Slashdot Mirror
Has GNOME Become LAME?
auferstehung writes "Nicholas Petreley (should that be KNicholas KPetreley) of LinuxWorld and VarLinux.org has taken his gloves off in the latest article in his KDE vs Gnome series. An unabashed KDE supporter, Petreley uses some choice fighting words in re-acronymizing GNOME as the Language Agnostic Morphable Environment
(LAME) Franken-GUI. Despite the sensationalistic flamage throughout the article, several of his GNOME criticisms (Gconf, file selector, features) echo those already voiced within the GNOME community itself. A happy GNOME user myself, please someone...tell me it isn't so."
If all of what this article implies is a reasonable "comparison" between the way KDE and Gnome function, why is it that so many prefer Gnome over KDE ?
I've used both for years and have finally settled on Gnome as I find it faster, more intuitive and less "bloated" than KDE, yet the authour of the article finds pretty much the opposite to be true.
I'm no programmer, so what happens behind 'the scenes' is not something I can use to compare the different desktops.
All I know is that I much prefer Gnome over KDE.
A slashdotting - you get the stick first and then the carrot !
I use gnome when I am root. As a user I like KDE. I am using Redhat 8 and I like KDE. I wanted to like gnome, I really did. I have upgraded it and KDE, and yes KDE is HUGE and its bloated big time. But so what, its linux, and its not like its gonna hit a swap file. Linux does what it does well, and on this laptop KDE is running smooth and sweet. I have eye candy, I have apps I use, and they work fine. Gnome just doesnt cut it when it comes to looking good. There are also several things in gnome that bug me. But the best part of all is that it doesnt matter, I also use Evil Entity and have been very intrigued with Enlightenment. When I get around to gentooing E17 as long as I get font antialiasing, I am going there and not turning back. Evil Entity Linux has shown me enlightenment and I am on board. How cool is it to double click the desk top and get a console? Very. How fast is it? Click and GO. So in conclusion KDE is huge, but has a reason to be, gnome? I dont know what your thinking. Its becomming more blurred every day.
As a Windows-only user for nearly my entire life (minus the DOS/Win3.1 years), I was actually looking forward to being forced into using nothing but Linux in my Computer Science classes. Windows has so many things I hate, and I have heard such good things about Linux.
To my disapointment, when I went into the *UNIX side of the computer lab at the start of this semester, I had noticed that they had upgraded from some pure-UNIX OS to Red Hat 8.0. I figured "oh well, *nix is *nix." The first time I logged in to GNOME, my first reaction was, "Holy crap, this is windows!" Then it began to sink in, the GUI is too much like windows. This was good for all the Math Majors who are required to take CS202, but for everyone else--EE and CS Majors--it seems pretty useless to have such a developed GUI that discourages people from learning handy terminal commands (atleast in a timely fashion). Sure the GUI makes it an easy transition for me, but I know that 'an easy transition' isn't going to help me in the long run.
I belong to the "something has gone wrong with Gnome" school of thought. I dearly want Gnome to succeed. It's got a different sort of style and sensibility than KDE and Windows, and there's a lot of great stuff there.
Love it or hate it, KDE feels like a unified desktop, while Gnome feels like a cobbled together set of unrelated tools.
The "Open File" dialog is a thing of shame, and I can't believe that it won't be until October until a replacement comes along. The fact that something so basic has been allowed to stay unchanged so long, in my mind, reflects the difference between KDE and Gnome.
I don't think that it's an organization issue, or even that one group is more clever than the other. My guess is that, at some level, Qt really is better than GTK. I don't know if it's C vs. C++, or KParts vs. Corba, Glade vs. KDevelop... Perhaps Nick's got it right, that it's the underlying objects. KDE doesn't seem to have suffered from having a C++-centric toolkit, and Gnome doesn't seem to have benefitted from having a C-centric toolkit.
The last release of KDE had some pretty cool stuff in it - I was eager to get my hands on it and play with it. In contrast, most of what I've heard about Gnome 2.2 has been about what it doesn't have in it anymore. It apparently won't even be featured in the next Knoppix release, since it's broken so badly.
But I wouldn't discount the future of Gnome. Maybe .NET/Mono will solve the problems (I wouldn't bet the farm on it). Maybe there are no problems at all - just a different desktop, with a different way of doing things.
