“Like clockwork, the GNOME project released GNOME 2.18 six months after the release of GNOME 2.16. The new version carries a number of improvements over the 2.16 release, but doesn’t bring many ‘must have’ features that would compel users to upgrade right away.”
…water is wet.
One of the the most common gripes i hear from people about linux is the lack of interface consistency in most distros. For a while KDE has had a very consistent appearance. Its nice to see that amongst the ‘incremental’ improvements to Gnome 2.18 is improved interface consistency. From using feisty, I would say that Gnome has finally reached parity with KDE. It looks very simple but polished, and I’m pretty sure its even more responsive than 2.16. Congrats to the Gnome project, I think this release is far more of an improvement than the author points out. He (she?) does a great job of highlighting those ‘incremental improvements’ but doesn’t highlight the general feel and appearance as much. I personally have used and Gnome for 3 years now (and still do) but have used KDE occasionally. Even though its advancing so nicely, I’m pretty sure when KDE 4 rolls out, that it’ll put the Gnome projects 2.20 release in the shadows.
Edited 2007-04-04 00:40
What I find funny though, that the people who complain the most about the lack of consistency ignore the ugliness in their own back yard, be it MacOS X or Windows.
Sure, GNOME and KDE aren’t perfect, but when it comes to a consistant user interface when running KDE native applications on KDE, or GNOME native applications on GNOME, nothing can *beat* the consistency I see.
On Windows, there is a mirade of skins and crap that add nothing to the user experience, distract people from the task, do little to improve productivity and is a nice diversion for customers to ignore the hideous ugliness underneeth and their failure to fix long standing bugs in their software.
What I find funny, is that as soon as Linux inconsistency is brought up, someone immediately brings up OS X and Windows.
Pointing out others flaws, does not fix your own.
“Sire, the peasents are dying”
“So? people die in my brothers Kingdom too”
If Linux developers would stop changing the subject and admit to inconsistency and fix it, Linux can be /better/ than Windows and OS X when it comes to consitency. Linux users need to get out of the habbit of always deferring every constructive criticism elsewhere.
If someone say OS X is inconsistent to me, I agree, and point them toward UNO, a fantastic fix for that.
If Linux developers would stop changing the subject and admit to inconsistency and fix it
What I find funny is that MS advocates often, ever so conveniently, leave out facts from their opinions. But it’s kind of understandable as they have learnt from the best: Microsoft’s Marketing Department…
So like the OP said:
“Sure, GNOME and KDE aren’t perfect, but when it comes to a consistant user interface when running KDE native applications on KDE, or GNOME native applications on GNOME, nothing can *beat* the consistency I see.”
It is already fixed.
Edited 2007-04-04 11:24
I’m a Mac user, not a Micosoft lackey; and secondly Linux is functionaly inconsistent, not visually inconsistent. Near to all Mac apps act the same. Quit is Cmd-Q, Preferences is Cmd+, and always in the Application menu (unlike Windows), and the clipboard even works everywhere, like it should.
I haven’t used a mac all that much, but the few times I’ve had a chance to play with OS X, I couldn’t help be reminded around every corner how much it reminds me of Gnome. This is a good thing, in my opinion. But let’s be fair, when you have a lot of different programmers / projects all working on separate pieces of software, there are going to be some problems with inconsistencies. As you said though, Alt+F4 will close any window, but of course that’s a function of the window manager more than the application.
Usually programs that are actually part of the Gnome project or the KDE project do follow all the consistencies, but then how many people stick to just the core Gnome or core KDE apps?
But at least the applications will look similar if they’re using the same toolkit (they all pick the theme from the same place, etc.) Comparing to Windows, there is no consistency even within Microsoft’s on software (for example, look at Messenger, Media Player, and Office (especially 2003 and 2007, neither of which fit into the OS at ALL!)
Mac does have a slick interface, and I think Apple more than anyone else makes that possible by helping developers with good interface guidelines and most of the programs are going to be done by professional publishing houses, and not by some guy trying to scratch an itch.
Linux is functionaly inconsistent, not visually inconsistent.
Once again, the Original Poster said that KDE apps are consistent, when run in KDE and likewise for Gnome, which is *true* even functionally.
You keep forgetting that basically we have Gnome and KDE users just like we have Windows and OS X users.
