A User's First Look at GNOME 2.0

Gentu writes: "OSNews has just published a review of the Gnome 2.0 desktop environment and its verdict is not so positive. The author feels that the new version is limited in many ways and with a UI not well designed."

Comments

Gnome and KDE are more or less the same these days

[ringbarer]

Maybe it's time to become the INNOVATORS, rather than copying the Win32 line of User Interfaces, which frankly, are getting stale.

Take a look at the visual inventiveness of Mac OS X for starters. There's a GUI that's worthy of the 21st Century.

Re:Gnome and KDE are more or less the same these d

[UnknownQ]

In my personal opinion WindowMaker is the best wm, but it is still a clone. Check out this promising distro.

Menu choices

[Ratface]

The reviewers comments about theme management menu choices seem very sound to me. As a long time user of Linux on the desktop I often find that default menu layouts for Gnome & KDE are confusing and unintuitive.

I'm also less than keen on what I have experienced of Nautilus so far and hearing that turning it off presents a naked desktop doesn't do much for my confidence in this product.

*sigh* I guess I'll be waiting for the next release before upgrading.

Re:Menu choices

[mpe]

The reviewers comments about theme management menu choices seem very sound to me. As a long time user of Linux on the desktop I often find that default menu layouts for Gnome & KDE are confusing and unintuitive.

Some of this probably depends on what distribution you are using. Anyway the way Windows does it isn't exactly intuitive.

Re:Menu choices

[GauteL]

Eh? Your comment on Nautilus makes no sense.

Nautilus is a desktop and file manager. Of course turning it off gives you a naked desktop, because you no longer HAVE a desktop-manager. How is this Nautilus' fault?

But please do not listen too much to what the reviewer said, because it is totally opposite to most others experience.

Firstly, for all persons I've ever spoken to about GNOME 2.0, it feels way faster than GNOME 1.x

Secondly, there is a centralized place for configuration. It is called "Desktop preferences" and it is either in the GNOME-menu, or in "start-here:". The reviewer got this fact completely wrong, almost on the edge of malciciousness.

He does have some valid points however. The theme-issue is inherited from GNOME 1.x, and was sadly not possible to fix in GNOME 2.0 without much delay.

The other issue, which does speak against intuitivity is the menu-panel. It makes no sense to move the menu-panel, as it is totally meant as a top-menu in all it's design.

However it is still possible to remove the menu-panel and just use a bottom GNOME-panel like Windows or KDE. You just have to create the new panel before you remove the menu-panel, as GNOME won't let you remove all of your panels.

Re:Menu choices

[Ratface]

Interesting points. I understand where the reviewer was coming from though regarding Nautilus - when I first installed Ximian Gnome I was using a slow machine and Nautilus chewed through pretty much all the memory I had - and then some. At that point I could kill Nautilus and return to my old desktop. However, Nautilus was still set as the default file manager and openenig it would start up Nautilus on the desktop again. The only way to stop it was to remove it from the running programs in the Gnome session list and restart Gnome. Pain in the rear!

Since I upgraded my machine I haven't really noticed it as much. I'm still not a grat fan of it as a file manager though...

Re:Menu choices

[Menthos]

Since I upgraded my machine I haven't really noticed it as much. I'm still not a grat fan of it as a file manager though...

Try the 2.0 version. Speed-wise and performerance-wise there is simply no comparison between the GNOME 1.4 and the GNOME 2.0 versions of Nautilus.

Re:Menu choices

[HiThere]

I wouldn't know about faster, but then I never minded the speed of Gnome 1.

OTOH, the Gnome 2 desktop seems to be missing some of the functionality. It's there if you dig through the menus, but you must dig for it. I wish I could be more explicit about this. It's almost as if the desktop has been stripped of the useful functions.

I suspect that this is largely because my standard Gnome installations were done by the Red Hat installer, and the Gnome 2 installation was done via the Ximian Red Carpet Gnome2 beta channel.

There's one bad exception to this. The dial up connection seems to be just gone. It was there when I first installed Gnome2, but as I updated it at some point it vanished. I've been using kppp to manage my dial-up connedtions since then. The Red Had network monitor will dial up a connection, but it never seems to connect it in any useful way with anything else (or even itself). And it doesn't hang up. Eventually I resort to killing the process. I suppose that I could use wvdial, but I'm not sure, as I haven't tried it. This has been a periodic problem with Gnome since the early days, though the Red Hat installations have usually fixed it (I think it was broken in 6.0, or 6.1, or both). Fortunately, so far I haven't had an install where both the Gnome/Red Hat dialer and kppp were broken since RH5.2 (for that one I ended up using wvdial from a text window).

This problem has persisted through several computers. It isn't only a problem with Red Hat, however. Progeny had the exact same problem. It would dial, but then it couldn't find the connection that it had established (and it wouldn't hang up). This was with whatever was the default window manager for Progeny. OTOH, so far Mandrake hasn't had this problem, but I can't remember whether I've tried using Gnome as my desktop under Mandrake.

Re:Menu choices

[Fizzlewhiff]

I loved GNOME until they added Nautilus as the default [insert anything and everything here] manager. While I haven't tried GNOME 2.0 yet I could relate to what the reviewer was saying because I had similar frustrations in 1.4. Hey, where did my wallpaper go, what happened to my icons? Between the Nautilus options and Sawfish options I became a little confused and I quickly went back to gmc. I guess little has changed in this department in 2.0.

We all bitch at Explorer being a file manager, desktop manager, and web browser too so I guess it is only natural for some of us to be equally annoyed with Nautilus. Luckily I can do everything I need to do on my box via ssh.

Re:Menu choices

[el_chicano]

Nautilus was still set as the default file manager and openenig it would start up Nautilus on the desktop again. The only way to stop it was to remove it from the running programs in the Gnome session list and restart Gnome. Pain in the rear!

I was playing with Gnome 1.4 on P-100 w/32 M RAM and Nautilus just brought that system to its knees! Nautilus was like a vampire, I could not kill it the way you described (it kept respawning when I did not want it to).

I ended up not running gnome-session; I found it is enough to run sawfish and optionally the gnome panel. Sawfish is great, I don't miss the desktop icons at all and the mouse-driven menus are nice. The gnome panel is O.K. too, but I think I like KDE's Kicker a little better...

Some valid things, and a lot of not-so valid

[JanneM]

Some of the issues he brings up seems valid. That said, I run Gnome2 and I don't recognize many of the problems he brings up.

First, for me, Gnome2 is far faster than Gnome1.4. This goes for most individual applications, as well as the desktop overall.

Lack of options: Well, yes and no. There has been a serious attempt at providing sensible defaults for a lot of stuff, and hide away rare and/or strange options into the gconf system. While some people like being able to tweak their desktops to hell and back, for many users it is just plain confusing to have as ridiculously many options everywhere as Gnome1 had. Note that for those serious about tweaking, gconf is there for your time-wasting pleasure. :)

Gedit: I've tried repeatedly, but I am unable to duplicate the marking thing he talks about.