But, gah... Please, fix that "Open File" dialog!
lets compare applications
Gnome/GTK vs KDE/QT
Web Browsing
Galeon
Galeon is better by default, it supports more since its based on Mozilla...and its fast. Konq's "tabs" are shitty, shitty when you compare galeon's tabs functions, and the speed shits all over konq.
Multi-media
Mplayer (gtk)
Nothing can touch this player, sure theres KDE frontends available, but the default is the GTK frontend, Arpi chose the better GUI toolkit.
IRC
Xchat
I've seen a few KDE IRC clients, they're either heavy bloated with shit (kvirc) or so featureless that its not usable.
News
Pan
Knode is trash, sorry to say, some attachments never quite work.
Email
Evolution
Kmail, see above...plus it cant thread my messages that well..Sure it looks like Outlook, big deal..Konq looks like IE.
Mame
gxmame or grustibus
You want ugly, QMamecat is *ugly*.
Music
xmms (gtk)
Noatun is so shitty, why do they bother even developing it? Using seperate tools for media playing, mplayer, and xmms for music...why is there a need for Noatun, there isnt. basically KDE try to make a KDE tool for everything, even if it sucks, even it crashes.
the only thing going for KDE is the actual desktop environment and Konq as a file manager, everything else is either gnome and/or GTK has the advantage. They simply offer better programs, more mature programs, less prone to crashing.
http://www.fanboy.co.nz/adblock/
I often swap things from one server through ssh (fish://) to another one with samba (smb://) by drag-dropping from a vertically-splitted konqueror window from my desktop that sits on a third computer.
Do that with builtin Explorer functions.
For that matter, do that with Nautilus, too !
-- don't discount flying pigs until you have good air defense
While it's nice Linux emulates these other OSes for "consistency" or such, why not develop a new "type" of GUI? Remove the desktop metaphor, the icons, the needless menues. Why not a simple GUI with no mouse where you cycle between things you want your computer to do? A circular node-based selection scheme, like the GameCube's OS except instead of moving around a cube you move around a sphere or circle, where the options are chosen by moving left or right in the circle and choosing things like "E-mail," "Write a paper," "Browse the Internet," "Write a spreadsheet," "Install something," or even "Have computer tune itself up" (so that it sounds easy to understand to a normal user, but it does all the stuff they don't care about like defragging the ol HD or updating virus protection - a technical support employee's dream - just name it something that makes it appealing for them to run it).
A friend of mine once said "If you could make something easier to use [than Windows], I'd buy it." Granted, he's not into computers, but the majority of people aren't "computer people" at all - they just want their computer to do what they paid $1000-2000 for it to do. And, being technical support for the freshmen in my hall, I can tell you that NOBODY who isn't into computers ever updates security packages, virus protection, or even software they're using, nor does anybody ever run defrag...
They leave out a lot of imho useful options, while there is an "input methods" menu item for every input box, which you can accidently set to "cyrillic" or "amharic" or ... and fsck up your input box. Yes, I can correct it, but a newbie can't. And yes, these things _do_ happen. A few weeks ago, my mom saw a secretary in a hospital who accidently lost a toolbar. She had to wait until the next day, when the tech guy was around. These things shouldn't be removable by default. Experienced users, the ones who use it, can turn the option on in a matter of seconds, but a lot of people who don't use it can't turn it off.
I switched from slackware to SuSE a long time ago. around SuSE 6.1, so I guess you could say I "grew up" on SuSE and of course because its a Germant Distro KDE was the window manager of choise. This was because the non North American's weren't as caught up in the debate over wether it was politicaly correct to use the non-open Qt libary and therefore spent more time polishing.
On my new SuSE 8.1 tho I notice that the KDE is slow, sometime drifting off into LALA land, and sometimes not coming back I've had actual screen freezes that required pulling the plug and rebooting. Now that I've got a cable broadband connect it's time to do some serious updateing.
Gnome on the other hand seems to work a lot better for me does slowdown occasionaly but hasn't actualy frozen yet,(a lot of the slow downs seem to come from Mozilla and not realy Gnome) I miss the more familar KDE interface Gnome is very spartan in comparison.
One advantage I've really noticed is that Gnome bassed app run a lot better in KDE than KDE app run in Gnome. A good running KDE is a pretty good desktop I've notice that the look and feel of Windows XP seems to be the same as KDE 3, and of course everything in the KDE. I've noticed that when I'm on the wife Windows XP I start to do someting, and have to stop myself because the software isn't in there.
