“I noticed some tiny disturbance in the force before Christmas as Thom Holwerda posted two articles about what he felt was the sorry state of free desktops. Seems most people in the GNOME camp simply ignored the article as irrelevant, but Aaron Segio of Trolltech and KDE let it somewhat get to him. Personally I felt Thom kinda pointed out some troublesome points, but that his context and conclusion was wrong.”
Have you really got nothing better to do on old years eve than updating osnews?
PS: Im going to a party in 10 min, so dont use “you’ve got nothing to do either” as a lame argument
I’m staying home until after 12, and then I’ll leave (where to is really none of your business). And you just had to squeeze in your plans, I see? First time allowed out after 12 by mummy and daddy?
Other than that, it’s Top2000 time [1].
Happy new year y’all.
[1] http://cogscanthink.blogsome.com/2006/12/31/shallow-meaningless-swe…
Edited 2006-12-31 19:34
Better than out dodging stray bullets.
I dunno, if you dig the bullets out of the wall you might get some money for the lead value . Where “internet addiction” isn’t far from having its own specialists…
Before everyone starts critiquing each others plans let’s remember that we might live in different timezones and have plenty of time to post useless comments Like this one.
My comment on the article is that I’m sure Thom’s original post was more about being read and generating interest than actually reviewing what most people think about “The Linux Desktop”. I don’t mind that, we should have people talking even if some emotion needs to be generated to do it.
Personally I use KDE so forgive me for glossing over Gnome (I don’t know enough to comment on it). KDE4 will be great, and of course there isn’t much to see yet; it’s called infrastructure and having one is a good thing. Keep watching and my bet is that Thom will change his tune as soon as some public previews become available and the developers are happy to say “Yeah we’re nearly done”. I’m so grateful there are people working on KDE that know how important these early stages are and I thank them for having the motivation to work on the dull stuff before getting carried away with eye-candy.
Happy New Year Everyone.
“Before everyone starts critiquing each others plans let’s remember that we might live in different timezones and have plenty of time to post useless comments Like this one. ”
Well, it’s 23.24 here and dinner is over. What should I do until zero hour and before I go to party? Internet to the rescue.
Somebody is definitely in denial about the improvements the competition is making.
First of all he critized GNOME for not having a clear vision for GNOME 3. Well this is true, but that is mostly due to not having any clear ideas for something that would require a GNOME 3.
Right, so there is no clear vision for GNOME 3 because… they really don’t know what GNOME 3 should look like. How is that a refutation of Thom’s point?
And to be honest incremental improvements is what everyone is doing these days. Windows Vista, MacOSX and KDE4 don’t really contain anything earth shattering, they are basically increamental improvements over the predecessors.
Vista contains more improvements than have been made in GNOME since 2.x. Sure, the UI is still inferior to GNOME’s, but that’s more a problem of a lack of taste, not a lack of progress. OS X has improved more incrementally, but OS X also had a big head start to begin with. OS X is at the point where it can afford to make little incremental improvements in Leopard, while GNOME doesn’t have that luxury.
I installed Fedora Core 6 on my Macbook the other day, and I have to say, it really is very slick. But, its very clear there is major consolidation that needs to happen. Compiz is not well-integrated into GNOME yet. When you select “Enable Desktop Effects”, your window decorations suddenly change, and your window preferences no longer respond as expected (ie: double-click to maximize). I know why it happens, but its something that’d be completely non-obvious to a user that was unfamiliar with the stack. There is a nifty Expose-like when you hit the top-right corner, but there is no mention of the feature in any of the preferences dialogs. The fonts in Firefox suck goat nuts, because Firefox doesn’t use the same font-handling paths as GNOME, apparently. They look great in Epiphany, however, and its beyond me how Firefox became the default GNOME browser… Desktop search is better, and Beagle’s memory usage and startup time is improved greatly, but of course it’s still not standard yet.
