Slashdot Mirror
Linus Switches From KDE To Gnome
An anonymous reader writes "In a recent Computerworld interview, Linus revealed that he's switched to Gnome — this despite launching a heavily critical broadside against Gnome just a few years ago. His reason? He thinks KDE 4 is a 'disaster.' Although it's improved recently, he'll find many who agree with this prognosis, and KDE 4 can be painful to use." There's quite a bit of interesting stuff in this interview, besides, regarding the current state of Linux development.
I used to use KDE 3 (Kubuntu) and I, somewhat recently, installed the latest version of Kubuntu with KDE 4. To be as clear as possible: KDE 4 is a trainwreck. At first I took it in stride and figured that a brand new release might be a little buggy, no harm. I'm using KDE 4.2 RC1 now and it's still horrible.
Yes, I am aware of and use gconf. And it helps some, but there's still some bone-headed design decision. My favorite, of course, is how they made it so that cursor blinking is a global setting. It doesn't matter if you use gconf or not, either your cursor blinks everywhere, including the terminal, or it blinks nowhere. That is, neither setting is acceptable.
I used to be a KDE user. I thought KDE 4.0 was such a disaster I switched to GNOME. I hate the fact that my right button doesn't do what I want it to do. But the whole "break everything" model is painful for users and they can choose to use something else.
I realise the reason for the 4.0 release, but I think they did it badly. They did so may changes it was a half-baked release. It may turn out to be the right decision in the end and I will re-try KDE, but I suspect I'm not the only person they lost.
I got the update through Fedora and there was a mismatch from KDE 3 to KDE 4.0. The desktop was not as functional and it was just a bad experience for me. I'll revisit it when I reinstall the next machine which tends to be every six to eight months.
Which isn't exactly the same thing, and probably not many people at KDE will be all that surprised. KDE4 is new, it has teething problems. It was risk, but we'll find out later if it was a risk worth taking.
KDE 4.0, and to a lesser degree 4.1, lacked quite a few nice customization features that KDE has had for the longest time. KDE 4.2 refixes the taskbar configuration...so you can actually do something useful with it again.
KDE 4.0 and 4.1 are nowhere near as functional or customizable as 3.5, 4.2 restores virtually all of it as well as adding compelling new standard/addon features.
4.0 was supposedly 'just a developer preview', and I personally think they dropped the ball on 4.1. Everyone was expecting it to just be 'ready'.
Though, one begs the question.
If Linus is an advanced user, why was he pressured to upgrade from 3.5 to 4.x in the first place? Couldn't he have just kept using 3.5 if that's what he preferred, rather than the GNOME which he hated?
I know the 'user friendly' distros tend to be a bit aggressive about pre-planned obsolescence, but that's little excuse not to find a supported and proper way to use the software and specific versions you prefer.
"A Goddess rarely smiles for she is forced by others to be an island unto herself." - Zephiris
I actually got sick of having to learn and spend time on every new thing, and having people laugh at me while I was reading the iwconfig manpage, while they just clicked a menu on OS X and joined a wireless network.
With Kubuntu Hardy and KDE3, the joke was on them. Everything Just Worked, out of the box, with far more configurability than just about anything else. All those "extra applications" includes things like wifi, bluetooth, sound, and USB mass storage hotplug, as simple, intuitive GUIs that require no more learning than "Let's try right-clicking the Bluetooth icon... Oh, I get how this works."
Then came Kubuntu Intrepid and KDE4. Bluetooth didn't work. Wireless became more complex, and no longer uses kdewallet to store things. I'm taking a long, hard look at things like xfce, fluxbox, or just rolling my own.
Don't thank God, thank a doctor!
