Slashdot Mirror
Release Team Proposes Gnome 3.0 Plans
benuski writes "Today at GUADEC, the Gnome User and Developer European Conference, the gtk+ team announced their plans for gtk+ 3.0; immediately after, the Gnome release team announced their plans for Gnome 2.30 to be changed into Gnome 3.0. This would mean a release date a year and a half to a year in the future. Details are short at the moment, but the Gnome team seems to be following in KDE's footsteps, but hopefully will avoid the problems that plagued KDE 4.0's release."
Just re-name 2.2 to 3.0 and you've released ahead of schedule!
It gets 40 rods to the hogshead and that's how I likes it.
Can Gnome 3.0 allow programs to render to the root window? Try running xplanet in gnome - you might catch a glimpse of something when you shut down. Try playing video on root with VLC - no uh uh. There are hacks to get screen savers and things to run on the background. This seems to be a fundamental design "feature" of gnome - the kind of thing you'd want to change in a major version bump. Or are they calling it 3.0 because 2.30 sounds too much like some really old software being patched over and over?
In other words, at this stage this is about the development team, not about the technical issues.
Gwow, this is Great Gnews! Let's Ghope they are Gstill Going to Geep Gusing the Gletter "G".
A kbit klike kthe kpeople kthen ksince kthey kdo kthis kfar ktoo koften. kmuch kmore koften kthan kthe kGNOME kpeople
The link leads to a tersely worded page which captures the entire essence of the plans for GTK+3.0 :) which in turn leads to another blog with a color scheme that threatens my corneal legerdemain.
"but hopefully will avoid the problems that plagued KDE 4.0's release."
instead they're gonna have all sorts of their own problems. it happened before, it'll happen again.
all major projects have this kind of stuff when major releases come out the door. examples ?
MacOS X 10.0
Windows Vista
Gnome 2.0
Netscape 4.0
.
.
.
maybe it'll be a set of completely diferent problems. but they'll be there. murphy is unforgiven.
What ? Me, worry ?
Luckily for Gnome, when 3.0 ships missing a lot of features, nobody will notice.
I don't want to upgrade to Web 3.0 yet.
Run and catch, run and catch, the lamb is caught in the blackberry patch.
KDE 4 is clearly the most future-proof desktop environment out there.
In terms of graphic capabilities, it can natively suppport every feature available on OSX and in Vista, besides a few new features that are unique to KDE 4. In theory, it would be possible to create a desktop that looks-and-feels EXACTLY like OSX or Vista.
However, the best features are not those, but rather the platform independence with native API support. This means that, unlike JAVA, you can create one piece of software that compiles in Linux, OSX and Windows, using the OS-specific APIs. So, the same software compiled in OSX and in Windows look completely different and they didn't have a single line of code changed. The platform independence is not available for everything... for now, you can only compile things like Openoffice. However, the multimedia API, as well as other APIs are being developed.
The other thing great about KDE4 is that it is done with SVG instead of bitmaps. This means that scaling to very small devices like smartphones is quite simple to achieve.
Fear is the mind-killer.
I know people will think I'm crazy, but I have a vision for kGnome.
QT 4 actually has a Clearlooks engine designed to look like Gnome. Dolphin can be configured to operate largely like Natilus (except it works better these days).
If QT 4 actully really does use less memory and runs faster, why not do a test and port a small Gnome app or two over to QT 4?
The app can run with the QT 4 Clearlooks engine, and look largely like Gnome apps, except they can take advantage of many of the KDE features like Phonon, Solid, Sonnet, etc.
As for the people who prefer C to C++, aren't there language bindings for both for QT and GTK?
I'd love to see just a few small apps as a proof of concept. It could demonstrate the feasibility of a Gnome desktop built upon QT, especially considering the annoucement of Gnome 3, and the decision to break API.
If you're going to build anew, shouldn't this concept at least be considered for a moment? Both projects can have their seperate apps, desktops, defaults, window decorations, features, etc. But more common libraries and toolkits are a win for everyone.
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
I see your $0.02 and raise you a nickel.
My problem with KDE 4 is that I can't drag a box over several desktop to select multiple desktop icons. That drives me nuts!
My problem with Gnome is the fact that I can't adjust the screen saver properties without some ugly hack.
I know, these are minor issues, but annoying nonetheless. And your post was probably the nickel's worth anyway.
There is no "I disagree" mod for a reason. Flamebait, Troll, and Overrated are not substitutes.