That’s just from the user’s point of view. Things are a lot uglier if you look under the hood. Window resizing and movement is admirably smooth with Compiz, but its clear that no attempt is being made to synchronize window movement with front-buffer swaps. While the system is snappy on my dual-2GHz system, it’s clear that there are some inefficiencies, as rapidly moving a window will push both processors over 60%, while doing the same in OS X won’t push either over 10%. There is no real accelerated vector graphics architecture, though that’s not really something the GNOME folks can do anything about. There is no official high-level-language story, though it looks like Mono is poised to become the de-facto GNOME high-level-language, by virtue of having the right people behind it…
That said, there is no reason for any apocolyptic thoughts. GNOME is great, and its moving forward just fine (if somewhat slowly). I have no doubt that a couple of years from now, it’ll gain all the features and infrastructure that its missing today. It’ll be a little bit late to the party, but in the grand scheme of things, that’s not a big deal. Given that, there is really no need to downplay the progress other systems are making.
Right, so there is no clear vision for GNOME 3 because… they really don’t know what GNOME 3 should look like. How is that a refutation of Thom’s point?
It would be better if you would RTFA before posting.
Gnome3 can’t have vision, because it will only exists if task will be impossible to get in 2.x. On the other hand Gnome2 has its vision.
In translation
Gnome3 is only reserved if thy would need to break too many things. All the incremental improvements will go into 2.
Gnome had a lot of improvements, just not visible ones. If you only check how 2.0 was relating with services, hardware and IPC. Everything changed, but… move along, nothing to see.
Another example, all apps now moving on NetworkManager are barely visible, but hell of a feature.
Gnome reduced memory usage and increased speed, again nothing flashy.
Compiz is not well-integrated into GNOME yet. When you select “Enable Desktop Effects”, your window decorations suddenly change, and your window preferences no longer respond as expected
And since when is compiz part of gnome? Compiz is WM, gnome is DM.
Right logic would be for Metacity to get flashy compositor support, not gnome integrating compiz.
p.s. Update compiz in fc6 with the one from rawhide, a lot of speed improvement there.
http://download.fedora.redhat.com/pub/fedora/linux/core/development…
http://download.fedora.redhat.com/pub/fedora/linux/core/development…
It compiled for me on fc6 just fine
Edited 2006-12-31 20:33
Gnome3 can’t have vision, because it will only exists if task will be impossible to get in 2.x. On the other hand Gnome2 has its vision.
The “GNOME 3” label doesn’t refer to an ABI incompatible GNOME 3.0. It refers to “the next major GNOME version(s)”. It’s a short-hand way of referring to the future of GNOME as a platform, not a denotation of a specific version. You can replace “GNOME 3” with “GNOME 2.20” and the basic problem remains. There is a coherent vision for what Vista will look like, or what Leopard will look like, where’s the coherent vision for what GNOME will look like?
I personally don’t think a coherent vision is that important, “evolving” the system as needs dictate isn’t a bad way to go (though its probably less of a good idea for an open-ended system like a desktop than a fairly constrainted system like an OS kernel). However, Thom’s pointing out that such a vision doesn’t exist is completely accurate.
And since when is compiz part of gnome? Compiz is WM, gnome is DM. Right logic would be for Metacity to get flashy compositor support, not gnome integrating compiz.
The fact that Compiz is not a proper part of GNOME is sort of my point. Compiz is clearly the way forward as far as compositing environments go. It’s light-years ahead of Metacity’s compositing, and that’s something that even the Fedora folks (who were originally pushing Metacity + Compositing) realized. Well, if Compiz is going to be the GNOME compositor (and it is in every major distro that ships with a compositor), then it better be integrated into the environment.
The “GNOME 3” label doesn’t refer to an ABI incompatible GNOME 3.0. It refers to “the next major GNOME version(s)”.
Nope, it refers to “when we can’t be compatible, but we really want this huge functionality”
You can replace “GNOME 3” with “GNOME 2.20” and the basic problem remains. There is a coherent vision for what Vista will look like, or what Leopard will look like, where’s the coherent vision for what GNOME will look like?
Gnome looks the same trough whole 2.0. My previous point was explaining that there are more happening on inside than outside.
OSX changed a lot more than Gnome, does that mean that Apple is not visually coherent? (in my opiinion they aren’t). Again look how Vista changed from XP. God damn basic functionalities are completely different.
Sorry, but you don’t know what you talk about. Either that or you don’t read about plans. Look how long they were planing D-Bus move or all other. It is all part of the plan where every bit must first fit to be fitted inside.
