnyRabbiera]

HOPEFULLY-a MASSIVE bug regarding reading audio CDs has been fixed. I'm a KDE4 fan and love it-but this was an EPIC FAIL.

https://bugs.kde.org/show_bug.cgi?id=180014

EPIC FAIL and showstopper.


There's also a glaring problem with Konqueror and mouse gestures that needs fixed...or rather mouse gestures need implemented again period.

Eh if its that recent you have another 6 month wait probably...

[Skripka]

Eh if its that recent you have another 6 month wait probably...

Considering how bad a fck-up it is, if they don't hold back 4.2 for a fix on it-the verbose locutions about 4.0s failings will sound like a holiday family dinner in comparison.

[MikeTheC]

Some thoughts on this...

While this should be true of most distros (there are exceptions to this, particularly in ones which are known and established as being more "hobby" or "experimental" in nature), it seems sadly that many have been willing to leap before they looked, and may possibly engender a similar reaction as that garnered in the 1998-2000 "Linux push" when, without a doubt, Linux was in no way ready for mainstream use nor the prime-time spotlight which it inhabited, and was quite rightly kicked out of.

That being said, Ubuntu (because of the reputation it enjoys as being a sort-of "premiere distro" amongst distros) needs to emulate more of Debian's cautious nature. Considering that a very large percentage (perhaps even most?) first-time Linux users are now "starting" with Ubuntu, Canonical has an obligation to the Linux community, since it is occupying a figurehead and leadership position, to ensure the user experience is smooth, solid and best-of-breed. The Linux community can hold them responsible -- at least in part -- for whatever satisfaction or dissatisfaction n00bs have, and quite rightly so.

I have no personal opinion, one way or another, on Gnome 3 or the other development work being done on it. However, I feel it is not just appropriate but critical that Mark does not allow Ubuntu to release something before it's truly ready, especially something with the capability to make or break the distro such as a D.E., and I hold him accountable for whatever mess has been allowed to occur in Kubuntu (again, as is right and proper).

Ubuntu has a reputation of being reasonably user-friendly, of being broadly compatible on diverse groups of hardware, and of presenting all of this in a very stable manner. Reputations are earned, and reputations can just as easily be destroyed. Care needs to be taken in regards to such things.

[etnlIcarus]

Correct me if I'm wrong but Gnome 3.0 isn't a complete rewrite, is it?

I don't expect Gnome 3.0 will be as broken as KDE 4.0 originally was as I'm reasonably certain Gnome 3.0 isn't intended to replace most of the DE's codebase, which is a large part of why KDE 4.0 was so broken and lacking in functionality. That said, I'm not expecting miracles since the last stable release of Gnome 2.0 I used was unreliable enough to send me packing to Xfce4.

[Paqman]

From this article (http://arstechnica.com/news.ars/post/20080714-gnome-3-0-officially-announced-and-explained.html):

Some have speculated that the GNOME 3.0 transition will be like the somewhat controversial development KDE 4, but such speculation isn't consistent with the details available about the release plan. Unlike KDE 4.0, which produced impressive innovation and accelerated development at the cost of user trust and overall desktop stability, the GNOME 3.0 plan is less ambitious, largely builds on the GNOME desktop environment's current strategy, and avoids significant user-visible changes or disruption to basic desktop usability.

[mips]

http://arstechnica.com/news.ars/post/20080714-gnome-3-0-officially-announced-and-explained.html


The new plan

All of the ennui and frustration eventually took a positive turn and culminated in the plan that was proposed by the GNOME release team at GUADEC. The GNOME development community will reach 3.0 but will do so without disturbing the users and without discarding the long-standing philosophy of incrementalism. There will be a GNOME 3.0, but it will not be what so many envisioned for ToPaZ.
"There are still lots of details to discuss but the important thing is that our proposed mindset for 3.0 is in place for discussion now. It involves a relatively smooth transition from 2.x to 3.x, a more focused and inclusive development process, long-term development cycles, and more," wrote GNOME contributor Lucas Rocha in a blog entry (http://blogs.gnome.org/lucasr/2008/07/10/gnome-30/).
http://media.arstechnica.com/news.media/gnome3.jpg
The GNOME 2.30 release, which will be about a year and a half from now based on GNOME's standard six-month release cycle, is what the developers have decided to call GNOME 3.0. During that time, the GTK+ toolkit will undergo its transformation and much of the desktop infrastructure that has been under heavy development will be more mature—this includes Clutter, Vala, PolicyKit, PackageKit, GVFS, D-Conf, and the GTK+ WebKit port.
Some have speculated that the GNOME 3.0 transition will be like the somewhat controversial development KDE 4, but such speculation isn't consistent with the details available about the release plan. Unlike KDE 4.0, which produced impressive innovation and accelerated development at the cost of user trust and overall desktop stability, the GNOME 3.0 plan is less ambitious, largely builds on the GNOME desktop environment's current strategy, and avoids significant user-visible changes or disruption to basic desktop usability.

[bruce89]

The only major sticking point over GNOME 3.0 is the fact its (quite likely) main shell (gnome-shell) depends on Mozilla's JavaScript interpretor, whereas the rest of GNOME is switching to WebKit.

Apart from that, it's just a case of removing depreciated functions in libraries and getting rid of other libraries completely.