Galeon has continued to work flawlessly for me, as have all other Gnome1 apps I have. he mentions that he does not have a Gnome1 installation; that may be an explanation as to why Gnome1 apps do not work...

As for 'scattered settings' - huh? I get all settings neatly in the 'Desktop Preferences' menu. That certainly includes things like xscreensaver settings and pretty much everything else he gripes about in this area. I do not have a 'Desktop theme', as he seems to have, but just the 'theme' option - as it should be.

I get the feeling there is something rather wrong with the reviewers setup; something like an incomplete install, or a mix of older and newer packages or something like it.

/Janne

Re:Some valid things, and a lot of not-so valid

[KeyserDK]

I think it's because he is using a mandrake setup.
It has some really odd menu/icon stuff in gnome2, which i know isnt in the 'default' gnome2.

He should try a clean slate, on top of that -O3 wiht gcc 3.1 is just _not_ an good idea, yet =)

Re:Some valid things, and a lot of not-so valid

[Mark Round]

s/he/she/g.

Take a look at the reviewer's name.... and her website http://www.eugenia.co.uk/.

Re:Some valid things, and a lot of not-so valid

[redtuxxx]

What the reviewer has done is done is very simple

Ignored release notes
Ignored Various READMES
Ignored known gotchas

The reason galeon wont work is that the mandrake rpm sounds like it is compiled with nautilus1 support, and nautilus 1 has been clobbered

The one thing loud and clear through all the development process is INSTALL GNOME2 IN A SEPARATE PREFIX!!

Personally I cant think of anything missing with my install of gnome2 (parallel with gnome1.4)

If people cant read release notes they should just pull down ximian RPMs

REDTUX

Re:Some valid things, and a lot of not-so valid

[mpe]

Lack of options: Well, yes and no. There has been a serious attempt at providing sensible defaults for a lot of stuff, and hide away rare and/or strange options into the gconf system.

What is and isn't a commonly used option is a rather subjective thing. That's before you consider that any configuation really should be settable or even mandatable by the sysadmin. Quite possibly on a per user/group basis to deal with tweakers who'd never get any work done if they could tweak all day.

Re:Some valid things, and a lot of not-so valid

[JanneM]

The current production version of Galeon is for Gnome1. If you want to run Gnome1 apps, you need Gnome1 installed. A major reason for bumping version from 1 to 2 is that the ABI (and API) is not backwards compatible. And as you point out yourself on the Mac, you need both systems to run legacy stuff. As more applications become stable on Gnome2, there will be less need for Gnome1 to be installed.

So, the oprions are: have Gnome1 installed as well; run Galeon from CVS; or wait until Galeon for Gnome2 is out.

/Janne

Re:Some valid things, and a lot of not-so valid

[Shelled]

Ignored release notes
Ignored Various READMES
Ignored known gotchas

Doesn't exactly sound like a ready for the desktop product to me.

Re:Some valid things, and a lot of not-so valid

[Xiphoid Process]

did you manually upgrade all the desktop and widget libraries when you updated from windows 2000 to XP? No? You just inserted the cd and let the whole OS updater do it for you? If you tried to manually update all the different interdependant libraries on windows without reading any documentation, do you really think it would work? i think not.

This is exaclty how it happens with gnome too: if you arn't a power user (ie, if you can't read and follow the instructions in the release notes) wait for your os distribution (ximian, redhat, debian, madrake, what-have-you) to release an update.

Most of the "reviewers" problems would never have come up if he A) followed directions, or failing the ability to do that B) let his distribution (do the update)

if you want to live on the bleeding edge and update packages left and right inbetween distribution releases, be prepared to read the instructions or pay the price, its really not that hard.

Re:Some valid things, and a lot of not-so valid

[Jeffrey Baker]

Funny. In my desktop preferences, I have nothing for xscreensaver. I have background, font, keyboard, keyboard shortcuts, mouse, network, sound, theme, toolbars and menus, and window focus. Which one controls xscreensaver?

Re:Some valid things, and a lot of not-so valid

[qweqwe]

> > Ignored release notes
> > Ignored Various READMES
> > Ignored known gotchas
> Doesn't exactly sound like a ready for the
> desktop product to me.

These are installation issues, not usage features. If you try to install MacOSX on the original Mac or on the PC because you didn't read the release notes, you'd get similar problems. You'd also get similar problems if you automatic upgrade from Windows 98 to Windows 2000 without checking if Windows 2000 supports the new hardware.

If you don't want to set it up yourself correctly, you have to do some reading. If you don't want to set it up yourself, wait for Ximian or your distribution to set it up for you.

What counts is, how does it behave when you install it correctly? That question wasn't answered since the system was not installed correctly. GIGO applies as always.

Re:Some valid things, and a lot of not-so valid

[e_n_d_o]

Doesn't exactly sound like a ready for the desktop product to me.

I'm not 100% certain on this, but I believe that if you are running Gnome2, you either compiled it yourself or use debian. That essentially means its not "ready for the desktop" in the sense that you are referring. Once it ships with distros and is available through Ximian, users shouldn't have to read release notes and readmes. But when a reviewer is compiling it himself and fails to take these steps, I believe the error falls to the reviewer.

Re:Some valid things, and a lot of not-so valid

[Eil]

Lack of options: Well, yes and no. There has been a serious attempt at providing sensible defaults for a lot of stuff, and hide away rare and/or strange options into the gconf system. While some people like being able to tweak their desktops to hell and back, for many users it is just plain confusing to have as ridiculously many options everywhere as Gnome1 had. Note that for those serious about tweaking, gconf is there for your time-wasting pleasure. :)

Sensible defaults are fine. But I take exception to the fact that a lot of the (arguably more useful) options have been completely eradicated from GNOME. The developers reportedly did this in an effort to make the environment more intuitive.

There's a quote from one of the GNOME developers floating around... something to the effect of: "Better to have one good default way of doing something instead of six equally broken ones." This just reeks of ego-centricism on the part of the GNOME crew. Obviously they think that their desktop preferences are somehow The Right Way and everyone elses's are "broken".

For example Havoc Pennington, author of Metacity, the new GNOME wm (please correct me if those facts are wrong), professed that he thinks the sloppy-focus model of window management is stupid. Did he offer a reason why? No, but it is clearly implicit: "It isn't my way of doing it, so it's wrong." I have to take offense to that because I think the click-to-focus model is utterly retarded. It severly grates on my nerves whenever I have to make that extra, pointless, unneeded click to focus a program in Windows. I got away from Windows partly because I hated the lack of flexibility in the UI, and now GNOME is trying to force the same old crap back on me again.