Look at another example, Long running tasks, really cool feature. Why is not yet inside? Because it is not ready. And after it will be there, then you can expect complete gnome to start fitting with this functionality. First POC, then inclusion and then fitting. (same goes for Compiz, but as you see… compiz will never be there, explanation later)
The fact that Compiz is not a proper part of GNOME is sort of my point. Compiz is clearly the way forward as far as compositing environments go. It’s light-years ahead of Metacity’s compositing, and that’s something that even the Fedora folks (who were originally pushing Metacity + Compositing) realized. Well, if Compiz is going to be the GNOME compositor (and it is in every major distro that ships with a compositor), then it better be integrated into the environment.
And it will never be. Sorry, Compiz is only 3D compositing, try using it with 2D only. Fedora never pushed Compiz as such. They simply had to take one solution that already works. And since Metacity has way more chances to get 3D compositing too than compiz getting 2D, you can guess which will be default WM. But this means that Compiz will have to fit inside and not otherwise around.
The “GNOME 3” label doesn’t refer to an ABI incompatible GNOME 3.0. It refers to “the next major GNOME version(s)”. It’s a short-hand way of referring to the future of GNOME as a platform, not a denotation of a specific version. You can replace “GNOME 3” with “GNOME 2.20” and the basic problem remains. There is a coherent vision for what Vista will look like, or what Leopard will look like, where’s the coherent vision for what GNOME will look like?
Babe, that isn’t the case, and you well and truly know it; it has ALWAYS been the status-quo within the IT community that a move to a x.0 means a break, a big step forward – “we’re doing something radical, so there fore, we’ll need to break some stuff” and the changing of the major version number demonstrates that publicly.
Want a parallel – Windows Vista has moved Windows from being Windows NT 5.2 (Windows 2003) to Windows NT 6.0 (Windows Vista); things have been broke, 100% backwards compatibility isn’t so, and there are some radical changes; compare that to Windows XP to Windows 2000, it was an evolution rather than a revolution.
I personally don’t think a coherent vision is that important, “evolving” the system as needs dictate isn’t a bad way to go (though its probably less of a good idea for an open-ended system like a desktop than a fairly constrained system like an OS kernel). However, Thom’s pointing out that such a vision doesn’t exist is completely accurate.
The issue isn’t about the lack of a ‘coherent future plan’ – no has denied that, what is being debated is whether that is a bad thing; Thom has stated that there is no coherent future plan, so therefore, the project is in a state of visionless limbo.
The simple fact is, he is looking at it from the angle of a company putting together a product and shipping something every 2 years, with paying customers and investors wanting profits on tap like water – GNOME and KDE don’t operate like that; heck, the whole open source model doesn’t run like that.
People work on things independently, then when the respective desktops request for ‘features to be added’ these maintainers come forward and offer their products to be included, if they’re within the scope and mission statement of the respective desktop, it is merged into the ‘mainline’ of the desktop distribution.
Add to that the fact that distributions can add even more applications on top of GNOME and KDE which aren’t officially aren’t part of the standard distribution, its difficult to claim that some how that GNOME or KDE is standing static.
Releases are quick, contributions come from a number of sources outside the GNOME and KDE community, and visions of ‘future development occur by those not necessarily within the GNOME and KDE community.
The fact that GNOME doesn’t need to have a giant overhaul is testament to the future proofed designed which has gone into it, same goes for Xorg and the other various parts which make up a *NIX desktop – compare that to Windows where by if something is fiddled around down the bottom, all hell breaks loose at the top, and any new features that need to be added have to be scaled back/castrated to maintain backwards compatibility with applications, which quite frankly, should have been killed off long ago.
p.s. Update compiz in fc6 with the one from rawhide, a lot of speed improvement there.
I couldn’t see much speed improvements, but it was nice to see that the 0.3 series of compiz now accepts metacity themes. This makes it look much more part of Gnome.
However there is a bug in the pager, as the cube doesn’t rotate when you select a different desktop from the pager applet. Or are there any new settings to enable it that I’m not aware of?
The cube effect still works, if you drag a window from one desktop to the other.
However there is a bug in the pager, as the cube doesn’t rotate when you select a different desktop from the pager applet. Or are there any new settings to enable it that I’m not aware of?