Apocalypse Cancelled, Sorry, No Ticket Refunds
A happy GNOME user myself, please someone...tell me it isn't so.
If you're happy with GNOME, why should it matter what anybody has to say about it? I've never understood the mentality that things in life are deemed "worthy" only if they're popular (pop songs, software, beer, etc.).
Petrely was talking about "clickable things", and I was simply using his terminology.
If [consistency] isn't a big deal to those people, then they won't mind if everything is consistent then, will they?
All things being equal, no, they wouldn't mind. But all things aren't equal: effort you spend on consistency won't be spent on other tasks. And you can't be consistent with everything at the same time.
Remote applications? That's something built into X, not specific to Motif.
As I was saying, KDE developers often don't even understand the issues... remote application usage isn't just about moving graphics from one screen to another, it's making configuration parameters, window management, and communications work correctly. Doing that requires toolkit and application support, but KDE and Qt don't even really try very hard.
Inter-app communication? Like DCOP and MCOP?
Yeah, that's the problem, like DCOP and MCOP.
Care to give an example?
Sure: the Konqueror window I'm looking at has about 50 distinct widgets and manipulable elements in it (buttons, menu entries, labels, scroll bars, icons, etc.), yet it only uses 7 distinct X11 subwindows. The intent of X11 was to use subwindows as a kind of structured graphics for building toolkits, but Qt and Gtk+ just do a whole lot of drawing inside big windows, and they have a whole lot of their own event processing logic. As an analogy, you can create graphics in PostScript by using its drawing commands, or you can create it by embedding a big bitmap. KDE/Qt and Gnome/Gtk+ are effectively doing the latter in a lot of cases.
So I'm giving KDE a try. It has problems too. The most annoying one to me is the way that it switches focus when I use my scrollwheel. It has options for what to do when clicking any of the mouse button (focus, raise, etc) but not the scrollwheel. When you scroll a window that is not focused and not on top, it gives focus to that window but does not raise it. This maybe wouldn't be bad, but then clicking on the window also does not raise it. You have to focus some other window then come back.
Another thing I don't like about KDE is that it is hard to add buttons to launch X (not KDE) applications to the sidebars. In gnome I could add a launcher easily. In KDE I have to add a non KDE app, it gives me a browse dialog. I don't know where my apps are, probably usr/bin/ or usr/local/bin, I don't want to hunt around, so I try to click on my terminal button so I can do a `which app`. The dialog has the sidebar. Doh.
The choices for applets in KDE is very underwhelming. In gnome 1 I was able to put applets for gaim and xmms in my sidebar. They are unobtrusive there and available on all my desktops. It was wonderful. KDE doesn't have these.
Sure the KDE apps all look the same and act the same, but they are not powerful compared to other stuff. I always use Mozilla as my web browser, open office as my word processor, etc. The KDE stuff are nice, but not as full of features. As soon as you add in non-KDE apps, you lose much of this consistency.
>And Gtk+ and Qt both make very inefficient use of >the X11 APIs, giving X11 an undeserved reputation >for being slow.
I don't know about Qt, but recent versions of GTK+ use the XRender extension. This bypasses entirely the regular X11 API. In fact, GTK+ makes VERY efficient use of X.
Give some concrete evidence (e.g. benchmarks) before making wild claims about performance.
Otherwise its just FUD.
The moving cursor writes, and having written, blinks on.
And what's in it for TT?
YOu start with an option where they may make no money at all, yet they have to give you support. Then, once you start to see that you will get money from your app (some of which would go to pay TT)... you switch to the other scheme, and they only get the flat fee!
What you want is TT to take the risk involved in the first scheme, while topping your expense at the fixed cost of the flat fee. No company is going to increase their risk FOR YOU!
Really, what you want is charity. Like the part about "I'd hoped Trolltech might have recouped whatever costs they have and then released Qt under something like the GPL"
Even ignoring for a moment that Qt *is* released under the GPL, why on hell would someone who makes a living selling a product stop selling it just because he has recovered the initial investment???? That makes zero business sense!
It's like saying "hey, I bet Disney has recouped costs from The Lion King, why don't they make it FDL?"