I used GNOME not because it looked good, not because it had sensible defaults, and not even because it was free software. I used GNOME because it offered a plethora of options to let me do what I wanted to do and how I wanted to do it without a whole lot of bitching.

I'm in the process of building GNOME 2 as we speak, and I will give it a try. But if my experience falls in line with most everyone elses, then it looks like I'll have to see what KDE 3 has to offer...

Re:Some valid things, and a lot of not-so valid

[Nailer]

The one thing loud and clear through all the development process is INSTALL GNOME2 IN A SEPARATE PREFIX!!

And? The development process for Gnome 2 is over. The environment is supposed to be ready.

Re:Some valid things, and a lot of not-so valid

[Nailer]

did you manually upgrade all the desktop and widget libraries when you updated from windows 2000 to XP? No? You just inserted the cd and let the whole OS updater do it for you?

Indeed. But Windows 2000 or XP weren't labelled as finished products until those installation mechanisms were finished and avaliable.

Re:Is it possible to Cut'N'Paste yet ?

[kpetruse]

Isn't this a problem with X itself, rather than the GUI?

This is one of those things that will keep Linux from the average user's desktops until it gets resolved. People love being able to paste stuff willy-nilly in Windows. Hey, Windows might be full of holes but at least it's easy to use.

Depends on the person.

[MartinG]

If you're the kind of person who decides which desktop to use based on reading a few reviews or asking your friends, then maybe this review is for you. Good luck.

If you're like me and you like try things for yourself, then you're probably already downloading it, and you probably already know that you're more different from the average person than you think, and you already know that you are constantly surprised by how much you disagree with reviews of this kind.

Seriously, I would recommend that everyone tries gnome 2.0 if you have time.

Re:Err... yeah, ok...

[yatest5]

I'm prepared to accept that the author of this article may not have English as her or his first language, but this sentence seems to display a lack discontinuity of thought that's not confidence-inspiring for the rest of the 'review' (or opinion piece as it seems to really be)...

Hmm, sweet irony.

Talk about laugh

[nagora]

The new Gnome 2 environment starts up much-much faster than Gnome 1.4 used to! It loads on my dual Celeron 533 in about 2-3 seconds,

Windowmaker loads in a fraction of a second on my 300mhz uniprocessor box.

TWW

Re:Talk about laugh

[Avakado]

>> The new Gnome 2 environment starts up
>> much-much faster than Gnome 1.4 used to!

> Windowmaker loads in a fraction of a second
> on my 300mhz uniprocessor box.

I bet Tab Window Manager (aka. twm) starts even faster! It must obviousely be far better!

Note: comparing the startup speed of software with completely different sets of functionality makes no sense.

Re:Talk about laugh

[satanami69]

Waimea ~1.5 seconds
FVWM ~4.0 seconds
Gnome 2.0 ~25 seconds
KDE 3.0 ~1 minute

hell, all i usez it for it to open mozilla anyway. I'll take waimea.

Re:Talk about laugh

[Glytch]

Amen to that! And ROX-Filer makes a nice, lightweight companion for Windowmaker, too.

Re:Talk about laugh

[SCHecklerX]

Well no shit. Window Maker is a WINDOW MANAGER ONLY. Gnome is a DESKTOP ENVIRONMENT with lots of stuff in addition to a window manager. I swear the number of people who continue to confuse this is so irritating. BTW I usually use Window Maker as well.

Yup. I use windowmaker too. And my environment is Windowmaker + Rox Filer (which is now using gtk 2 as well...but I have not upgraded yet b/c I am highly annoyed at the decision to use one big screen sized window instead of the root :(

Re:Gnome and KDE are more or less the same these d

[Avakado]

> Maybe it's time to become the INNOVATORS,
> rather than copying the Win32 line of User
> Interfaces, which frankly, are getting stale.

How is it getting stale? Do you have any mind-blowing new ideas that counter well-established knowledge about the usefulness of GUI widgets as we know them today? Let me remind you that a good feature of a GUI is to be useful, not to be innovative.

> Take a look at the visual inventiveness of Mac
> OS X for starters. There's a GUI that's worthy
> of the 21st Century.

While the GUI of MacOS X might be "inventive", I find it extremely cumbersome to navigate, dreadfully slow, overly full of bells and whistles pointless animations, non-intuitive, obstructive et.c. In short: a real pain to use. While the animations might be funny to look at the first time, and the GUI looks very sleek, it generally reduces productivity. Most of the work devoted this GUI, is clearly meant to improve visual appearance, and not usefulness.

It appears obvious to me that people claiming the MacOS X GUI is intuitive have either not really tried it themselves, or never tried anything else. In the same manner, stating that "GNOME and KDE are more or less the same" shows that you haven't really tried both.

configurability

[Random Walk]

anyone who complains about a lack of configurability apparently never had to deal with:

• people who managed to tear off a taskbar by accident, and could not figure out how to put it back in place,
• people who managed to switch off a taskbar by accident (this evil M$ Word ...), and could not figure out how to switch it on again,
• countless other examples ...

Many, perhaps most, users use their PC only occasionally, are not familiar with configuration options, cannot 'fix' even the most trivial issues, and would rather need a well thought out configuration that cannot be modified by any means.

Re:configurability

[scabbers]

agreed about he first point, but "newbie-friendliness" is also offering a good standard setup from the beginning so that you _can_ cope with complexity but need not.
I like to have choices, but I like to to have the choice, not to make decisions (esp. about trivialities), at least sometimes, too.

Would a big "reset-to-default-desktop"-button help? :-)

Re:configurability

[Omnifarious]

This whole "Linuxs on the desktop" thing is going in a very strange direction, IMHO. Why do people choose UN*X as a desktop system in the first place? I for one don't care about newbie-friendliness that much, all I want is a powerfull, efficient, open and stable system that is a pleasure to use and doesn't treat me like an idiot.

If I were a CIO who had bought into the ideas of why Open Source provides enormous benefits to my business, I'd want an Open Source desktop I could stick in front of any random administrative assistant and have it work. If you don't want such a beast, then don't use it.

Also, Linux on the server is doomed, in the long run, if Microsoft continues to control the desktop. Microsoft will use that control to do everything they can do makes things difficult for Linux on the server. Eventually people will just give up and switch to NT because it's too hard to run a Microsoft Windows network any other way.

Face it, as a business strategy, keeping your OS a newbie unfriendly 'l33t' OS is dooming it to failure. It's much the same problem as trying to sell high-end servers when you have cheaper alternatives that keep getting more and more powerful. You keep being pushed into higher and higher niches in the market until your pushed so far into the stratosphere that you only have 5 customers and end up going out of business shortly afterwards. In this case, the cost is the learning curve, not the cost of the hardware, but the principle is the same.

Re:configurability

[Dixie_Flatline]

Exactly! Thank you, thank you.