The cube effect still works, if you drag a window from one desktop to the other.
That’s the difference between “workspaces” and “viewpanes” (or however one wants to call them). Workspaces are completely separate from each other, viewpanes are just sections of a larger workspace. Metacity did away with this awkward separation and only supports workspaces. The Compiz cube uses viewpanes instead (since all the sides of the cube are part of one, it would be difficult to use workspaces since you couldn’t have windows sitting on the edges of the cube for example).
Compiz now added workspace support, probably for better compatibility with Metacity. But to get it working right you should switch to a single workspace, which will then be displayed as one large workspace the size of your cube.
> they really don’t know what GNOME 3 should look like.
> How is that a refutation of Thom’s point?
Missing the point. There is no clear vision of GNOME 3 because the developers don’t see a reason for breaking compatibility. Breaking compatibility will require rewriting apps and is generally a bad idea unless you really have to. There *are* however, specific goals for the next release of GNOME and some ideas for future releases. Have a look at:
http://live.gnome.org/RoadMap
A great deal can be done while still keeping backwards compatibility. For instance, since GNOME 2.x was released, dramatic changes like changing window managers, adopting an XGL backend, reducing the need for GNOME specific libraries by moving more functionality into GTK+, memory use reductions, memory speed-ups, adoption of D-BUS and removal of Bonobo as a core technology, adoption of GStreamer, and other major changes have been done in a backwards compatible manner (e.g. rewriting the old libraries to be implemented in terms of the new ones and making the old libraries optional and changing internal data structures).
It’s much the same way with the Linux 2.6 kernel. One day, there may be a Linux 2.8 release, but right now there’s no need to break backwards compatibility. This doesn’t mean that the Linux kernel has no plans or that no progress is being made.
Again, wasn’t going to post over the festive period, but I’ve found myself perusing the articles again :-). Happy New Year (in advance)!
Vista contains more improvements than have been made in GNOME since 2.x. Sure, the UI is still inferior to GNOME’s
I don’t think it is. There are some gaping holes in Gnome’s UI you can drive a lorry (truck) through, despite the clean image they’ve tried to cultivate.
Instant apply is still poorly thought out and some Gnome applets have instant apply and some have the usual OK/Cancel. I don’t have a Gnome desktop around me now but many of the Gnome applets were horribly inconsistent as well, and the accessibility one stood out. Spatial browsing has also been an idea that seems good when skim reading a usability theory book, but doesn’t work in practice.
I also cannot abide a desktop that simply doesn’t have the right technological underpinnings to create even moderately complex user interfaces. An example is that Gnome does not have a theme manager that gives me a useful and usable preview of what might desktop will actually look like, without me having to apply my settings first, like every other desktop in existence.
When you select “Enable Desktop Effects”, your window decorations suddenly change, and your window preferences no longer respond as expected (ie: double-click to maximize).
Compiz is not finished yet (it will be included in RHEL5 as a “technology preview” feature rather than a supported feature). For example, the 0.3 series supports Metacity themes so there are no decoration changes when you enable it.
It’s not completely consistent with Metacity in other ways but there is work being done there. I’m sure the double-click to maximize feature is also a standard in compiz, so you must be thinking of some other preference. The most significant one that I can think is the keyboard shortcuts capplet, which is also being fixed to work with both wm’s.
Vista contains more improvements than have been made in GNOME since 2.x.
I can only disagree with that. Vista is in many ways a regression from Windows XP, and a significant number of people will skip it for as long as they can. OTOH I can’t find one soul that would rather use CentOS 3 (with GNOME 2.2) than a modern GNOME 2.16 distribution.
Really, I think Thom and anyone else that thinks GNOME hasn’t made huge improvements in the 2.x series should download and use CentOS 3 for one week. That should be fun.
The crappiness in Vista is a result, as I said, of poor taste, not poor technology. Technologically, it’s an enormous step forward from XP.
Right, so there is no clear vision for GNOME 3 because… they really don’t know what GNOME 3 should look like. How is that a refutation of Thom’s point?