It becomes inconsistent only when you mix KDE and Gnome programs. It is a similar situation to running Mac apps in Windows (e.g. Quicktime).
Edited 2007-04-04 14:21
Keyboard shortcuts are usually quite identical between KDE apps and Gnome apps and between QT apps and GTK+ apps.
Quit is always Control-Q – this is true for QT apps, KDE apps, GTK+ apps and Gnome apps. So that part is quite weird.
Copy and pasting between KDE/QT and Gnomde/GTK+ apps also work fine (in most cases) though Drag’n’Drop often depends on implementation of FreeDesktop “standards” (often drafts). (Shame on Nautilus devs.)
But there are some important shortcuts which are different and which are annoying in everyday work when combining best of both world apps:
– Switching between tabs in an app, for example gedit/kate, gnome-terminal/konsole
– Redo: strg-y vs. strg-shift-z
– Fullscreen mode: I think nearly every app has a different shortcut for that…
– …
And KWin and Metacity/Compiz should use the same shortcuts so that it’s easer when switching between both desktop environments.
This can be amended. In Gnome and KDE you can modify the keyboard shortcuts.
You know, you can file bug reports on these issues
Cannot edit my post, so it will be 2 replies.
Fullscreen is usually F, though I don’t have many applications supporting fullscreen. It would be Totem, VLC. OO.o Impress does not have a fullscreen function, but shows presentations automagically in fullscreen.
Firefox (for GNU/Linux) however uses F11 for fullscreen. I don’t know if this is true for the OS X version of Firefox. Evince does also use F11. However, most people expect media players to use F for fullscreen, so it’s not really a inconsistent that documentviewers (including browsers) all use F11. Eye of Gnome does also use F11 for fullscreen.
I have 1 QT application that supports fullscreen mode DjView. It has no keyboard shortcut for that. Scribus apparently uses F11 for showing the whole page. But it is not a fullscreen mode.
There is some consistency, but it could probably be better across the toolkits and guidelines.
Actually, most windows apps are functionaly consistant also, hotkeys generally work across apps, and in most cases where they don’t, that is the developers fault and not MS’s. Which I would believe would be the same if a 3rd party Mac application didn’t follow the standard conventions. You can’t blame the OS vendor when 3rd party apps don’t follow. Even with office, which often has a different visual theme (inconsistant, I agree) the hotkeys remain the same, and are the same across Windows and Office
“You can’t blame the OS vendor when 3rd party apps don’t follow.”
Oh, just how you cant blame GNOME/KDE when 3rd party apps don’t follow?
Oh, I don’t blame them either, I was just commenting that Windows is functionally consistant also, as the previous post just focused on OS X (other than declaring themselves “not an MS Lackey”). THere is only one group to blame in this situation, regardless of the OS, and that’s the 3rd party developers
Windows isn’t consistent. Even the standard dialogues for Desktop Preferences in Windows are quite different from each other. And one Microsoft application is quite different from another. Try blending parts of Office XP with Office 2003 with Windows XP or Windows Server 2003.
It’s really a mess. It is the one platform without consistency even in basic dialogues.
I never talked about visual consistency, I was talking about functional consistency, where hotkeys and menu items tend to be the same, I actually consider Gnome and KDE much more visually consistent, but when you share apps across DE’s, you get functional inconsistency. I think Gnome and KDE do a great job blending in visually with one another. I think OS X is actually the worst, visually, but it is once again getting better.
Menues in KDE are quite like menus in Gnome. Basic hotkeys are identical. Which is no surprise since basic hotkeys are identical in Windows, KDE, Gnome and (replace Ctrl with Cmd) Mac OS X – and BeOS, Haiku, SkyOS, Syllable and probably a dozen other obscure OS’es and DE’s.
Excuse me, but if you make a statement honey, you’re implicitly compareing something with an ‘other’ – whether that ‘other’ is MacOS X or Windows, you’re still doing a comparison.
You can’t simply just look at something in an isolated situation without the context of having something else, or otherwise you would never know what consistency was unless you actually saw it and used it.
People crap on about “inconsistant user interface’ with GNOME and KDE – how about this; less talk, more facts. How about using examples of where this inconsistency exists within GNOME, because for the love of me, I can’t find it!
While this is true in some ways, if you could only compare against things that exist, how would the state of the art ever improve? You can also compare things to an image of what you believe they could be, and that might be what the original poster was trying to do.