Here's something people fail to realize: even if you dislike your interface in some way, with a well designed interface and some training, you can be faster with the interface that is subjectively offensive than the one that you feel is somehow 'comfortable'. Configurability is the hallmark (in general) of a poor UI design. It means that you didn't know how to do it properly in the first place, so you're passing the buck to the user.

The advantages of a rigidly stardardized interface are often completely ignored, but they're what allow most people to sit down at any computer and start typing.

Re:configurability

[frantzdb]

And to those 31337 h/\xors who want more configurability, use gconf-editor.

--Ben

Re:configurability

[HiThere]

How about this one:
I recently installed a Ximian Gnome2 snapshot. And since I wanted more than one workspace, I switched from MetaCity to Sawfish. After awhile a GUI crash occured.

When I had recoverd, the control panel appeared gone. I couldn't even switch back to Metacity, but I couldn't alter the Sawfish options either. The didn't exist.

Up until the crash, I could have examined the menu call to find out how to invoke the controls. After the crash, the menu editor was no longer accessible.

. . . . . . . . . .

Well, since that time I've largely recovered. But not totally. I did this by using Red Carpet to install upgrades. A real fix would probably have been possible too, but I don't feel like removing the Beta version, and then downloading it all again.

OK. Gnome2 was clearly marked beta. And Metacity was the clear safe choice for the window manager. So I knew I was living dangerously. (Well, KDE is my main desktop, I was experimenting with Gnome, and didn't mind taking a few chances.)

But it sure isn't obvious to me how one finds out how to do a recovery. I'm sure that it must be possible to do one without that whole "remove and reinstall" routine, but this doesn't tell me how to find out how do actually do the recovery.

Re:configurability

[EastCoastSurfer]

If I were a CIO who had bought into the ideas of why Open Source provides enormous benefits to my business, I'd want an Open Source desktop I could stick in front of any random administrative assistant and have it work.

Actually CIO want complete control of the desktop in the least expensive way possible. Complete control can happen in any environment(windows, *nix, etc...) with the right admin. The admin removes all configurability from the users and gives the users access only to things they need in an interface consistent across the enterprise. This gets to them to the stick in front of anyone and have it work thing while still having a free OS. The only real thing holding back *nix on the desktop in the enterpise is application compatability(inhouse custom apps, office file formats, etc...).

Home use is a totally different story for linux. Until Joe can go to walmart, buy deerhunter, and 1 click install it onto linux, linux will be a niche in the home market.

Re:configurability

[eddy the lip]

This isn't intended as a flame, but I'm curious what you want from some high-fallutin' desktop environment. I suspect I may be missing some important point, because I think I'm probably in a similar user category as you (long-time *nix user, like to configure things as best fit my work habits, and don't mind investing a bit of time to make me more productive in the long term), and I see no benefit to me in environments like KDE or Gnome (or Windows or OSX for that matter).

With a full desktop enviornment, I can drag and drop stuff from one app to another, keep stuff on my desktop and use some kind of file/app manager front end thingy (I'm sure there's a technical name for the likes of Nautilus). But why would I want to? None of those things save me any time. (I especially don't understand the drag'n'drop thing - even when I worked as office admin (gah) I don't think I ever did this once).

What I need is multiple desktops, configurable window placement and focus rules, window shading, menus I can edit to my heart's content and some themability so I can set the colors and sizes and fonts to something easy on the eyes.

On the other hand, for new users, a lot of office types and more visual people (my designer co-worker, computer savvy as she is, would never work from a CLI - it's not how she thinks) full environments are perfect. What's more, they want them simple, and I can understand that. I want more people to use linux (because it means I get asked fewer Windows questions), so I'm happy there's projects to make it more friendly for people who work different than I do. So long as I can have WM or Blackbox or something, I'm happy.

Or am I really missing something?

Re:configurability

[raistlinne]

need a well thought out configuration that cannot be modified by any means.

Ok, here's how to accomplish this: administrative lock. The administrater gets to lock his users out from making any configuration changes, even local ones.

Problem solved. We don't need to gut the usefullness of a program because some people can abuse it, we just shackle the offenders.

Here's a good analogy: because some people get into accidentys, should we remove the steering wheel on cars? Why not just make the people who are at fault for getting into accidents take a bus (i.e. take away their driving license). I don't know what country you come from, but here in the US (where much of gnome development is done), we have a princple called "innocent until proven guilty" (howevermuch the current government finds it inconvenient). Why treat all users like blithering morons, when that's only some (or perhaps most) of them? Why imprison everyone for the sins of some?

Re:configurability

[Eil]

Then either these inexperienced users can find a different, non-confusing desktop or the experenced users can find a different desktop if either group decides their current one doesn't fit the bill.

Most of us are complaining about the fact that GNOME went from a desktop for "experienced developers" to an "easy-to-use" desktop, dropping support for some of our favourite features along the way.

Many, perhaps most, users use their PC only occasionally

Many, perhaps most, users should not be using a Unix-based OS in this case. They should be using an operating system that was designed from the ground up to cater to their needs: Windows or MacOS. Because by the time that you finally manage to modify Linux (for example) to be as user-friendly and intuitive as possible for inexperienced users, it would then cease to be Linux as we know it.

Re:configurability

[Eil]

Here's something people fail to realize: even if you dislike your interface in some way, with a well designed interface and some training, you can be faster with the interface that is subjectively offensive than the one that you feel is somehow 'comfortable'.

There are a number of things wrong with this sentence.

1) Who defines what a "well-defined" interface is? That is a loaded term if ever there was one. If you were to take some kind of accurate poll of what constitutes a well-defined interface, I'm sure you'd get a whole range of different answers to include:

a. how intuitive it is to use
b. how efficient it is to use
c. how useful the included applications are
d. how it good looks
e. how quickly the computing environment responds to user input
f. the number of options available to control it

Everyone's going to have a different answer as to what they feel constitutes good interface design is in their opinion. Anyone who asserts that there is but one good interface is much too ego-centric; they're really only thinking about what works well for them. As if they expect the rest of the world should simply adjust to their preferences.

2) When you say "faster" what exactly does that imply? If by "faster" you mean "efficient", how can one possibly be considered efficient with a setup he or she is not even comfortable with? That doesn't make any sense.

Configurability is the hallmark (in general) of a poor UI design. It means that you didn't know how to do it properly in the first place, so you're passing the buck to the user.

That is patently false and moreover, ridiculous. Configurability is the hallmark of flexibility, nothing more. And it is most definitely not passing the buck to the user as long as you use sensible defaults. It's when you take away the ability to configure options and features that you immediately take away anything resembling usefullness. There's a reason I use Unix instead of Windows in my daily routine and that reason it has nothing to do with how standard the interface is. (By "interface" I refer to all user-visible aspects of the operating system, not just whatever happens to run in X.)