I've used KDE 4.1 for a month. Then I switched back to KDE 3.5, and was the happiest guy in the world to have my good old desktop back! I use Kate a lot for programming C++ and Actionscript projects. In KDE 3.5, Kate rocks. In KDE 4.1, they have, on purpose (by design) ruined the search function of Kate (no whole word option, it doesn't search for the same word in the different open documents), making it unusable for programming (especially refactoring). They have totally made the file managers unusable. No proper working tree. Konqueror can have a tree, but it has the most annoying horizontal autoscroll thing ever (again by design), and you can't drag anything to it. The unzip tool (Ark) is a joke (I've never seen it working). No possibility to have two rows in your taskbar. I *need* to have one row that acts as quick launch for programs, and another row that has the buttons of open windows, one for every window, and only the windows on the current desktop of the multi desktops. Terribly annoying behaviour in file managers and file open/save dialogs, it's so extremely hard, almost an annoying computer game, to select multiple files. Anything from dragging a rectangle around multiple files, to using ctrl + clicking, are all not working properly due to various reasons (such as when beginning to drag the rectangle, it thinks you want to drag 1 file, instead of dragging something around rectangles). Filenames in such lists are clickable everywhere, instead of only on the text of the name, and are in a very wide column by default, which is a second cause for making it hard to drag a rectangle around multiple files. The non-SVG cards in the card games are rescaled in a terribly ugly way, and the SVG card decks all have an ugly design.
But the productivity loss with kate and the file managers is still the worse of all, KDE has become unproductive as hell for me, and I use KDE 3.5 as long as possible.
There is actually one tool I use alot is Ubuntu Tweak which fixes some of the gripes that we Gnome users complain about.
Features of Ubuntu Tweak
* View of Basic System Information(Distribution, Kernel, CPU, Memory, etc.)
* GNOME Session Control
* Auto Start Program Control
* Show/Hide and Change Splash screen
* Show/Hide desktop icons or Mounted Volumes
* Show/Hide/Rename Computer, Home, Trash icon or Network icon
* Tweak Metacity Window Managerâ(TM)s Style and Behavior
* Compiz Fusion settings, Screen Edge Settings, Window Effects Settings, Menu Effect Settins
* GNOME Panel Settings
* Nautilus Settings
* Advanced Power Management Settings
* System Security Settings
http://ubuntu-tweak.com/
Well, he's using fedora and it doesn't have kde3 since F9.
I'm also using Fedora for several years. I've been a die hard kde fan, but switched to gnome after I had enough with kde4 as well.
The kde developers said you can always stick with kde3, but truth be told you can't.
Not to mention that kde 3.x hadn't been properly maintained since they started working on kde4, with latter updates sometimes braking things that were working fine previously.
Is is cynical to inherently distrust a Microsoft Web site called "getthefacts.com"?
We experience the opposite. We are 4:1 a Linux:Windows shop, yet spend twice as much time fixing Windows boxes. We charge for our time, so we love, and hate Windows.
Linus is promoting the best option available at the time, without bias. Which is perfectly sane, and valid.
In all fairness, "best" is one of those things that is in the eye of the beholder. When KDE 3.5 was the latest, GNOME was still "the best" for many people.
Basically, KDE has great tech. BUT core developers seem to have some sort of arrogance about listening to the community
Please elaborate, without using mailing list threads where these core developers get flamed endlessly because people don't like something in KDE 4. On the other hand we are always interested in receiving reports of what we could do better (although reasons of "KDE 3.5 did it this way" does not exactly prove the point...)
and some sort of project-deathwish which manifests in a horrible release process,
Horrible?... How so? I ask because the release process is mostly unchanged since KDE 3.5, where apparently it worked well. What do you think has regressed since then?
minor versions that don't work until x.4 or so,
So you're saying that you've had issues for both 4.0 and 4.1 not working until 4.x.4? 4.1 would have been much the same as 4.0.4, with the exception of extra features. I personally did not notice tons of trouble from 4.1 on (although obviously I'm biased ;)
and poor support for non-core developers.