There was a clear refutation of Thom’s point. There is a vision for the future, which is what really matters. That vision involves more Gnome 2.x releases. Gnome 3 will happen when backward compatibility can no longer be maintained, which is probably more than a few years away. From what I’ve seen, the biggest complaint most people (myself included) have about Gnome 2 is that too many pieces still aren’t complete. Having to recode everything to Gnome 3 would just make that situation worse.
Maybe Gnome developers should pick some big, fancy feature, build up a lot of hype about it, and then call it Gnome 3 when they release it. (just kidding)
“Do it wrong, do it better next time.”
I’m a KDE/Gnome user, i’ve switched to gnome after instaling FC6 and realized that all the simple things i needed were compleatly functional in gnome. Bluetooth support, compressing files, managing USB HD, MSN, IRC, WWW, Mail,… all in a blink of an eye, and I saw that was good.
My point is, gnome is behind of KDE in features but the ones he has are functional. Maybe my lack of knowledge of gnome full power (since i’ve left gnome far at RH9) and since then must has change. The featureless of gnome was realized by me when a friend of mine said that know i can pick a taskbar and move it to the Desktop Area i wanted (in Gnome 2.16), and i said, omg.. that is something that i’ve being using in KDE for ages.
For conclusion, GNOME is a clean, functional desktop enviroment, and KDE is a powerhouse of features and good looks, both are far greater than winXP DE and becoming (if not already) greater than Vista. As of GNOME’s plans… there’s no need to a 3.x version because it already does everything we need. KDE is going for a major version release because QT4 is a huge move for developers to let it pass by.
Steady Improvement and stability is far better than earth shattering. At one time I disliked Gnome, now it seems to be my desktop of choice. There are things in KDE I like better, but it seems Gnome is the more user friendly desktop for my day to day use.Gnome and KDE both rely on a community of developers, yet they have eclipsed anything Microsoft and Apple have accomplished in any comparative time period. Don’t scream at me about “how popular” Microsoft and Apple are, I’m talking about features and usability, you cannot top Gnome, KDE, or GNU/Linux.
I’ll give you the time-span bit — it’s really incredible what GNOME has accomplished in the five-or-six years the 2.x series has been in development. However, from a features and usability point of view, Aqua is a bit ahead of GNOME. The OS X GUI is both more powerful and more complex than GNOME’s, though it does a very good job of managing that complexity properly.
Going deeper, Aqua has the technological base that GNOME has always lacked. The modern iterations of Carbon*. GTK+’s API is an order of magnitude more sensible and well-designed (of course, it doesn’t have 20 years of baggage to carry around), but in terms of features it really can’t hold a candle to to Carbon + Quartz + Core*. This will be even more true in in Leopard. Composited layers, pixel-shader based special effects in widgets, resolution independence — GTK+ is still a long way from competing with that.
*) Which have been refactored around the new HIToolbox APIs.
I can’t debate technical(LOL)! I’m just an average PC user, under the hood I know just enough to be dangerous! All of my opinions are based on my personal experiance with Windows. My “Mac Friends” think I’m out of touch with reality by not using a Mac(I won’t say fanatical, just very entusiastic about their OS). My comments are based on my limited knowledge, and I admit to limited knowledge about “under the hood.” What I have enjoyed with KDE, Gnome, and GNU/Linux is rock solid stability(Debian Sarge, and now Etch), and things “just work,” I haven’t been able to say that about Windows. I will concede that many of the programs in a Windows enviroment is not all Microsoft software, such as RealPlayer, Quicktime, and so on, but I don’t like having to put up with all the nagging and reminders that these and Microsoft programs pester you with upon login. Yea, I know you can disable it, but the GNU/Linux, Gnome/KDE desktops are far more user friendly to me. I still believe Gnome/KDE have eclipsed anything Microsoft and Apple have accomplished in the same time period, but in the interest of honesty, I will concede my lack of technical knowledge, or programming skills. My two cents worth
I use OSX all the time because I do a fair amount of video editing with final cut studio and while I must say it is a very polished os and aqua is a very nice ui, I really prefer KDE to it and in recent months I also perfer Gnome. OSX has it’s strengths, but but gnome and kde seem a little more user friendly to me. I must say this is slightly bias because I have used KDE almost exclusively for years and I am also a bit of a computer geek.