For an example, why don't we try Mozilla. Mozilla has some very sensible defaults and in most cases the user does not have to do modify a single one of them to start using it right away. But there are still something on the order of a hundred or more options that can be set "behind the scenes" by users who choose to modify them. Those options literally touch on every single aspect of the browser, from look and feel to performance and behavior. By your reckoning, the mere presence of these options (despite the fact that they are well hidden and have reasonable defaults) implies that Mozilla is a completely worthless browser. (Personal feelings aside.) That can't be the case, since two of my friends and my fiancee, all of whom are inexperienced computer users, think Mozilla is great. From what I've seen, they aren't the only ones.

The advantages of a rigidly stardardized interface are often completely ignored, but they're what allow most people to sit down at any computer and start typing.

The words "rigidly standardized interface" honestly send a chill down my spine. That's a socialist statement if I ever heard one. If you're proposing that all interfaces be the same, I'll be the very first in line to oppose that. I do not want my interface to be the same as anyone elses, simply because I'm not everyone else and my preferences for a comfortable and efficient interface are completely different from every other person's on the planet.

You might as well take it one step further and propose that all schoolchildren in the world have the same exact cirriculum, despite the obvious differences in culture. Or that everybody should be required to drive the same car. Wait a moment, that has already been proposed... think a fellow named Hitler thought of that one.

However, I'm not about to suggest that most interfaces shouldn't carry among them common elements. But it's important to note that those common elements are already in place. Application windowing is almost universal across all computing platforms. Everybody expects a scrollbar to scroll content, a checkbox to check, and radio button to select a particular option.
I can think of a hundred different examples that include cars, video games, and microwaves, but I think you get the picture.

There is simply no need for a universal interface the way you describe it. The current ones are far from perfect, but they are the best technology has to offer. You simply cannot create one interface that will allow everyone be efficient with it. Nor will absolutely every person even be able to use it. (The physically disabled, for example.) Most importantly, not everyone would even happy with it.

As long as there are computers, (and it's looking like there always will be, in some form) then there will always be different ways to inferface with them. I, for one, am thankful for that. There is no larger killer of innovation, spirit, and individuality than a large group of people conditioned to believe that there's just one way in the whole world to perform a task.

Re:configurability

[Dixie_Flatline]

A well designed interface is one that doesn't detract from the task at hand. Ideally, an interface should be as close as possible to a nipple (since the nipple is the ONLY intuitive interface, and everything else is just learned. :)

Anyway, a well designed interface is transparent. If the interface were perfect (so, imagine the most ideal sense) you wouldn't even notice it, you'd do what you wanted to do, and leave. That's all. And that's what we should aim for. Ironically, time spent CUSTOMIZING the interface detracts from the task at hand. The act of selecting preferences makes the interface LESS transparent.

Your thought that there is no perfect interface for all users is incorrect. There IS. The fact that nobody has been able to conceptualize it doesn't make it not so.

Moreover, I haven't been made aware of any studies that show that high customizability make UIs any better to use in any OBJECTIVE (ie. measureable) sense. Before you say that intagible benefits like comfort or aesthetics are important, in the world of interface design, they're very difficult to factor in. Treated as a science, UI design can be very concrete, measuring the time that it takes to complete a task, for instance, or find what you're looking for.

On the other hand, there HAVE been studies that have shown that despite a SUBJECTIVE dislike for the interface, an OBJECTIVE, measurable benefit can be found when the interface is meant to optimize productivity. Alas, I can only cite one example right now (Tullis, Thomas S. "Predicting the Usability of Alphanumeric Displas", Ph.D diss., Rice University, 1984). Sure, the study is old, but my only reference at the moment is Jef Raskin's book 'The Humane Interface'. I'll give it a plug right now. You seem interested in interfaces, at least to some extent. It's an easy read, and it'll expand your mind.

Anyway, back to the conversation at hand.

Now, 'efficient' and 'faster' aren't subjective terms. I can measure how fast you type, the amount of time it takes to move your hand to your mouse, the amount of time it takes you to point the mouse pointer at an object on the screen, etc., etc. I can certainly assert that something is faster, even if someone isn't comfortable with it. For instance, I could give you a very uncomfortable mouse to use, but it may be extremely precise and high tech, which yields a measureable decrease in the amount of time it takes you to manipulate data on your screen. On the other hand, you could use the old, dirty mouse on your desk, which doesn't roll very well, or click as accurately, but feels nice in your hand. The subjective experience tells you that your old mouse is better, but the objective measurements show that you're not as productive as you could be.

The problem that we're running into here is that you think you know what's best for you, when that's clearly a false statement. You're not a user interface designer, and despite your gesticulating, you're not an expert on what makes a good interface, merely what is aesthetically pleasing to you. You wouldn't try to diagnose and treat diseases on your own, so I don't know why you think you should be 'prescribing' user interface solutions for yourself.

Hitler comments aside, your comment is actually somewhat more astute than you realize. There ARE standardized curriculums (so we can measure everybody on a normalized scale) and despite their obvious EXTERNAL appearance, cars are almost all exactly the same from an interface perspective. There's a gas pedal, a brake pedal and a steering wheel. Ignoring any transmission issues, when you know how to drive a car, you pretty much know how to drive all cars.

It's easy to see why you feel this way about interfaces, however. When all your options suck, any change that you make will feel like an improvement. Don't think that I'm not customizing MY interface. The interface design was poor in the first place, and THAT'S why we're still making changes.

Try not to make this a bigger issue than it is. There's no political agenda behind my statements. I've read books, studied interfaces objectively, and tried to see where interfaces fail us on a day to day basis.

I don't know why you think that there is no need for a universal, optimal interface. Sure, it's a lofty (and likely, impossible) goal. However, striving for an interface that is so mediocre that the user is FORCED to customize it just to be able to get work done is ludicrous.

In any case, go out to your local computer book store, and pick up some books on interface design. If you're still in school, go take some classes in it. It's an interesting subject that actually has a scientific element to it. It's 1 part psych, 1 part art, and one part Computing Science.

Re:configurability

[Eil]

First, thanks for your reply. I was probably a bit more abrasive than I meant to be and that kind of thing can sometimes get out of hand.

Anyway, a well designed interface is transparent.

I'll agree with that, even though as someone interested in computer science, I don't tend to think of them that way.

Your thought that there is no perfect interface for all users is incorrect. There IS. The fact that nobody has been able to conceptualize it doesn't make it not so.