> Pointing out others flaws, does not fix your own.
True, but it’s often brought up by Windows users as a negative point – linux isn’t consistent. Well, that might be true, but compared to KDE or Gnome, Windows is a mess…
I know it’s a drag having to install 3rd part software to achieve the consistency desired, but personally I cannot install Mac OS X 10.4.x on any Mac without installing UNO (including the Firefox theme to make it look like Safari) as the first thing, along with any system- and security updates!
http://gui.interacto.net/index.php?option=com_content&task=view&id=…
Anyhow I’d say it’s about the same thing as running Ubuntu’s Human theme to make GNOME look like you want it to…
I’m pretty sure when KDE 4 rolls out, Gnome 2.20 will be but a distant memory
Well, maybe everybody will have forgotten 2.20 when 2.22 is out, but then again – most ppl have better memory.
gnome 2.20 will be out in 6 months, right? that means sometime around september? 4.0 is slated for late october … let’s say we even miss that by 3 months bringing us to january. gnome 2.22 still won’t be out by that point. the only way i can see 2.20 being a distance memory when kde 4.0 rolls out is if people switch away from it to something else.
I gotta admit, I was a bit disappointed, as last years summer of code produced several projects I was excited about, but it looks like they are slowly dying off
I was in paticular looking forward to the new gnome scanner program as xsane sucks and gsmartmix for improving the sound experience in gnome… I think those types of incremental improvements are important for gnome. I hope in 6 months, we’ll get a release that adds all of that stuff… its starting to look like google SoC wasn’t as successful as I thought it was.
One natural characteristic of 6-month spring/fall release cycles is that a combination of weather and social constructs cause spring releases to be modest and fall releases to be more ambitious. This is especially true in free software, which gets a huge boost every summer (in the dominant northern hemisphere) when many students are off from school.
I realize this, but a lot of the stuff I mentioned was developed last summer… during the release cycle, it looked like it was going to be an impressive release, with gnome-scan and network-manager, which IMO bring the gnome desktop above windows/mac in both of these areas from using the beta versions of these programs IMO… I do hope that the devs and other contributors have time during the next release cycle to fix the showstopper bugs in each and get them nice and polished and integrated into the gnome desktop… I’m also praying ubuntu focuses on their printerdrake spec for the next release cycle… these 3 components actually have decent support on Linux at this point, but not an impressive interface to show it off.
Running NetworkManager on my laptop with all the VPN plugins impresses people at work and school who see how much easier the Linux VPN client software is then the Mac and Windows counterparts, and the wireless dropdown menu as well for the Windows users (Mac users at work insist on denying the network-manager dropdown is better then the Mac one.. the ability to reselect the connected network to “Repair” it makes it better for me, I see Mac users rebooting all the time because they can’t repair their IP when the DHCP server screws up)
Edited 2007-04-04 03:19
just to tell you that gnome scan is *certainly* far from dead. I’m not the author, but I can see that Étienne is putting a lot of effort into it as we speak. At the moment, I think he’s rewriting some big chunk of it though.
im looking at the change log to see how many features were removed in this release lol lmao
Edited 2007-04-04 03:35
And your point is? I prefer software developers to stop and fix the bugs, rather than just keep on releasing pointless upgrades with new features that are half baked and unreliable due to unfixed bugs. If Gnome releases 2.18 with minimal new features and fixes a lot of bugs, I think that’s good of them.
Dave
I’m sure the Gnome team will kick it up a notch when they realize big bad KDE 4 will be knocking at their door come Fall.
The best new feature in 2.18 is Seahorse
So, the best new feature is that someone look at kgpg and thought, let there be light
I’m underwhelmed at the changes between 2.16 and 2.18
If the changes improve enough, they don’t necessarily have to bring revolutionary features. In this world, we value improvements more than rushed out half-ass shiny junk.
There’s one improvement the article fails to notice: You can change Clearlooks theme colors! I’ve got a green Ubuntu, finally …
But seriously, only six months have passed, and many things have been worked on: a lot of work has been done in improving the Cairo graphics layer to make it faster and leaner; a lot of work has gone into gstreamer, too; a lot of work has been put into cleaning up the gtk+/gnome libraries; and many bugs have been killed. These changes don’t show on a static picture, but they make Gnome a much better product than it was six months ago.