Have you guys seen Slicker?
http://slicker.sourceforge.net/
-john
Slashdot: you'll not find a more wretched collection of villainy and disreputable types...
I think GNOME needs something like Sawfish -- something with useful features rather than just a Microsoft clone. If the GNOME people have gone off Sawfish, that's a shame, because there's nothing else like it.
In my experience Sawfish versions 1.2 and 2.0 are not even ready for beta testing. They crash readily and badly. Don't try them unless you're interested in development.
Sawfish 1.0.1 is fairly solid, and no other free window manager I have heard of comes close in features. It makes it easy to work efficiently. For example, if you maximize, restore and close windows a lot, you can put "Maximize window toggle" and "Delete window safely" on keys easily. You can do the same with about 251 other functions including XMMS controls.
John Harper commits some user interface howlers like the fixed-size "Edit binding" window, but you'll find that sort of thing in all software. Refreshingly, he doesn't readily make assumptions about what features users don't need. Don't want 251 other functions? Don't use them. If you want a particular window manager feature, try Sawfish 1.0.1 first. It's more likely to be there than in any other window manager, and it will probably be easy to use.
The Sawfish list is busy, John Harper is there, and development seems to be going on.
Most of it is written in the author's own personal lisp dialect. One language per developer is a bad principle, but in this case it helped Sawfish become very useful quickly. You seem to suggest that Lisp is the problem. Does it make software hard to maintain?
Metacity is good for Windows users. It's a better default than Sawfish was with that ugly Crux theme and the settings it came with in the old gnome defaults. But it's a shame that there's no longer a modern, sophisticated and efficient window manager in the project.
The article in question is obviously nothing anyone should take seriously. It does not even pretend to be an honest comparison between two windowing environments; it's basically one person's rant about why he prefers one over the other. Good for him, but mentioning only KDE's good points and only Gnome's bad points isn't a useful comparison to anyone else.
I'm surpised at how poorly informed the people who discuss Gnome vs KDE are. No one has mentioned any of the new accomplishments both environments has achieved. It's still all "file selector" this and "configuration options" that. Dudes, I stopped fretting over thing like that years ago. There's plenty of other things that need focus for a good desktop environment, and are being worked on as we speak, but no one has mentioned them in any of the comments I've read.
Anyone here even know about the massive time spent on building a rich and powerful "accessibility toolkit" ATK? Or the very well thought out multimedia framework GStreamer that's currently in development. I've only seen a few mentions of the establishment and accomplishments of freedesktop.org - whose goal is to set standards (such as the HIG) which both Gnome and KDE can follow to achieve consistency and inoperatability. How about the universal adoption of Unicode (using UTF8) throughout so that proper internationalization is finally possible?
These are important things, and much more forward looking than all the nitpicking that's so prevalent in these discussions. These articles and the bickering that ensues are no better than "celebrity tells all" and "other celebrity makes rebuttal" shows on TV. Totally pointless - fun to watch sometimes - but pointless.
One thing I need to add: Most complaints about GConf that I've read are miss-informed. Yes, the closest approximation is the Windows registry. But it was created with the strengths of that registry in mind, and steps taken to get rid of the problems that the registry had. For example, ALL keys are documented. Which is easier? Hand editing a text file, or going down a list of fully documented options in a gui editor - toggling boolean keys, editing strings, etc.
I'm obviously a Gnome user. I know KDE has it's own list of accomplishments, but I don't know them well enough to list. My point is, why the hell are you choosing a desktop environment based on which has a better file selector? There is plenty more to look at.
Loban Amaan Rahman ==> Anagram of ==> Aha! An Abnormal Man!
As a long time GNOME supporter and user, the latest incarnation has completely turned me away from it. I was a very happy Gnome v1.4 user and still run it at work. The problem now is that my favortite apps are starting to convert to v2.2 and soon I'll either have to run v2.2 or stay with obscurity in v1.4 My biggest gripe with v2 is Metacity. And yes, while you CAN switch WM's, Sawfish just doen't work the same as it did in v1.4 After not running KDE since the original v2, I have now returned to it with v3.1 I was truly impressed at how much has changed in it and the improvements are unreal. Sorry boys, but your choice of taking MY choices away in Gnome have caused you to lose yet another supporter.