And I still don't think so. In regards to computers, an interface is usually defined as the link between the hardware/software and the whole human. The "interface problem" that we're discussing right here is that the hardware/software combination, in your agrument, doesn't account for variations in the human element. Sure, you could have your universal interface if all humans were the same! But they're not. Pretty much drastically so, both phsycially and psychologically. The problems resulting from this range from repetitive stress disorder to grandma erroneously double-clicking on every hyperlink in sight. Granted, these are problems with today's interfaces, but these interfaces were designed by so-called experts. The point is that no matter what kind of interface you design, somebody either can't use it properly or won't like it. (And yes, someone that doesn't like a particular interface is probably not going to use it either unless they quite literally have to.)

Moreover, I haven't been made aware of any studies that show that high customizability make UIs any better to use in any OBJECTIVE (ie. measureable) sense. Before you say that intagible benefits like comfort or aesthetics are important, in the world of interface design, they're very difficult to factor in. Treated as a science, UI design can be very concrete, measuring the time that it takes to complete a task, for instance, or find what you're looking for.

I doubt that the benefit of being able to customize a UI can be measured objectively. And yes, you can measure how effective a particular interface is... let me break off from this for a moment.

I'm getting the feeling that when you talk about interfaces, you're referring to a smaller unit than I am. Say, an interface for one particular program. I was originally referring to an interface as the whole of a computing desktop. In designing a UI that complex, the user is not typically going to be performing a few pre-determined tasks, but a whole slew of different tasks that cannot be predicted at design time. I think this is part of what the GNOME developers faced... when they rewrote GNOME to make it simpler, they had in mind the type of tasks that they performed regularly. From the complaints I'm hearing about GNOME 2, they didn't stop to consider that other users out there would be practically breaking GNOME because the types of and number tasks they perform are so much different. Particularly it's the advanced GNOME users who have been using it for years who seem to be frustrated.

On the other hand, there HAVE been studies that have shown that despite a SUBJECTIVE dislike for the interface, an OBJECTIVE, measurable benefit can be found when the interface is meant to optimize productivity.

I can admit that an interface can be effective even if the user personally dislikes it for being boring or what have you. Some people are forced to use such an interface to do particular tasks. A task at work such as data entry, for example. But if the pendulum swings the other way--if the interface is aestheically atrocious--then productivity is lost. Easy enough to design around until you take into account that each person has their own view of what is aestheically atrocious and what is not just as much as they have their own idea of what is pleasing. For example, the person who wrote the webmail software that I have to use to view my email probably thought that the solid bright-blue bar on the left hand side of the screen looked alright with the rest of the page having a white background. But I can't stand it because it makes my eyes go screwy. I can't look at it for more than a couple minutes at a time... lost productivity.

Particularly in the computing industry where impressions alone drive sales. Your argument might apply to ATM machines, but not personal desktop environments.

For instance, I could give you a very uncomfortable mouse to use, but it may be extremely precise and high tech, which yields a measureable decrease in the amount of time it takes you to manipulate data on your screen. On the other hand, you could use the old, dirty mouse on your desk, which doesn't roll very well, or click as accurately, but feels nice in your hand. The subjective experience tells you that your old mouse is better, but the objective measurements show that you're not as productive as you could be.

That's a pretty good analogy, (rare on slashdot, I know) and I'd admit that any person who chose pure aesthetics over efficiency is indeed a moron. However, you can't change that the person is a moron. What you can do is change the aethetics of the new mouse to match what the user prefers. That will make him want to use the new mouse, which is technically a better interface. This is one of the primary problems with interface design. Like I said before, overall impressions drive sales and adoption. That's why nearly every open-source project with an interface has screenshots on its website. That's why OS X has candy-colored buttons. That's why we have color TV. It would be awfully nice if interfaces were judged on technical merit alone, but they are not. If they were, Linux (and other Unices) would probably have a heck of a lot more users than it does now.

The problem that we're running into here is that you think you know what's best for you, when that's clearly a false statement. You're not a user interface designer, and despite your gesticulating, you're not an expert on what makes a good interface, merely what is aesthetically pleasing to you.

I'm sorry, but I think I do have a pretty damn good idea of what's best for me. As does every other person in the world out there, including yourself. No, I'm not an interface designer, and I'm not an interface expert. But I have common sense. Common sense tells me that if I don't like an interface for whatever reason, then I am going to resist it as much as I can. THAT makes it an ineffective interface. It does not take a rocket scientist to realize that people do not all come out of a factory with the same set of specifications, each one is different from the next in some way, shape, or form. Some even interact with their environment in totally different ways than 99.999% of the rest of the world. From what I've seen, current interface design caters to the common demoninator, i.e. "which types of people are most likely to do this sort of task?"

despite their obvious EXTERNAL appearance, cars are almost all exactly the same from an interface perspective. There's a gas pedal, a brake pedal and a steering wheel. Ignoring any transmission issues, when you know how to drive a car, you pretty much know how to drive all cars.

In using the car analogy, I was referring to the car as a whole. I'll assume that you knew what I was talking about when I mentioned Hitler. He proposed that in all of Germany, (and its subsequently acquired countries) that there would be just one kind of car that everyone drove. Having just one nationally-endorsed car for everyone would have make a lot of things easier: the cars being mass-produced would be much cheaper, streets built according to the width of a single type of car rather than estimating the average (and getting it wrong), social class differences would be somewhat blurred (cars are treated in this society as status symbols more than anything else), and perhaps most importantly repairs would be a breeze because every part on every car is exactly the same. You'd even stand a pretty good chance of knowing how to fix it yourself if you've had the exact same type of car for a few decades.

Now ask yourself how many people you know of would want to buy this car rather than an SUV, or Ford Mustang, or Chevy Truck or whatever they decided suited them best (regardless of whether it does or not). Same goes with desktop environments, my friend.

It's easy to see why you feel this way about interfaces, however. When all your options suck, any change that you make will feel like an improvement. Don't think that I'm not customizing MY interface. The interface design was poor in the first place, and THAT'S why we're still making changes.

And my opinion is that creating a desktop that's perfect for every single kind of user out there is impossible, hence having the option to make changes and tweak things is essential.

I don't know why you think that there is no need for a universal, optimal interface. Sure, it's a lofty (and likely, impossible) goal.

I know it's getting late and my mind isn't working at full-throttle any more, but that last sentence seems like a bit of contradiction to what you've been saying. There's nothing wrong with a universal interface for smaller components of the desktop. I wouldn't be the least bit worried if every media player on the planet, for example, changed their interfaces to be completely identical as long as it worked well. But the entirety of a computer user's interface (desktop plus programs) is a whole 'nother matter because there are virtually limitless tasks that one can perform on a desktop computer. More over, many people's desktops are quite dynamic. Interface designers (such as the GNOME team) simply cannot predict and plan for every single scenario that the user will encounter. Sure, adding options to the interface might result in some that end up as unnecessary bloat. But those can be discovered and gotten rid of over time through feedback. If those potentially useful options never get added in the first place, many users (myself included) might not even try the interface in the first place because it doesn't suit their needs and the developers wonder why their product is doing so shabby because they're getting no feedback other than the fact that their product isn't being used. See, you have to go a bit deeper than just the design of the interface alone in order to get anything remotely resembling an accurate guage of its success.