No offense but this is a troll unless you have something in particular that you're talking about. The same mailing lists, API documentation, and support tools are available now as were available for KDE 3.5. In addition we now have a Wiki available instead of the crusty old KDE 2.x material, KDE TechBase, and the number of developers has only been increasing.
For instance, the latest KDE Commit Digest shows commits by 249 developers, up from 231 a year before. If we go back to the last Commit Digest from Derek Kite in October 2005 there were 195 developers. Argue about seasonal effects or whatever all you want but the data doesn't support your argument.
Moreover they've alienated some of the very groups they tried to encourage early in the KDE 4 brainstorming process.
Well there are definitely "alienated groups" but who are you talking about specifically?
Finally, they generally seem to suffer from lack of manpower, which they have never really tried to solve.
Well not only is that not true as I already mentioned, but your latter point is also not true. I know it's easy to blame the shift of focus that we employed in KDE 4 on everything, but the fact of the matter is that it actually brought in quite a few developers as well... We have people working on the art, basic desktop and games, areas which were mostly unmaintained in KDE 3. Things like the KDE TechBase I already mentioned were created as part of making it easy to develop for KDE. Again though, if you have something specifically that you have in mind then say so as developer support is a very high priority for KDE.
If you believed the hype the core devs were spouting, KDE 4 was going well, and no help was needed, until the product actually appeared as a release and everyone saw the real situation.
Here's the announcement about the Development platform release where the library API was declared stable. "With a lot of issues facing KDE hackers before 4.0 is a usable desktop, all work on new features and UI is stopped, and efforts focus on fixing the inevitable, long list of bugs." Where's the hype?
Here's the Plas
If Linus is an advanced user, why was he pressured to upgrade from 3.5 to 4.x in the first place? Couldn't he have just kept using 3.5 if that's what he preferred, rather than the GNOME which he hated?
Because he is running Fedora, the alpha/beta (depending on the release, but not planned in advance) for RHEL. Note that he did not complain that Fedora did this. When Red Hat Software announced that free Red Hat was going away and being replaced with Fedora, they explained that it would be bleeding-edge, and provide a testing ground for new technologies before they made it into RHEL. Linus simply said that he's not using KDE because Fedora pushed it at him, he wants to run Fedora ("for historical reasons") and that since KDE4 wasn't ready for use, he's using GNOME instead.
It's worth mentioning however that Kubuntu did this too. Fedora is supposed to be beta quality software. Kubuntu is supposed to be release quality software. But my perception (based on quite a bit of use across both desktops and servers) is that Ubuntu has been sacrificing quality for an aggressive release schedule and new features with every release.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
MS paid a company to license it's zip technology, and it's very basic.
-]Phreak Out[-
KDE 4 is unfinished. It says everywhere in the official sources. Since KDE 2 the .5 releases basically where the stable targets. It's only with 4 that with the .0 release they didn't care about finish at all, and thus provides Über-suckage. 4.5 will be the stable finished 4 release. No news here. What's the big fat hairy deal?
That said, KDE 3.5 still kicks Gnomes ass usability and integration wise. However - and this *is* true - Gnome has actually stopped sucking in 2008. For the first time in history Nautilus is usable also for non-total-fanboys, and allthough the featureset and power is no where near that of Konqueror, it also has become intuitive to use. For the first time ever since I moved from Debian, I'm using Ubuntu instead of Kubuntu (also due to the flak KDE 4 has gotten) and for the first time I didn't remove it after 10 minutes.
It is far away from the KDE featureset and I'm still convinced that a well configured KDE 3.5 is the best desktop in the world and also outperforms Mac OS X usability wise (fyi: I'm typing this on a mac), but the Ubuntu foundation work done on my Dell Volstro is so awesome, I don't really care that much about any nitpicky details. Maybe I'll dick around with E or something if I get bored by it. Both Gnome and KDE are so far beyond Windows - which I use at work - that it doesn't really matter that much to me. Especially with the improvement Gnome obviously has seen lately.