Apps have also improved due to this improved infrastructure: Jokosher has advanced a lot by also pushing gstreamer forward. And Inkscape is considering Cairo for all their rendering, instead of their own layer. Most apps in the Gnome suite have improved.
And there’s a lot of work that is almost done, but not quite yet. Compiberyl is almost there, and so are the desktop applets. When Glitz arrives, we will have a lightning fast 3D-accelerated full graphics layer.
We shouldn’t expect a revolution every six months: It’s impossible, and it would be bad if it weren’t. But whoever thinks Gnome has stagnated, please install Gnome 2.0, and watch its evolution in 5 years. And wait 6 months more and watch it go a bit further.
Will KDE4 be a kick in the a$$ that will propel Gnome developers into an improvement frenzy? Dunno, might be. KDE4 is looking good indeed. But I hope that Gnome keeps getting more solid, more reliable all the time as it has done so far, even as it gains new features.
You don’t need to install Gnome 2.0. The PC I’m using at the university lab has Gnome 2.10. I remember that when I was using Gnome 2.10 at home, two years ago, I was completely satisfied, but now that I compare it with Gnome 2.16 (I haven’t upgraded to 2.18 yet) I see that in 18 months the Gnome developers introduced so many major or minor improvements that the user experience is incredibly different. I didn’t notice it when I regularly upgraded to 2.12, then 2.14 and finally 2.16.
Indeed, that’s a disadvantage of a faster release cycle (though, normally KDE’s cycle is just 2 or three months longer). Gnome 2.10 vs 2.20 will sure be very different, but it should, after 2 and a half year of work. It will take KDE more than 2 years to get 4 out, so you can KDE 3.5 vs KDE 4 with the same difference. See how it turns out 😉
actually, i think if you look at the difference between gnome 2.10 and 2.20 and kde 3.5 and kde 4.0 you’ll notice some qualitative differences in the type of changes.
while there are many of the kinds of changes that happen in a “regular” release between 3.5 and 4.0 (namely bug fixes and feature improvements to existing applications as well as some new apps), there’s a huge number of fundamental changes that really, really difficult to accomplish in a series of quick release cycles.
in theory one can take multiple release cycles to do bigger framework type changes that affect the entire environment and introduce such changes one at a time as each as is ready. however, this is problematic in that it’s not easy (and sometimes infeasible) to track the main trunk long enough in a branch as they diverge increasingly to get the work done. do this along several fronts of change and it’s just insane.
and so shorter release cycles with no allowance for variance tend to result in smaller sets of changes. this is not inherently a bad thing, but it does put a damper on larger changes.
while i’m all for regular releases, and i’m quite happy how well kde did in both the kde2 and kde3 series, sometimes it’s necessary to step back and allow the time necessary to get big things done in the guts.
adding features here and there is all well and good, but that’s a short-to-mid-term strategy. practiced long term with invariance it will lead to a certain flavour of stagnation, a sort of convergence on a limit (of the math sort) inherent in the frameworks (aka ‘a rut’). this is great if you have a near perfect framework. it’s not so hot if you don’t; and, trust me, there are no near perfect frameworks
i do think this is partly why there are issues getting gnome3 going and why some gnomers are suggesting in their blog things as radical encouraging someone/anyone to come along and purposefully fork gtk+ to get things rolling.
disclaimers/explainers: i could, also, be completely wrong on this. i don’t think so, but hey, i’ve been wrong before. i have shared my feelings on this with people in the gnome project while sitting around with them at conferences, so this isn’t just me sniping at someone from the safety on the intarweb.
ok, all that said: congrats to everyone involved with 2.18 on the release =) i look forward to seeing it in action.
there’s a huge number of fundamental changes that really, really difficult to accomplish in a series of quick release cycles
That’s why Gnome is more suited for these like it is. Because Gnome is so modular, these kind of changes are actually possible, through the course of several releases.
This has already been done in several Gnome modules. But in KDE, it would not be as easily possible, even insane.
i do think this is partly why there are issues getting gnome3 going and why some gnomers are suggesting in their blog things as radical encouraging someone/anyone to come along and purposefully fork gtk+ to get things rolling
This is pure BS. There just is no incentive to go Gnome3, as nobody know what it is, and it’s not even needed.