However, striving for an interface that is so mediocre that the user is FORCED to customize it just to be able to get work done is ludicrous.

I would never claim that a decent interface should be required to be customized right out of the box before it can be used. That is silly. What I advocate is the ability for users to be able to customize their desktop at any time as they see fit, so that the interface can remain flexible to their needs. If I recall your original post, you stated that the mere presence of lots of options was indicitive of a broken interface, with all other considerations being overlooked.

In any case, go out to your local computer book store, and pick up some books on interface design. If you're still in school, go take some classes in it. It's an interesting subject that actually has a scientific element to it. It's 1 part psych, 1 part art, and one part Computing Science.

Aha! Art! You admitted it! Aesthetics do play a part after all. :) Anyway, I shall make certain I do look into it further, as I will hopefully be taking some more CS classes this fall on my way to get a degree. Thank you for your interesting conversation.

Re:configurability

[Dixie_Flatline]

You have to think of this 'ideal' interface more mathematically or philosophically. The 'optimal' interface is a thought experiment - sort of like infinity. It is possible to conceive that such a thing exists, but you can't hold it in your hands, look at it, or otherwise 'have' it. Pick up 'Goedel, Escher, Bach' for a better example of 'perfect' objects. It's a good read for someone interested in CS, anyway.

The fact that our experts have so far only been able to come up with cumbersome interfaces is not proof that we can't come up with something better. All sciences strive to come up with something better. You may have seen that new input method that they came up with for palm devices headlined on slashdot a few days ago. New ideas and methods are popping up every day.

So, despite the fact that people are very different in many ways, the philosophically optimal interface is a concept that we can work towards.

I REALLY, REALLY recommend Jef Raskin's book to you, again. Part of the problem that I *think* I see is that you haven't been exposed to many UIs. There have been a lot of successful (from a usability standpoint) interfaces that require no customization, but allow the user to get their work done without lost productivity. Creating the right user interface has involved changing the interface right down to the filesystem level. I'm actually talking about interfaces at a much larger level than you are - at the operating system level. Current operating systems may expose too much of the system layer to the user (and when I say user, I mean user. If you're a systems programmer, you'd clearly have access to different things at some point, though you should be able to reach those things utilizing the interface, for the most part.)

Now, your example of the terribly coloured web-mail thingme would also be measured by a proper UI engineer. Colour intensity and the like are matters of ergonomics. Nobody would ever create an interface that had red text on a lime green background and expect people (normal people, let alone the colourblind) to use it. Things like that are quantifiable, much like font readibility.

In regards to the aesthetics of the mouse, you'll note that despite the wide range of hands on people, you don't actually see configurable pointing devices (beyond button configuration, which is another OS/application customization, really), and yet everyone can use one with reasonable facility (I'm excluding the disabled, but they'll always have problems in this world until we can 'fix' their disabilities.)

Now we're back to discussing this so-called 'ideal' interface. If it worked, would you REALLY care that it was the same as the next guy's? I'm talking about a level of optimality that you've likely never experienced as a user. You're stuck thinking about how the interfaces that you use right now don't work, and thus force your hand to make changes. People RARELY know what's best for themselves, I find. (Don't take that as some sort of political statement that promotes some sort of despotic government. :) It's why we have doctors, and mechanics, and programmers. They've been trained to know best. I know what's best for me GIVEN WHAT IS AVAILABLE, but I'm still looking for the day that someone drops a UI in front of me, tells me to try it out, and it JUST WORKS. It may happen.

For our discussion to go much further, you'll have to undergo a slight mental shift. Your experiences are telling you, correctly, that interfaces today don't work for everyone and require customization. But there IS a better world out there. :)

As for being 'required' to customize your environment, how many people do you know that haven't changed a SINGLE thing in their environment? If nobody changes it, then it's an interface element that's transparent. Nobody knows it's there enough to care that it can be changed. On the other hand, almost everyone customizes their colours in some way...why would they make colours so ugly that EVERYONE needs to change them?

There are a bunch of other considerations that I'm skirting over that involve modality, various quantifications of interfaces...stuff that fills books on UI design.

And Art is something that's always present in Computing Science. I'm a programmer with an interest in Graph Theory and UI design, Systems Programming and Compiliers. I'd like to think that there's a little artist in me that makes those things a little better when I work on them. :)

Anyway, it was good talking to you, but I'm starting to hate this little typing box. If you want to continue on, mail me at BioWare and we'll chat. And, as an amusing exercise, try to look at interfaces from my perspective for a while. Pretend you're me, and try and find the elements in your UI that could have been standardized without removing productivity from the interface for any potential user. You may be suprised with what you come up with. :D

Err, Well, mostly on topic

[smithwis]

Something I'd like to know is...

How well does Gnome2.0 work without Nautilus? I've never been a big fan of Nautilus and so always removed as much dependence of Nautilus as I feel safe removing from my instalations of Gnome. I've noticed that as I've updated Gnome, that Nautilus has been more and more integrated. For instance Gnome tries to get you to use nautilus to navigate to different control panels, Fortunately I was able to dig up the Gnome Control Center utility last time I updated. Anyways, with Gnome now using a new and incompatible GTK do we lose the gnome control center in favor of the not so nice Nautilus interface?

It's a shame if we have to use Nautilus. One of the reasons I liked Gnome so much was that you weren't really forced to use much of anything. You didn't have to use Sawfish(or now metacity) for your window manager and you didn't have to use GMC or Nautilus(I prefer an XTerm window for the most part)

Thanks for any light you may shed on my questions. And excuse me for being a lazy ass and not doing to much research b4 asking;-)

Steve

Re:Err, Well, mostly on topic

[kigrwik]

No problem here: my nautilus is gone, and everything is still accessible through the Applications menu.
(it'll be back as soon as I apt-get it back, but I'm lazy :)

Actually, Nautilus is a *pain* when used w/ transparent terminals, as each redraw of the icons will cause each terminal to redraw. Besides I don't like desktop icons anyway. What I miss in nautilus is:
1. middle-click to open new windows
2. split-screen for easy DnD
3. Konqueror, that's what I miss, actually. :)

But with nautilus gone, everything is quite snappy on my PIII450. Of course "everything" is a lot damn smaller now :).

Re:Gnome and KDE are more or less the same these d

[anpe]

Yeah! Lets be original, stop to copying Win32, copy MacOSX !

Sade??

[Anonymous Coward]

The guy listens to Sade. How much can he really know about computers?

He didn't even know enough to hide the Sade before taking the screenshot.

Re:Sade??

[Salsaman]

What do you mean ? Sade is *obviously* a big computer geek. What are her two most famous songs ?