The thing is…… when in put Gnome or KDE in front of a long time Windows user they can figure it out, the same is not true for OSX.
I like OSX, but only because I can make it do what I want it to do via cli when it is necessary, and OSX took a long time for me to get used to after coming from a linux/bsd world and ultimately a windows world.
Edited 2007-01-01 01:35
I use OSX all the time (…) while I must say it is a very polished os and aqua is a very nice ui, I really prefer KDE to it and in recent months I also perfer Gnome. OSX has it’s strengths, but but gnome and kde seem a little more user friendly to me.
But you can easily port KDE or Gnome to OS X, using Darwin ports.
I applaud Thom for agitating both KDE and GNOME so much that they felt they had to explain to the world what’s going on with their next-gen desktops.
Because I sure didn’t know, and it’s interesting to read about. Most of the information may have been out there already, but scattered over many different places.
Nice.
This POS site still exists?
I don’t get why anyone would listen to some random f–king moron on a shitty site. I wonder why this ugly, stupid, fud spreading down-syndrome face “Thom” isn’t dead yet. Here (the internets) he can spout his retarded nonsense in a physically safe environment, but you’d think that someone who pisses off people with his constant retardation would get the shit beaten out of him all the time.
Can’t wait until we meet actually…
Hi everybody:
This is an offtopic comment, but I want to wish you all a happy new year 2007, OSNewers !!!
Happy new year to the OSAlert crew also: Thom, Eugenia and all your colleagues and collaborators!!!
Well, happy new year!
I Wonder why the many WM’s always seems to be forgotten, in the “Desktop Linux” debate? There is a very healthy selection of different user interfaces, that doesn’t follow the Windows GUI paradigm, but isn’t that exactly what’s fresh and “new” on the desktop?
Even a simple wm like FLuxbox/Blackbox can be picked up rather easily by new users (the concepts atleast), just right-click and select program – with the 2-3 most used software split between a couple of desktops. Enlightenment is my personal favorit for the last ~4 years, and I so much look forward to E17 (writing this from Elive, coincidally I must say that E16 has served my workflow perfectly the last many years, with the darkone/lightone theme, to the point where I feel slowed down, when I’m using Gnome/KDE/Windows. So nothing but respect for Rasterman and the Enlightenment Dev-team!
Will be interesting to see what the PS3 can do for Linux, and Enlightenment 0.17, in 3 years time.
There’s a reason. These WMs lack significant functionality and integration. They are simply a non-solution for Linux on the desktop. Back in the day when all graphical apps were written in bare Xlib or Motif, using a plain WM was probably fine because that’s all there was. Now that we have desktop environments with their own toolkits and development APIs, as well as a number of configuration utilities, games, productivity software titles, webbrowsers, etc., window managers simply cannot stack up.
“””
There’s a reason.
“””
Yes. I always have to wonder about the sanity of the people that these kinds of articles tend to draw out. The ones that say things like “No one needs bloated desktop environments! I use TWM and some hand-cobbled scripts and I’m happy as a clam!”
“Crazy as a loon” is more like it if they think they are going to get the rest of the world to agree!
However, I suspect that it’s not insanity at all, but just some sort of weird geek bragging right. Kind of like using Debian.
Edited 2007-01-01 22:09
When it come in infrastructure while not publicly marketed, E17 does have a lot of potential if taken up and used as the basis for a desktop environment. I do agree that they have alot of potential, its just that no one besides the actual Enlightenment devs seems inclined to release a distro based on it (such as Elive). All that I think the E17 guys lack is marketing. I for one have always thought to my self that many strategies would be best shared between say GNOME and E17 or KDE and XFCE.