I wonder why you people want to change sth that works perfectly well. The truth is that the big changes you see in KDE4 are perfectly possible to do (and happening) in Gnome 2, despite what you say. And that’s thanks to its modular concept. The switch to Cairo was easy, the switch from Bonobo/ORBit2 to DBus is smooth too.
These two kind of changes alone require an overhaul of KDE. Because KDE needs to do this, doesn’t mean Gnome have to, that’s just a fallacy.
Some people were ranting that the move to Cairo made things slower, which was true, but a necessary step to ensure a smooth transition. That’s the amount of problems that arise with current Gnome framework, to make a major change. Now it’s very fast already, and all the naysayers that said Cairo would never be optimised were proved wrong.
i have shared my feelings on this with people in the gnome project while sitting around with them at conferences, so this isn’t just me sniping at someone from the safety on the intarweb
And yet, this is plain FUD. Fortunately, it’s far from being the truth, as facts show different things.
Edited 2007-04-05 09:19
Now I love aseigo, but it seems to me you are right. Gnome DID change some serious core stuff over the last few years. I wonder, though, at what costs… Can you still compile a gnome 2.2 application on 2.18, or did they just dump backwards compatibility?
And if they didn’t, isn’t there a huge amount of duplication, with frameworks from the 2.2 time and and the new ones available? wasting disk space, memory and maintainer time?
> Because Gnome is so modular, these kind of changes
> are actually possible, through the course of
> several releases
i would agree except that it’s not a matter of the number of releases or of modularity. if the releases were not hammered down to a fixed time span that would help a lot, imho. if there was actually a bit more true integration effort across apps it would be easier to float new concepts from beneath to the benefit of all apps. the gnome project seems fairly resilient to both of these things, however.
as it is, gnome tends to get new features and new apps on top of the stack of the sort that fit nicely into such release cycles. it’s a matter of what is encouraged by the dynamics of the release process and software infrastructure.
honestly, if that’s all that is needed and desired then the system is just great. and for many window managers and “old school” desktop environments it is exactly what is needed and desired. is it what’s needed for gnome?
now, the original postulate was that given enough 6 month release cycles, you get the equivalent of one longer cycle. i think the theory of it can be be argued, but the reality of it is another thing given the incremental creep of gnome’s last several releases.
> There just is no incentive to go Gnome3, as nobody
> know what it is, and it’s not even needed.
i suppose it depends on who you ask. reading planet gnome, there’s not a clear consensus on that matter at all. whether you call it “gnome3” or “gnome 2 on steroids” hardly matters, from havoc’s recent suggestion of an online desktop to villa’s occassional “hey, try something radical!” blogs to the suggestion someone should “beryl” gtk+, the sentiment is there.
there’s a very real inertia towards maintaining the present course. which is -great- for the present course: it means it continues to be refined and what not release to release.
what gets thrown under the wheels in such a situation is innovative leaps.
one gnome developer (who you’d recognize by name) explained it to me this way while waiting in an airport: “gnome is being cde’d.” which is to say, conservativized and kept as a platform for Liux/UNIX applications such that eventually we wake up 6-10 years hence and realize that it’s mummified.
this concerns me. but then i’m hoping for a continuously relevant free software desktop that competes with, or better yet, defines the state of the art. i’m less interested in something that is good enough for today but which results in another CDE in the future.
i don’t think it needs to be an either/or situation. i do worry about the longevity of the top free software desktop projects; “worry” in the sense that i care about them and consider the issues at length, even regarding the ones i’m not directly contributing to. this is because i’m here for the free software in general, not to mention we all share an ecosystem and thereby each other’s successes and failures.
i think mozilla has a great future; i’m a bit concerned about the long term realities of openoffice, but that has been eased with the advent of ODF as a standard; i think x.org is finally on its feet again in a meaningful way … in general we (all of us involved in the free software desktop) are on an upward swing. this is good; i do think it’s valid and even desirable that we can observe each other and say things plainly when there are risks, dangers or even outright mistakes being made.
there’s been a lot of feedback regarding kde 4.0 as well, much of it constructively critical rather than simply optimistically brazen. that has been healthy.