"Klaatu, verada, necktie!" -Ash
The differences between DOS 2.11 and DOS 3.2 were so slight very few manufacturers at the time made much effort to promote what they had. I think on a technical level, the differences really were limited to internals like hard disk support (sizes of filesystems, etc) and support for 3.5" disks which meant that, at the time DOS 3.2 was around, most people didn't care. DOS 3.3 added the "CALL" command to DOS Batch scripts, which was the first major bit of new functionality.
It wasn't until DOS 4/5/6 that major new functionality was added from a user's perspective, enough to actually mean people wanted to upgrade. The spur for this was competition. Digital Research, Gary Kildall's old stamping ground (not to be confused with Digital/DEC, an entirely unrelated company), created DOS Plus (based on CP/M86 but with the MSDOS APIs added), and then DR DOS, with various new features (the former had preemptive multitasking, though in a somewhat crippled form, the latter a usable file editor and other nifty add-ons)
The "Everything works for the third version" thing doesn't really apply to Windows either. Ignoring point releases, Microsoft released at least four versions of Windows before 3.0 (Windows 1, Windows 2, Windows 286, and Windows 386, the latter of which was supposedly better than Windows 3.0 when it finally came out, according to people who used it.) 3.0 still languished, it was the first version after Microsoft started a campaign to have it actively bundled with DOS (earlier versions were occasionally, but manufacturers rarely did because it cost more. Microsoft changed the DOS pricing and put pressure on manufacturers to ship 3.0), but few machines ran it well, and most people who used it at all used it for Solitaire... 3.0 had major problems with speed, with screen refreshes, and with the DOS compatability box, which were all fixed in 3.1.
3.1 was the first version of Windows that took off. By that, I mean people started actively using it at that point. This happened because all the pieces came together with it - PCs just about hit the point where they were fast enough to run it, and 3.1 was the first usable version of the 3.0 series.
BTW, I rather liked Word 2. I thought Word 6 was a bloated pile of junk... Oh well ;-) Too many people disagree with me for me to be able to disagree with you on that one, but I think it's much underrated.
You are not alone. This is not normal. None of this is normal.
Both GNOME and KDE suck wind. Arguing one over the other is like arguing wich is better, a broken leg or a broken arm.
CDE is even better than either of them. If you want something that really works look at Xfce. The current "production" version is xfce3 and it can do everything KDE & GNOME can do, and much more. It also is very nice on system resources. It runs as light as BlackBox or IceWM and is just as fast. And the development version of xfce4 will blow your mind. It'll make you cry it's so good.
The fact is that GNOME and KDE are, functionaly and from a usability standpoint, damn near identical. Under the hood they are vastly different but for a regular user they are interchangable. Bluecurve proved that.
--
If I actually could spell I'd have spelled it right in the first place.
I want to know why the Gnome developers decided that control-A should "select all" instead of "move to beginning of line" like it does in almost every other (read: emacs editing command-compliant) X application. I can't seem to change it anywhere, even in gconf-editor.
These are the sorts of changes that make me, as an experienced Unix user, want to look elsewhere. I personally grow tired of the drive to "simplify, man!" and yearn for the days of configurable sawfish and a galeon with 1,000,000 options in the preferences dialog.
Anyone have any suggestions as to where I should look? I'm completely open.
[ home ]
That's a bit of a straw man argument don't you think? You're putting up a fragile target and then tearing it down. If I had a bunch of key/value pairs like that, why wouldn't I instead use:
...and so on for hundreds of entries. AND(!) I can add hierarchy at any point if necessary whereas the key/value pair file must be completely reworked.
As for coding difficulty, using the DOM in C++, the code looks something like this:
Assuming you are using libxml++ of course. Personally I prefer using other programming languages, but the code is very similar no matter the language. Let me say that again because it's important: "The code is very similar no matter the language."
And with XML, you get i18n for free. Can your program read Cyrillic characters? What about Chinese characters? Does it handle accent marks correctly? Or are you using ASCII with no contingency for i18n? With XML, it's a no-brainer. If you don't need it, don't use it. If you do need it, it makes recoding much much faster. UTF-8 you say? That works in most cases, it's true, but it doesn't address the other issues of a config file.
With your own config file format, you must handle comments, multi-line values, conversions from this format to the next, incompatible version. All of these are easier when using an XML parser. But you're a 1337 coder. Don't let me stop you. If you like spending time on config files, go right ahead. I prefer to spend time on the rest of the application.