Come to think of it all these major debates seem to boil down to the attempt to use Windows as the reference. Those other DM/DE such as XFCE, E16/17, Fluxbox/Blackbox I think simply need more endorcement by way off utilization. Most of them have passed feature parity years ago, its the polish that marketing like which they lack I think.
from Dropline to Gnome/Edgy Eft, from KDE/OpenSuse to KDE/Slackware, I gotta say that, I personally prefer Gnome to KDE for its simplicity and functionality. There are a ton of rich features in Kde. It is good and far more enough than I need. A rich KDE options to configure Desktop make me feel lost. I am not a *DE geek but I can see Gnome serves all my needsquite well. I simply use some essential works such as Evolution for mail/ gedit / Abiword/ GBaker / Gthumb Image Viewer/GIMP for web graphic design /gFTP and of course, Firefox (i prefer swiftfox) and OpenOffice, as well as Litespeed webserver SQlite are my favourites but they are not G(DE) :-). I dun think I need to look further more DE to start over. All really is well integrated in Ubuntu Box
Happy new year y’all
Ken,
I will be installing FreeBSD 6.2 when it gets released but before I do, are there any major differences between KDE and GNOME that would have me installing one over the other?
The two are completely different. KDE is prettier, flashier, and more featureful. GNOME is simpler, more consistent, and places a huge emphasis on proper usability practices.
There is no alternative for trying out both and seeing which one you like better. If you try GNOME, and you’re really a KDE’er at heart, you’ll feel confined by its simplicity. If you try KDE, and you’re really a GNOME’er, KDE’s UI will drive you slowly insane.
Thank you for the information. It sounds like I will be using GNOME and I was leaning toward it anyway because of AbiWord, Epiphany, and Eye of Gnome.
“””
KDE is prettier, flashier, and more featureful. GNOME is simpler, more consistent, and places a huge emphasis on proper usability practices.
“””
Indeed. But is that dichotomy absolutely necessary?
What if KDE started putting the same emphasis on proper usability practices, but implemented “User Levels”. One option that you could set to one of:
Novice: Gnome-like simplicity
Intermediate: More options
Expert: The full KDE UI chaos that many seem to love
There is no reason that the objective of usability for new users *has* to clash with flexibility for experts.
I just pulled those labels out of the air. In Quake, they used “Easy”, “Normal”, “Hard”, and “Nightmare”.
Every UI has to have a design, a philosophy, an asthetic. Its difficult enough to create a single interface that is consistent in all of these parameters. What you’re talking about is creating multiple such interfaces. If it’s even possible, its a feat that has not been accomplished on any substantial scale in any product.
But I’m not really talking about 3 different interfaces. I’m talking about offering subsets of what already exists. Do newbies who would never consider changing the positions of the window buttons really need to see those options? Maybe later, after they are ready. But not while they are ‘novices’.
Gnome shuffles the whole kit and kaboodle off to the configuration editor. Firefox shuffles everything off into about:config, for that matter. That seems suboptimal to me… as if Quake had only had two levels: ‘Easy’ and ‘Nightmare’.
To make the ‘Easy’ level, they just had to take some monsters out. Same with KDE, I think.
To make the ‘Easy’ level, they just had to take some monsters out. Same with KDE, I think.
*ROtFL*
Thank you SO MUCH – you just made my crappy morning SO MUCH better
KDE needs way more than just taking some monsters out. It needs fundamental level redesigning.
KDE needs way more than just taking some monsters out. It needs fundamental level redesigning
No, KDE’s fundamentals are pretty solid. KDE’s emphasis to date has been less about HIG and more about building an accessible framework of reuseable components and libraries for developing integrated applications. It does a pretty good job of this. As a user I personally care less about the layout of my toolbar or configuration options I rarely modify than I do about having consistency with components like kparts, kio-slaves and the accessibility provided by the KDE file picker across the apps I use.
But that’s not to say useability doesn’t count. Where KDE has fallen down to a certain extent isn’t even so much the configurability of the GUI and core apps, it’s the presentation layer. They acknowledge this. There’s a very strong emphasis in KDE 4 for common interface standards and useability. They’ve been heavily involved with OpenUseability.org. They’ve been conducting controlled and measureable user interface testing. They’ve even been documenting and cross-comparing Windows, Gnome and OSX to see what works well and what doesn’t. Heck, a couple of weeks ago one of the design members posted a blog on the applicability of elipses (…) in menu options, when and where they should be used, for consistency. In the past that would have been laughable, for the most part. But most importantly, everything will ultimately be documented and enforced.
Having said all that, it’s still superficial. KDE’s inherent Kiosk framework already allows a considerable amount of simplification for KDE applications. Menu options can be removed, toolbars can be simplified. Although that’s mostly intended for administrators looking to lock down desktops, some distros such as Kubuntu made configuration changes to simplify the appearance and use of some of the core apps, such as Konqueror or the control panel.