> These two kind of changes alone require an overhaul
> of KDE
actually, we use d-bus in kde3, and in much the same way gnome2 uses d-bus. so no, it didn’t take an overhaul.
what we didn’t do was replace dcop in kde3 out of respect for the user base and developer community who had large investments in dcop. changing that sort of thing on people midstream is not good stewardship of a platform.
with kde4 (and any such major number change), we grant ourselves the permission to break compatibility in fairly radical ways. so that was when we swapped out dcop wholesale for d-bus.
but we’d been using d-bus for various tasks for quite some time prior to that.
respect for compatibility is why a lot of changes were “sandbagged” for kde4. i’m not sure if it’s simply that people don’t rely on architectural features of gnome or if there isn’t the same value put on preserving those investments within their community, but for kde this sort of discipline is a very real issue and one that wins us a lot of friends.
> The switch to Cairo was easy
as is any “switch” to a qt with different rendering code paths. i think you’re confusing apples with oranges.
the rest of your cairo “point” is a straw man argument.
> Because KDE needs to do this, doesn’t mean Gnome
> have to, that’s just a fallacy.
oh, of course gnome doesn’t -have- to. but the current path does have certain consequences in addition to its benefits. it’s not a zero sum game. but to say “oh, there are no consequences” particularly in the face of the reality of development is a bit … odd.
honestly, i hope gnome developers sit up and say, “that’s bullshit!” and then take to the streets, er, svn and prove that a 6 month fixed cycle and loose integration doesn’t optimize towards a conservative product. that would be the best outcome, even it would prove me wrong
I see I’ve been modded down severely for a comment most people apparently has misunderstood.
We all know it is no news that Gnome is improved incrementally and my post should be seen as critique of the critique of Gnome being improved incrementally and not as a critique of Gnome or Gnome being improved incrementally.
Obviously people didn’t get that part
What is more interesting is that Gnome after several versions with removal of functionality in order to be HIG compliant has begun to add in functionality in order to satisfy power users. And at the same time KDE has become cleaner with a better organized GUI in order to satisfy people asking for a clean power-user oriented DE. This is especially true for KDE4 which is basically Gnome on QT4 – or a Gnome HIG compliant KDE
(yes, I know it is blasphemy but whatever).
Seems like Gnome and KDE is converging to some extent – which is no surprise after all.
EDIT: Fixed some typos.
Edited 2007-04-04 12:32
Dare I say that the Gnome-i-fying of KDE is exactly what I’ve been waiting for. There are many things in KDE that make me go WTF? right now, but if they clean up their interfaces so that they are simpler, yet just as powerful, then I’ll give it another chance (I try out KDE every point release and while it has gotten better, it’s still a big tangled mess in a lot of areas. Gnome on the other hand can usually be a bit too simple, though I like that as well. If there were a happy medium between the two it’d be the best DE ever.
And I might add that the KDE-i-fying of Gnome is exactly what I’ve been waiting for
There’s been quite a few WTF-situations in Gnome – especially when migrating from Gnome 2.6 to Gnome 2.8.
Personally I consider KDE4 as very interesting. It could be the one to make me switch from Gnome. Unless of course Etoilé has reached a truly useful level the coming fall
This is an oft heard complaint, but seriously: I can’t think of a single feature that’s been removed for the past five years. Admittedly there were both deliberate and not so deliberate regressions in v2.0, but that version was released in June 2002. Please, I’m curious: which features have been removed since v2.0?
The GUI mime type editor, the GUI MetaCity titlebar-button editor (you have to use GConf at the moment – quite suboptimal) and the former menueditor (editing through Nautilus). We got recently “smeg” as a replacement, but for quite a while we had no editor.
It’s not so much functionality that has been removed. What has been removed is easy access. Some of this has been re-added to Gnome, though “smeg” is a suboptimal solution in many situations (why not just right click on the menu item and choose “preferences”? – but oh no, can’t do).
Ugh, the old mime type editor was hideous. It never felt like it worked right (for me anyway, your mileage may vary). What functionality do you feel is missing in the new system?
AFAIK, Metacity has never had a graphical titlebar-button editor — are you perhaps thinking of a third-party add-in supplied by your distro?
The GUI Mime type editor was pretty bad, but the GUI Metacity title-button editor was only for switching the buttons from the right side to the left if I recall. Hardly an important option to have right there in everyone’s faces.
Alacarte is the official new menu editor, and you can just right click on the Applications / Places / System menu and select “edit menu”
Is that easy enough for you?
