At linux.com, I just read an article about the new GNOME 2.18 release (although 2.18.1 was just released). the line that caught my attention is the following:
“[..] I’m underwhelmed at the changes between 2.16 and 2.18. It’s a solid release, but it doesn’t move the ball forward very far in terms of improvements, new applications, or new features.“
I’ll be hated for saying this, and am partially responsible for this (I’m spending only limited time at gnome-media at this time) myself. But I felt something similar when I saw the GNOME 2.18 release announcement. There’s simple things that make me feel this way. Compare, for example, the 2.18 and the very detailed and excellent 2.14 release notes. Of course, all honour to those who do the actual work, but somehow the 2.14 notes look very professional and like they were done with a lot of love. Compared to that, the 2.18 release notes look like they were pulled together at the last minute. Not very impressive, not very impressed.
But there’s more, it’s not just this. Compare the still very sketchy developer pages for GNOME (I know, I know, people are working to migrate to a new web service system, but the thing is: it’s not there yet) and compare that to KDE’s plasma, phonon, decibel or solid pages (and for fun, compare those to some comparable GNOME technologies: GStreamer, Telepathy or HAL). There’s a multitude of differences. The KDE pages are targetted at both developers and users. The GNOME (basically FDO) pages are only targetted at developers. They lack information for the user on what it is. More importantly, they don’t associate directly with GNOME. Unfortunately, GNOME doesn’t associate with those projects either, even though all of them have large backing by GNOME developers and community.
Lastly, apart from the obvious pimping of those very cool technologies, those technologies should be embraced also. If GStreamer is the one true love, then make it work for Totem (and ditch Xine). The Firefox plug-in, for example, should work with playlists, which is what every internet site will stream. While I’m at it, please know that GStreamer can still not play DVD menus, shame on you, go fix it instead of make up excuses (or just rip it from Xine and release it under the GPL in a separate module). GConf, Evolution and the panel should use DBUS (work is being done on GConf [see Jeff earlier today] and Evolution, but none of this is upstream yet…). Let’s integrate Telepathy, add Gossip (even if it only does Jabber and GPhone). There’s some very obvious stuff out there which basically already exists (and it’s far more than the examples I’ve mentioned up here, e.g. Novell’s new start menu, GnomeScan, etc.), it only has to be brought back upstream. GNOME as a whole would profit greatly, those projects would attract more developers making them (hopefully) develop quicker (releaseintegrate early & often) and reviewers would be a whole lot more happy.
In addition to all of the above, I’d love to see exciting new experimental projects such as Gimmie enhance my experience (and I’m as excited when Mirco puts new screencasts online), but some of that is probably further away than “the next release”.
One word : Thanks !
I agree with some parts.
The most one will be ‘stagnation’ situation in this 2.18 release. In fact the best part of the desktop is being done by small individual projects where gnome as a whole should be doing the most part of the job.
As I see it, there were a lot of developers and engineers developing the critical parts of the desktop to build gnome. Now most of them are done and nobody cares.
I will just take an example, the panel applet window list and his resizing problem. It’s there since … the beginning?
Personnal Message to the gnome developpers: _please_ do not reinvent a new desktop ala KDE4 (plasma, etC..) and ignoring current bugs or desktop flaws. Please fix the desktop, and make it rocks (and rock solid). The gnome desktop is an excellent base, we need to fix the most critical parts.
And that’s why I’m all with your point of view Ronald.
The answer of some of your GNOME requests is the new web. At the time it is changed, all those projects you talk about can be documented and I hope anybody can help with this.
I always liked GNOME simplicity but it’s website is disorganized. It should be homogeneous, keeping it’s simplicity and clarity.
I’m waiting to the new website and see what can be done with it. Please, take a look in a few months and post a comment like this again… I think then should be easy to make any page needed and keep all organized.
In the other hand we have the bugs of some essential applications for the desktop. New times should come with the stabilization of GStreamer and Telepathy, I expect that GNOME will be up to date with them.
I think that GNOME should make a way to manage bug reports and make easy the life of an user. When a bug is filed, a lot of needless messages are sent back to the reporter. It can make the user not send any more reports.