Basically KDE and Gnome have taken diametrically opposite approaches. Gnome chose to emphasize HIG and interface above all else, with the intent to build an actual framework behind it. KDE emphasized the framework and components first with the ultimate intent to now build an effective and useable interface on top of it.
They’re both converging on the same point, though I suspect adding clean interface guidelines to an extensible framework will be easier than doing it the other way around.
Just my 2c…
KDE needs way more than just taking some monsters out. It needs fundamental level redesigning.
I don’t see any evidence for that (nor do I see any examples), although I’ve seen more than one person try and hint at it before ;-). More usability and accessibility support in kdelibs as well as an adherence to the upcoming HIG and discussion on usability issues (something that Gnome does not have – their HIG is merely pulled out of a theory book) is what’s needed.
As mentioned previously, Gnome’s control panel applets are still hellishly thought out and inconsistent, instant apply has been pulled from the Mac world without any thought whatsoever and Gnome still cannot give me a preview of what my desktop will look like if I change the theme, colours or style, because it cannot even create interfaces of even moderate complexity. Is not providing that being usable? Well, no it isn’t. Even Mac OS gives you all that, without any loss of functionality. Just read your first post for more.
If I was being cynical, the drive for usability and simplicity in Gnome was, and is, being used as an excuse because they just don’t have the technological underpinnings or resources to achieve what they need, rather than just admitting they don’t have the resources like other OSS projects (even Linus Torvalds hinted at that in one of his e-mails).
That is not working towards a usable desktop, and I think many people should look up the meaning of that word because the meaning has got lost.
The chances of there being a Gnome 3 are not entirely great because unless they make their technological base more flexible, and build it on a sensible object oriented and natively compiled language or toolkit for developers to build off (native gcj Java with Classpath or something), it ain’t going to happen regardless of what anyone thinks.
“””If I was being cynical, the drive for usability and simplicity in Gnome was, and is, being used as an excuse because they just don’t have the technological underpinnings or resources to achieve what they need,”””
And you would be demonstrably wrong. Gnome 1.4 had all the knobs, buttons, and other play toys that ‘advanced’ users always seem to whine^Winsist that they need in order to be productive.
Say what you will about the details… spatial… configuration editor, etc. But if the Gnome guys wanted to dazzle us with a dizzying array of config options, they could.
Fortunately for my users… they don’t.
And you would be demonstrably wrong. Gnome 1.4 had all the knobs, buttons, and other play toys
No it didn’t. Gnome 1.4 simply didn’t have what a other desktops have had for years.
But if the Gnome guys wanted to dazzle us with a dizzying array of config options, they could.
No they couldn’t.
Fortunately for my users… they don’t.
Fortunately for your users they’re not likely to be able to change their theme effectively to something that works for them and they won’t have a reliable print dialogue either. Wonderful.
Actually Breadbox Ensemble comes with that kind of functionality… but I don’t know how well it works, if it works at all.
Perhaps a worthy experiment for the Wii or PS3?
There is no reason that the objective of usability for new users *has* to clash with flexibility for experts.
Usability:
The extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use.
In short: Yes there is.
“KDE is prettier”
“Pretty” is in the eye of the beholder. Personally I think KDE is butt ugly, and I’m sure I can’t be the only person on the Planet that thinks so.
I’m totally new to Linux, but my experiments with various distros have shown me that I much prefer GNOME purely because of its aesthetics (weird maybe, but true).
However, the purported ‘simplicity’ of GNOME just irritates me. I find the lack of configurability a major annoyance and an impediment to accommodating my way of working.
The goal of GNOME isn’t to be simple, it’s to be consistent. Eventually GNOME will have most of the complexity of KDE (it’s been accelerating over the last few releases), but (at least attempt to) do so in a way the the complexity is hidden from those who don’t care (e.g. through sane defaults or higher level features that set sane defaults en mass), but still be available for those who need it.
Providing a consistent, easy to use, easy to learn interface that helps you move across the “chasm between novice and intermediate” and the “chasm between intermediate and expert” is hard, but for the most part, but GNOME is evolving nicely into this sort of desktop.