Even easier, to launch Alacarte, the menu editor, you just have to right click the applications menu (IIRC, the first one), instead of left click.
Then, you have several options, one of them being “Edit menu” or sth like that in english.
“Windows Vista Shows an Incremental Improvement”
I mean really, besides the annoying UAC and the new interface and a few minor things like dropping “My” from Documents, Computer, etc. it really is just the same old XP. (and before you argue whether or not UAC is just plain annoying or it’s the same as sudo on Ubuntu, try creating a folder outside of your home directory. You are literally prompted four times! First for creating the folder, then you have to enter the password (at least if you created an account besides Administrator) then it creates “New Folder” and has an entry field to input the name you actually want, then it prompts you again, then asks for the Administrator password again!)
Gnome 2.18 has been a beautiful construct. From what I have seen though, Gnome 2.20 is where the good stuff is going to be. One example of this is the Conduit project for doing file synchronizations.
http://www.conduit-project.org/
Also someone else mentioned the Gnome-scanner utility. It’s working fine here on my Feisty Fawn install. Though since I happen to have one of the series of HP PSC all-in-ones, I had to find a newer hplip package for my scanner to work (Hopefully Feisty will have the 1.7.3 for final release, since it fixes a LOT of bugs)
Gnome simply rocks and is getting better all the time. I do hope they finally sort a lot of the idiocy behind the screensaver fiasco though. For the most part I gave up and turned off my screensavers, though it is always nice to have at your work computer, since there are a few things that usually impress people, Screensavers, Wallpapers, and sometimes the general theme.
the last time i used kde you could set something in one spot then set something else in another to undo the first. you had to get the order just right. I still like kde thought you can kill hours playing with it.
I had gnome 2.16 crashed and all desktop icons disappeared after I have installed xwWidgets 2.8.3 and I don’t think that this is realeted to the crash.
System is UPS protected and uses the brand new RHEL 5.
It reminded me with the problems “explorer” was showing in windows xp and vista.
Is it so difficult to produce a rock solid GUI?!
Anyway since then my azureus client started to shutdown for no appearent reason, going to crash log files of it showed nothing, probably the crash caused azureus to shut down before allowing it to log the cause of this strange behavior.
Java and Azureus didn’t make for me any problems till the moment of first GNOME crash.
“””
I had gnome 2.16 crashed and all desktop icons disappeared after I have installed xwWidgets 2.8.3
“””
“””
It reminded me with the problems “explorer” was showing in windows xp and vista.
“””
“””
since then my azureus client started to shutdown for no appearent reason
“””
Make sure that the processor fan turns freely. Run memtest86. DON’T OVERCLOCK YOUR CPU. And file a bug report at bugzilla.redhat.com, if appropriate.
No Gnome crashes on any of my machines for years.
Edited 2007-04-04 15:34
I never overclock any of my systems and the memory I use are heat sinked and both (dual channel) are set to Auto timing (the default) from the BIOS; No CPU overheating too 38-44 oC.
Same here, I have Corsair memory that have the heat sinks on them and the bios had set the auto timing, but it was wrong and set too high, as soon as I raised the number from 2.5 to 3, all my stability problems went away.
I never overclock any of my systems and the memory I use are heat sinked and both (dual channel) are set to Auto timing (the default) from the BIOS; No CPU overheating too 38-44 oC.
Neither do I, but you can’t always trust your motherboard timings. My motherboard set the Trfc to 75 ns but the RAM specifics indicate that the modules need 105 ns; I don’t know if the problem is in the motherboard bios or in the SPD chip. Then last year I had a DDR2 module who advertised 667 MHz but didn’t work (random gcc failures) at more than 533 MHz.
As the other poster said. Also check your memory timings in your Bios. I was having crashes left and right after I had reset my BIOS to defaults because of a bad overclock. The Memory timings it had set were too high and after I fixed them, my system became rock solid, and I’ve been using Feisty Fawn for months and wondered why it had been the buggiest release yet, but it was due to my memory timings.
I’ve been testing Feisty since Herd 5 and I would agree with the incremental improvement point which is kind of at odds with the inclusion of “desktop effects”.
What I mean to say is that the core of Gnome has been and continues to be rock solid. So I am kind of disappointed that Desktop effects is even implemented in Ubuntu Feisty. I would much rather see a drastic improvement to the font anti aliasing than layering on the gee whiz of wobbly windows, etc.