- I don't need to go outside, my CRT tan'll do me just fine.
Actualy, using
:)
<FullScreen>true</FullScreen>
is almost as amateur as using
[TRUE]
FullScreen
UseJoystick
UseOpenGL
You can always find the worst way to do the things, if you're creative enough.
You know, it's not even all about people like your Dad -- people like me want stuff to Just Work too.
I'm a Unix sysadmin by trade. (Mostly Solaris, but now some HP-UX as well) At work, I do nothing _but_ futz with configuration files and tweak things and make it go --- not so much because things are busted, but because I'm trying to make them run _better_. Things like setting the sd_max_throttle setting in Solaris, or various ndd commands to disable source-routed frames and so on
Anyway, when I get home at the end of the day, the LAST DAMNED THING ON EARTH I want to be doing is tweaking config files, downloading Widgetx 0.234567 and finding out it's incompatible with the libc libraries in my OS, so I have to upgrade to glibc and now Widgetx still doesn't work because glibc broke my window manager.
I remember those days --- the libc-to-glibc migration
March 24, 2001 I bought my first Mac. And I haven't looked back since
Now stuff Just Works. I come home and I can read my email, play games, connect to work, and even watch DVD's and edit home movies and play with my digital camera
Some folks can't see the forest from the trees --- The end-goal here should be for the system to just GET OUT OF MY WAY and let me work
How funny you are. The typo in your example "valuw2" really illustrates exactly why XML is a good thing.
You didn't include all the custom code that you have to write to validate that the numbers on the right side of the equals signs actually define valid settings for the variables. XML Schemas make this very easy.
Also, suppose I'm starting at your app and I wonder if I can use
name2 = value2, valuw2
or do I need
name2 = "value2, valuw2"
or does it even make sense to give name2 a list? Or do I separate lists with ";" or maybe ":" like in the environment profile that you hold up as a model? Now I have to read your code. Don't pretend that you put this in the documentation, because you didn't mention the documentation when we were measuring "simpler syntax".
Of course, even on UNIX, nobody can agree on the exact syntax for the profile. Csh is different from bash is differnt from korn shell is different from blah, blah, blah. And nobody on non-unix platforms knows what the hell you are even talking about.
XML is THE standard for platform independent data. Bash script syntax is not.
Because it is the standard, and an open standard at that, you don't have to worry when you move your app to win32 or BeOS or Mac or Tru64 unix or fooOS about the availability of tools that can read and write the format.
I am quite a heave XML user. I think XMl is good for two reasons:
... ... ... ...
a) I will NEVER write a parser again. That is already enough reason to use XML. Just eliminating all the string digging makes it worthwhile.
b) I think a move to tree structures as opposed to everything-is-a-table is a good idea in general.
That said, one thing about XML that is irritating is when it gets used for things where it not good.
XSLT is a good example of this. A programming language in XML sytax? Why for chrissakes??!!
The problem with XML in a PL context is that it is not possible to build trees with multiple branches without complicated syntax. For instance, the following sort of thing is not possible:
if x
elseif y
elseif z
else
endif
Simply because a tag in XML can only have a start and en end and NOTHING in between, which leads to things such as XSLT's stupid CHOOSE tag.
Also, an example such as
5">
is dumb when if (x >5) then is su much easier.
Using XML syntax for a language just because is just plain dumb. Syntax matters. This (sadly) is one of the prime reasons LISP is dead. All the LISPers chout that "Yeah, but its easy to learn once you get used to it". Bullsh*t.
(if not (> x 2) (+ 4 y))
is NOT easier to read than
if x > 2 then return 4 + y
And for the same reason XSLT is a pain in the butt. But I digress....
The dangers of excessive individualism are nothing compared to the oppressiveness of excessive collectivism
Yeah, and therein lies their problem. For all the money that goes into usability testing, they are BULL-FUCKING-SHIT.
No offense, but they sit these people who've never used gnome before in front of a computer for 15 minutes. Watch them go: duh.. I don't get it.
and then try to draw conclusions based on this as to what would make gnome better for everyone.