As for GStreamer / Telepathy, what it comes down to is that they are essentially vapourware. They’ve been around for many year now, both have been taken up commercially and there’s companies hacking on them. In that way, they are a success.
I want them to be unvapoured _for the larger community_ and become reality, but that has to be done by their respective developer communities. Showing some cool prototype that does something technically significant (e.g. playback Ogg/Theora or allow a one-on-one webcamchat with CVS versions and an unknown protocol) with no functional relevance (since outside the test environment, nobody is going to be able to use this) is cool, but not useful in the long run. Some of us want to watch Windows Media in the browser, and we want MSN / iChat webcamchats.
So let me state that clearly: some of the GNOME technologies need to be less vapouPRware and more real. No more excuses, just do it.
(I agree with the website, I hope it’ll become very cool when it’s on – but it has to be on!)
Actually all those KDE pages give me the impression that they’re spending more time on flashy websites than _writing actual code_. Last I looked at least they were mostly just shell sites saying “we’re going to write this cool new thing .. later”.
@Michael:
http://cia.vc/stats/project/kde
21 messages so far today, 279 messages yesterday
664 messages so far this week, 2562 messages last week
3595 messages so far this month, 9273 messages last month
319981 messages since the first one, 3.25 years ago, for an average of 5.34 minutes between messages
http://cia.vc/stats/project/gnome
15 messages so far today, 108 messages yesterday
322 messages so far this week, 1195 messages last week
1607 messages so far this month, 6208 messages last month
247920 messages since the first one, 3.25 years ago, for an average of 6.9 minutes between messages
Well I don’t feel that 2.18 was such a letdown but some of your points are very valid (and sad!), mostly the missing DVD support in gstreamer is a huge thing. Totem even uses gstreamer by default now and I really wonder why I have to use mplayer to watch DVDs now…
About the explanation of not having DVD menus on GStreamer, this is a patent issue:
http://thomas.apestaart.org/log/?s=dvd+menus
I was also uninpressed about the release notes, but it seems Vincent Untz and Lucas Rocha want to tackle this, with the Roadmap Gang.
http://live.gnome.org/RoadMap/Process
I want a pony
Totem’s browser plugin already supports playlists…
It’s true a lot of GNOME web pages are:
Empty!!
Outdated
Not informative for those not intimately involved in development.
One problem is a list of depends with many things given as optional but no indication of why you would / wouldn’t include them in the build.
This may or may not contribute to why GNOME doesn’t seem to have a clear direction in the way that KDE does. Admittedly in some ways GNOME is more ambitious but it’s not so focused.
GNOME has a personality crisis, it doesn’t want to dictate to people what to use but it can’t be integrated (enough) if it doesn’t. It want’s to be a desktop on top of ones OS of choice but desktop/user services reach deeply into (ARE!) the core OS.
Certainly it is good to be able to use your language of choice but a modern OS needs a well thought out, deep, wide reaching development environment which provides simple access to sensible default technologies such as dbus, gstreamer, EDS(Dbus!!!) Telepathy Etc.
Is Bonobo deprecated or not? Some seem to think it’s a boat anchor and others cling to it from affection/religeon but it’s hard to find clear information on the status and whether some things depend on it just because they always have. Is GOffice providing the application object code now? Is it enough or does GNOME need to bite the bullet and adopt an OO language for application code?
The GNOME hare might find the GNUStep tortoise finishes the race first. (This is not a plug for Obj-C adoption!)
I agree with your assessment of 2.18. I love gnome, but I’m finding it interesting that most of the interesting work going on “in” gnome isn’t in gnome at all…
Sorry to say, but the release notes were done at the last minute and suffered from having too many cooks in the kitchen.
“As for GStreamer / Telepathy, what it comes down to is that they are essentially vapourware.”
How are they vapourware? I regularly watch videos and listen to music using GStreamer (Totem, Rhythmbox), and IM using Telepathy (Gossip/Gabber/Butterfly). Sure, there are other things they could do too, but “room for expansion” is not the same thing as “vapourware.”