This is not a Gnome only issue, KDE 4 portends to be beautiful but unless we start seeing polish in fonts, its just going to be the beautiful graphics and environment capabilities next to the ugly sister fonts.
It detracts from the perceived polish of Linux in general.
“””
I would much rather see a drastic improvement to the font anti aliasing than layering on the gee whiz of wobbly windows, etc.
This is not a Gnome only issue, KDE 4 portends to be beautiful but unless we start seeing polish in fonts, its just going to be the beautiful graphics and environment capabilities next to the ugly sister fonts.
“””
Am I the only one in the world who is perfectly happy with my desktop fonts?
I hear this bit about fonts being crappy over and over and I’m always quite puzzled. Because my Ubuntu, Fedora, and CentOS desktops (all Gnome) have had great fonts for years.
Is there some popular distro out there that has crappy font handling?
Edited 2007-04-04 17:48
Actually Fedora has suboptimal font rendering as default. Replacing the official FreeType2 EPM’s with unofficial RPM’s with BCI enabled will fix it.
However the fonts themselves are quite good.
EDIT: I might add that the difference isn’t particularly visible for many fonts. It is most visible for Century Gothic (or Avantgarde) in smaller fontsizes with “hinting” off. (A misnomer btw.)
I recall the letter combination “liti” rendering as a big blurp for Century Gothic Bold, 10 point.
Edited 2007-04-04 18:05
In what way are they substandard? I don’t have a Fedora desktop in front of me right now, but last I looked, I decided that i could not imagine the font rendering looking any better than it did.
Looking very closely at my Feisty Beta desktop right now, I cannot imagine what I might criticize if I were called upon to do so. Rendering *speed* maybe. But not rendering *quality*.
Since I don’t know how your system is configured I cannot tell for sure.
But the auto-hinter results in less wellshaped glyphs than the patented BCI.
http://www.freetype.org/patents.html
The latest freetype 2.3.3 is out, with ftdiff.
ftdiff allows you to compare objectively the fonts hinting.
But all this “fonts looks like crap on Linux” is BS, and mostly nitpicking, even with the auto-hinter.
ftdiff will show that.
Font rendering is definitely good on Linux. Depending on the actual configuration it can be as good as Mac OS X – or if you want it – more like Windoes 98 (or Windows 3.x) if that is your preference. Choose as you like
Personally I have a hard time seeing the difference between the autohinter and the patented BCI. I can see the difference when using Century Gothic Bold in small fontsizes, or when using low resolutions. But apart from that it requires an A-B test with images side-by-side. The difference is small for most fonts in most situations.
Using good quality fonts with autohinter gives better results than crappy fonts using the patented BCI. That is – if we are talking about TrueType outlines. It doesn’t matter for PostScript (Type 1/3/42 and OpenType CFF) fonts since they don’t contain such bytecode (one of several reasons PS is superior IMHO).
I have since researched the issue and found the Debian and CentOS both have BCI enabled – which may explain why you have not had any issues. Based on instructions provided at Ubuntu forums, I have enabled BCI and now my founds in Ubuntu Feisty Beta 1 are flawless. Gnome is gorgeous. I love it.
http://ubuntuforums.org/showthread.php?t=237383&highlight=bci
That should answer that one. It should be enabled by default in all Debian distributions.
But there are still problems which should really be solved. For example:
– kill your X server and change to a virtual console. Then you can see Gnome processes still be in memory, like bonobo-activation-server, evolution-data-server, …
– the systray has always had problems showing KDE systray icons. Even with Gnome 2.18 sometimes some KDE icons are not shown in systray but in a window. That happens for example when loading Kontact, Klipper or sth. like that at Gnome startup. In KDE I had never problems with systray and Gnome applications.
Maybe small things but they are annoying…
But I prefer having a clean interface in Gnome otherwise I would prefer KDE cause it seems more reliable/matured in my point of view.
> Evince will open a document at the last page you were viewing, but it doesn’t allow you to create bookmarks throughout the document. Evince isn’t alone here — I haven’t found a free software PDF viewer that does do bookmarks — but I don’t understand why this obvious feature hasn’t been included in any of them.
Ehm… Kpdf2 can do it (kdemod on archlinux ships it by default), and so does Okular, the successor of kpdf in KDE 4.