What utter fuckin bullshit. Do you buy a computer, use the desktop environment for 15 minutes, and then throw it away because you're done? NO! You use your computer for at least 3-4 YEARS. You learn. You're not a computer-illiterate-monkey for the rest of your life with a goldfish memory. In other words, of ALL the people using gnome *right now* only about 2 of em are within the first 15 minutes of using it. All the rest are at different stages of being used to it. Therefore the whole basis of these usability studies is that they're great for making gnome usable the first 15 minutes you sit down with it.. but that's it. Then you're out of the target market that benefits from them.
You want to do R-E-A-L usability studies? get 100+ people who use gnome NOW. Of all different levels of skills, just like distributed in the user community. Study them for at least 3 months. Thouroughly. Then make your decisions.
Liberty.
And what about code-size? Every progam it's own parser. Shudder.
Ease of programming?
I'm intentionally taking this slightly out of context, but I think this is one of the biggest overlooked benefits of XML.
Sure, there might be slightly better data formats out there for specific cases in terms of speed or file size, but as far as development time goes, XML parsers can't be beat. You should basically be able to take a parser, wrap it with a few calls or a class to deal with your specific file, and you're done.
You don't have to re-implement and fully test your own parsing engine; you're using a parser that's pretty much optimized and tested to completion. This lets you develop more robust software in a shorter amount of time, and also lets you focus on coding the things you actually *want* to be coding.
Unless anyone you're someone that actually gets off on writing parsers (people like that are out there; I've met them...), I think there are very few reasons NOT to use XML.
It's only software!
> An 18% increase in size after compression is nothing to boast about. (and the XML can never be smaller, it has more information in it)
You missed the point more important to me:
>> As if space for config files would matter...
A increase by some factor is debateable. A percentile increase not. Especially not in such a limited benchmark. On a sidenote: My compact example is even one byte smaller than the given INI-example.
> There's a lot more to validate in XML than simpler formats.
Yes, but you don't have to do it. Well, more importantly, I don't have to do it.
>I'm a programmer and XML files hurt my eyes
And it hurts my eyes (as a programmer) to have ever look at a textual-file beside code. I want the frickin data as binary in memory.
> This is a cop out [...]
> [...] you might as well go for a binary format
No, it is not.
It is human-readable. Visually not as pleasing as an INI-File. But not something you have to rack your brain about.
All those people answering here wrote portions of valid XML, in a simple text-box without the use of some XML-editor and with minimal knowledge about XML.
But additionally to editing it with vi, cat, echo, sed, awk, or dontknowwhat, you have the possibility of using any XML-editor which represents the data in a more eye-pleasing form, performs automatic syntax-checks, and saves you some typeing.
Think source-code: You can use your spartanic editor, or your favourite developing enviroment with code completion, syntax check and the like.
> I'll say. I'm a programmer and XML files hurt my eyes. Screw John Doe, *I* don't want to look at them.
I think that is the main problem. Your personal dislike for XML. I don't think, I can discuss your aversion away.
I'm not emotional attached to XML. At most I'm slightly unhappy about the redundancy introduced by the opening and closing tags, but it is there, it works, it is becoming standard.
"Between strong and weak, between rich and poor [...], it is freedom which oppresses and the law which sets free"
Now, don't get me wrong. I love Gnome, have used it for years (although I am typing this on OSX), and never got on with KDE. I think the article is a big bunch of trolling hooey. But... Many moons ago, I ran the Ximian Gnome packages, probably Gnome 1.0 or possibly 1.2. The File Selector back then had a fourth button at the top of it, next to new folder / delete file / rename file buttons. This button took you to ~ when you clicked it. Where has this bloody button gone? I remember reading that it was a Ximian patch to the file selector, and presumably it never made it into the upstream sources. When I left Ximian (the day I switched to Debian testing), I lost that button forever and I still miss it. It seems like such an obvious thing to me. Bah! Humbug!
You win again, gravity!
Mandrake may have gotten its start by bringing KDE to Redhat, but it has had little to do with Redhat since about version 5 (and now Mandrake's at version 9). Just because a distribution uses rpm packages doesn't mean it's Redhat based. Mandrake has all its own administration tools, and even its own package manager that's sort of an apt-rpm hybrid (urpmi). It has left Redhat far behind with respect to this, as well as in installation and configuration, where Redhat is still playing catch-up in many ways. So don't equate Mandrake with Redhat. It's certainly no more Redhat than Caldera or Suse are, both of which started as modified Redhat too, but neither of which have much in common with it anymore.