I made the folly of installing KDE-4 on my mom's new computer (she had KDE-3.5.x before). There were no "problems". There was a total disaster.
The amount of features available in KDE-3 for years, that did not make it into KDE-4 is staggering... Add bugs to that.
And I was not entirely unprepared — I knew better, than to try KDE-4.0, when it came out with the enormous (and Google-sponsored) hoopla. I waited for 4.0.2... You can't even move widgets around on your task-bar yet — that's "scheduled" for version 4.1!
The all-new "plasma"-desktop can't show you the contents of files in ~/Desktop/ — that's still "in the works". Showing the list of files themselves is buggy — every time you login, a new set of icons (one for each of your files) is added to the desktop.
And to think, that I was getting impatient with FreeBSD KDE-team for not upgrading the KDE-ports! These guys were simply protecting me, but no, I wouldn't listen... I installed the much tauted Kubuntu and paid the price (don't even get me started on Ubuntu itself)...
In Soviet Washington the swamp drains you.
On top of that you have Aaron Segio now suggesting that users should have less control over configuration, fewer choices, and saying that end users are dumb. He also has suggested repeatedly lately that if you're not a coder, then you can't comment on UI issues.
Can you prove those 2 statements? Can you provide links to statements where he says that?
From my use of KDE 4.1, I, a user, have the exact same configuration menu in konqueror that I used to have, and I now have dolphin, with simpler configuration, that has been added which I can use standalone, or along konqueror or not.
As a user, it seems I now have more choice.
Plasmoid seems a little raw right now, but I have the feeling they are the equivalent of firefox extensions.
Basically, they are putting the desktop in the hand of the users. You will have extension, sorry, plasmoid, whith little or no configuration, and some some with heavy configuration and you will just choose and build your own personnal desktop. Just like firefox with its extensions.
So your comment about them dumbing down the desktop or removing it from the users hand is pretty much out of the picture, it's quite the opposite.
As for aseigo, I follow his blog and I can't remember him saying users can't comment on UI issues. If you'd give links to that than I might find your comment informative, right now, it seems mostly flamebait.
(My bet is that he said that as long as the underlying technology is not ready, the discussion about with or without 'insert your preferred desktop item or usability issue' are irrelevant.)
Community software should mean that people can easily post bug reports and get issues like these addressed.
Open a bug for each issue and hopefully they will be addressed.
I think it is beneficial to the entire community when people report these things.
The problem is that these don't appear to be bugs, but design choices. I believe that the gnome developers intentionally removed the option to configure each of the different screen savers and that the KDE dev's set up their horrid desktop icon system by design.
What's to file?
There is no "I disagree" mod for a reason. Flamebait, Troll, and Overrated are not substitutes.
GNOME HCI guidelines are one of the best I know of. Following the HCI leads to surprisingly good physical and mental health. 1) Navigating the GNOME dialog box with just the keyboard provides a rejuvenating and rigorous finger and mental exercise at the same time. 2) The font choices make pupil dilation effortless 3) The occlusion of "OK/Cancel" in elongated dialog boxes make accepting/rejecting dialog boxes into a fun hideAndSeek activity.
this is one reason why I continue to use gnome or xfce instead of the new KDE. Of all things they removed one feature most important to me:
the ability to change tabs in konsole by pressing alt-# (ie, alt-1 = go to tab 1, alt-2 to tab 2 etc.)
I asked in the #kde-devel channel if it was removed intentionally or just hadn't been re-added. Aaron's first response was to claim I must not use a terminal much (I'm a systems admin and programmer, I spend nearly all day in a terminal.) He then said that terminal programs should bind as few keys as possible because terminal programs have already assumed nearly all possibly combinations.
I offered a patch that would re-insert them as an option -- not enabled by default but there for people that decided they wanted to set it. It was turned down.
Fuck it all, KDE is going the same way GNOME did. I'll stick with vim, mutt, and move back to freaking wmaker or fvwm if it's the only way to have a system that doesn't treat me like I'm five years old.
Community software should mean that people can easily post bug reports and get issues like these addressed.
Open a bug for each issue and hopefully they will be addressed.
I think it is beneficial to the entire community when people report these things.
Here is the GNome developer response to the screensaver thingie:
Comment #1 from William Jon McCann (gnome-screensaver developer, points: 22)
2005-09-19 13:32 UTC [reply]
I don't have any plans to support this. My view is that any screensaver theme
that requires configuration is inherently broken.