Since Linus actually cares squat about the Desktop, as long as it works, his statements actually make sence in current context. No surprise here either.
We suffer more in our imagination than in reality. - Seneca
Common misconception. OS X doesn't use PDF for screen drawing. Its graphics engine (Quartz) can generate PDF data with the same commands it uses for screen drawing, and has built in support for PDF rendering. And PDF is used commonly throughout the system to store things like vector graphics elements and documents sitting in the print queue. But things that are drawn on screen by Quartz commands don't exist as PDF data anywhere along the way.
Pre-X versions of Mac OS didn't have PDF integrated at all.
NeXTSTEP did use PostScript (officially licensed from Adobe) for on-screen drawing, but PDF and PostScript are very different. PDF is data and PostScript is executable code, so the notion of doing on-screen drawing with PostScript code makes some sense, while the notion of doing on-screen drawing by generating PDF data and then rendering it doesn't make a whole lot of sense.
Anyway, PDF licensing is royalty-free.
This space unintentionally left unblank.
I'm using H&R Taxcut this year, probably the only software I still purchase on a regular basis.
Have you tried TaxAct.com? Their $17 "Ultimate" bundle includes both Federal and one State, plus e-filing for both is included! I got so sick of TaxCut and TurboTax becoming more and more expensive each year, and either charging me to e-file or else making me send in a rebate. I've used all three, and TaxAct is as good as the others.
Plus unlike the others they don't play the Vista game of offering multiple versions each with different features that try to get people to pay more out of fear they'll get stuck with the wrong one. It handles self-employment income, capital gains, deductions, etc. all in one package. I had my reservations at first since it costs about 1/3 of the others but I used it last year and was very pleased.
This is incorrect. Quartz models the PDF object graph. It even has the legacy bottom-left coordinate origin in its views which you must flip in your custom view (if you prefer). In the data sense, it does use PDF for screen drawing, which is why a PDF graphics context is available for any view that wishes to render one. It's also why a view can use the same drawing commands to render to a printer as it does to the screen, and why for many years, screenshots you took in OS X created PDF files on the desktop.
From Apple:
"Quartz's feature-rich drawing engine leverages the Portable Document Format (PDF) drawing model and offers Mac OS X applications professional-strength drawing functionality."
Have you ever tried Almeza Multiset? If you have to deal with Windows often it really is a Godsend. With Multiset I only have to install software ONE time on my test machine and it makes a nice software unattended installation. It will remember all the checkboxes I check, all the buttons I push, and all the tweaks I do to the software as I install it. Then I simply burn the software to a CD/DVD and voila! No more sitting there for hours going "clicky clicky next next next". It is WELL worth the money, believe me.
They have a trial on their website. Try it and I bet you'll like it. One piece of advice though. It was designed for single .EXEs, like what you get with FF and Opera. So if you want to install something like MS Office with it simply drag the folders off the office disc into the correct folder during CD/DVD creation. Since it simply drops the contents into a folder for you to burn later this is trivial to do. And since you can choose between having it install all the software on a CD/DVD or picking and choosing what you want you can tailor the install for the client. I cut down my Windows software installs from 3 hours to under 10 minutes with this baby. Trust me, it is REALLY worth having in your toolbox!
ACs don't waste your time replying, your posts are never seen by me.
A PDF object graph only exists if you're drawing into a PDF context, which isn't how most on-screen drawing occurs. If you're drawing into a bitmap or window graphics context, you're using C functions to put pixels on the screen; no PDF data exists anywhere along the way.
See here. That doesn't describe a system for drawing to the screen by creating objects in a PDF object graph.
Those PDF files simply contained (and contain; you can still enable saving screen shots in PDF) a single bitmap image of the entire screen. If the contents of the screen were represented in memory as PDF data, one would instead expect to see those files contain separate bitmaps for each bitmap displayed on the screen, vector data for text and shapes drawn via Quartz, etc.
This space unintentionally left unblank.