Is developer arrogance a bug or a feature?
There is no "I disagree" mod for a reason. Flamebait, Troll, and Overrated are not substitutes.
What I would really like to see from the GNOME team is a pledge to keep the framework free of unencumbered technology. Specifically, this means we need them to promise that both the framework itself, and its core applications, will not be built with .NET (Mono).
Miguel de Icaza may enjoy appeasing Microsoft, but most of the Free World does not.
Tired of FB/Google censorship? Visit UNCENSORED!
Except that the Gnome apps aren't built on top of gtk, but on top of Gnome libs. And porting the Gnome libs to QT4 is what would be the pain.
I have yet to figure out what the Apple fascination with the prefix i- on absolutely everything is supposed to signify?
The only person that matters to Steve, course.
If that isn't enough, Trolltech will also provide QGtkStyle which will draw using native GTK widgets in the same way it uses native Cocoa/Carbon on OS X to make all Qt/KDE 4 applications have a Gnome look and feel, including things like the order of the dialog buttons.
My other account has a 3-digit UID.
Here is the GNome developer response to the screensaver thingie:
Is this a troll or do you suffer from short attention span? This was his first comment, but the discussion on bugzilla was very long, and further down he identified technical issues that prevent this from being done sanely atm, wrote an FAQ on the matter, asked for help from those who see this feature, and so on. Anyone interest in the issue is well-advised not to rely on the parent but read the discussion themselves.
"When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
I hear this kind of complaint all the time from linux peeps. They want two very different, conflicting things to happen at once. First of all, they want linux to evolve while maintaining all the flexibility that it is known for, while also wanting so desperately for each year to be the "year of the Linux desktop." This is an either/or situation. Gnome and KDE are both aiming to be user-friendly desktops, and therefore shouldn't be criticized because they don't meet the productivity needs of a sysadmin. Like you said, vim, mutt, and wmaker are still around and kickin'.
Similes are like metaphors
> I asked in the #kde-devel channel if it
> was removed intentionally or just hadn't
> been re-added.
It just hasn't been re-implemented.
You should have been pointed at me rather than Aaron. Terminal related queries will reach me if they are sent to konsole-devel@kde.org , robertknight on #kde-devel or be filed as bugs against Konsole at http://bugs.kde.org/ . Your patch hasn't crossed my path yet and I cannot comment on it until I see it.
Also a recent comment of his was that non-coders on the whole shouldn't be allowed to comment on design issues.
Well this is not going to make it feel any better but those who do not have experience coding often do not understand why their proposed design change does not or cannot work. Not always, but if you're good enough to design it you're typically good enough to code it. Code is just transferring a design into a language syntax. Designing it in the first place to work correctly (or not... ;) is hard.
He also repeatedly said that if you don't read the code, you can't understand the UI. That itself is a problem.
Again, this is probably actually true. Of course you can *see* the UI and point out what sucks about it but sometimes a "trivial" UI change involves a large code change. This is easier to see if you understand how the UI is actually formed from the code in question (i.e. Containments, Applets, Activities, etc. in Plasma-land).
Frankly, end users should be able to pick things up and learn them intuitively. Suggesting that if you don't read the source code, you can't understand the project means there is a serious usability issue.
Sure, but there's also the type of user (and I don't know if this is true in the case you're talking about but bear with me) that does something to the effect of:
User: Hey I noticed this is going on and it sucks, fix it!
Dev: One of:
User: Why don't you just do something like integrate the frobnitz?
Dev: Because it doesn't work like that.
User: No seriously, just integrate here and you're done.
Dev: No you don't get it. The code does not work like that. It cannot work because of reason foo
Now most bug reports we get are good reports and if the dev is actually here to work on it even get resolved to everyone's satisfaction. But we do get reports like these and when it turns into a pissing match between the user and the developer politeness is usually the first thing to go out the window. I've seen Aaron be ganged up on by multiple users in this fashion and it's really disheartening to see as a developer.
Hey, sometimes the developer is even wrong and it can be implemented somehow but that typically happens with a patch (oh, maybe it does work...), which you're not going to get from the same developers by pissing in his Cheerios and acting like a jerk. And in the end (at least in KDE) those who actually do the work get to decide so if Aaron is holding off on changing something because no one has presented a satisfactory technical solution (i.e. no "evil hacks" for bug fixes) then that's how it'll be.