Smooth Operator (an obvious reference to overloaded operators in OO programming languages); and Diamond Life (obviously a reference to hardened carbon based nano machines).

Re:Gnome and KDE are more or less the same these d

[bokketies]

True, KDE and gnome may not be innovators. But I am not looking for that right now, and nor will I in the near future. What I am looking for is

a stable desktop

easy to configure, out of the box solutions for multimedia

applications with fast response times

So far gnome seems to implement only the latter, while KDE scores points in the first two departments. KDE is becoming faster though. And yes, win2K (and possibly MacOSX) seems to own all these points, but is not

free, open source. as gnome and KDE are.

Ah, memories

[Gryphon]

Reading this review of Gnome reminded me of the days (about a year and half ago now) when I was still trying to use Linux and Gnome (somtimes KDE) as a desktop machine.

Nothing coherent about the UI design, hunting around to find configuration panels, getting messages that tell the user to download this package or that package (which leads directly to Dependency Hell).

These days, I use Mac OS X. Sure, it's UI isn't perfect. And I know, it's an apples to oranges comparison, Free Software to commercial. But man, do I ever enjoy using a coherent desktop with one place to change settings (System Preferences). No fuss! No muss! I'm far more productive.

And my Linux server continues to hum away in the basement, quietly powering my website.

Life is good.

Re:Ah, memories

[foobar104]

It sucks that font size is fixed.

Don't forget that OS X is still evolving rapidly. Lots of the stuff that people are bitching about-- including the inability to change font sizes in the Finder-- are already implemented in Jaguar.

If Apple had tried to implement every UI feature before shipping the first release... well, we'd still be waiting.

Re:Ah, memories

[foobar104]

These days, I use Mac OS X. Sure, it's UI isn't perfect.

Speaking of which, the coolest thing about the whole OS X development process to date is the way things changed between the public beta and the first release, and so on heading into Jaguar.

Lots of stuff was different in the first public beta. There was no Apple menu. The Finder didn't support multiple windows. Other stuff that slips my mind at the moment.

Rather than releasing the public beta as a publicity stunt-- that is to say, solely as a publicity stunt-- Apple actually solicited, accepted, and acted on user feedback. The first release of OS X featured a dramatically different UI because of it.

I just think that's cool.

Re:Ah, memories

[Pierre]

Hehe - I remember using FVWM and tkgoodstuff and being amazed when there was gui configuration at all :)

I have to agree about the configuration consistency. The multiple themes (GTK/Window Manager) is a problem that is going to have to be solved for Gnome to move forward as a desktop machine.

I can't imagine telling my mom to change themes, not not THAT theme, the other theme. You know the widget theme. A widget? Widgets are the buttons and text inside the windows. Ya the other theme is for the window manager. Oh it has it's own little buttons that aren't controlled by the widget themes......

Wow...

[Anonymous Coward]

By this point, I expected about a hundred Linux jockies on here personally attacking OSNews for this.

Anyway, sounds like Gnome 2 is a lot like Gnome 1... very amateurish and lacking the 'polish' of the commercial OS's, especially where the help files are concerned. At least the fonts are better and Nautilus seems workable but from reading the review, it's nothing to write home about.

Re:Wow...

[MadAhab]

Boring. Blah, blah, blah, oh, my computer.

Yeah, I'm sticking to the minimal but polished and useful WindowMaker for now. It doesn't hold my hand, but it doesn't get in my way. If I could settle on a good file manager, I'd forget about gnome and think of it as an application framework for some really good programs. I could care less about the whole desktop thing.

10+ mod points to whoever gets the reference in my first sentence.

Re:Wow...

[eddy the lip]

Couldn't agree more....I don't understand the whole desktop environment obsession thing. I don't work more effeciently with them, and they clutter my screen.

I used to have little time for file managers either until I discovered Rox. Which you've probably tried, because it does seem to be all the rage lately, but if you havne't, do. I didn't think anything would get me using a graphical file manager again, but this did.

(crap - I know I'm missing some important reference with the first sentence, but even google couldn't help me fake it).

Re:Gnome and KDE are more or less the same these d

[weo]

You are way off base here. I have been using linux since '93 and saw all of the great advances in Window Manager design. Going from TWM -> FVWM -> elightnment -> windowmaker each went far and away improvements to usability and features. OSX is now my favorite UNIX desktop. It takes windowmaker and nextstep to another level. Plus it is all scriptable with applescript. Applescipt is like having shellscript for GUI... very slick. As for performance every new iteration of 10.1.x releases have shown alot improvements, evem on older hardware.

There is a rule in every industry...

[Raleel]

that got described accurately in the software industry. Don't use the .0 version.

The short of it is that it's still a tad beta, and still needs work. It looks gorgeous, but then again, I'm a Gnome user (KDE feels too bubbly to me, and I can't find a theme to trim it. yes I've looked on a couple of sites). it still needs to be tweaked. I'm sure the Ximian guys will have a very solid version in not too long.

Perhaps left-over G-1.4 settings?

[deaddeng]

I've been using Gnome-2 betas/RC1-2, and found that I must erase/rename all the standard Gnome configuration files (.gnome, .gnome2, etc.) in order to avoid problems and see the default desktop that the developers intended.

Re:Gnome and KDE are more or less the same these d

[MaxVlast]

For the love of god, don't copy Mac OS X! It will inevitably be awful, ugly, make a thousand geeks cum in their pants, and annoy everyone who's actually used OS X.

For example: Check out the Mail.app tries-to-look-alike in GNUStep. Most people think that it's a reasonable clone of Mail.app from NEXTSTEP. It's not by a longshot. The thing is butt-ugly, misses all of the GUI tweaks that make the latter a sheer delight, and feels like it's going to completely give up and go away at any time. I hate most Linux GUIs. They all seem to be unacceptably fragile and have the artistic sense of an unguided tractor. I simply feel that if I click too much the program is going to crash. That's not a feeling I want in software on which I'm going to rely.

Re:Gnome and KDE are more or less the same these d

[t_hunger]

You are invited to join the Berlin (soon to become Fresco) project at http://www.fresco.org/. We are going very slowly these days, but we are trying to do something new. We can do all (even more) then MacOS X can with our architecture. Of course lots of stuff is missing. We are not ready for even the most adventurous of users, but we could definitly need some developers.

Regards,
Tobias

Quick Analysis

[Leimy]

GNOME-1.4: Still hard to figure out when you first sit down on it.... I personally had trouble changing an Emacs icon to use Xemacs and ran around looking for a "property list" for it... I think you have to manually edit some text file is what someone said... I stopped using GNOME immediately.... That's no way to do a GUI IMO.

KDE3.x: Slow... very slow. Too many virtual functions need to have code relocated at runtime. Luckilly This site is addressing some fundamental linking issues with C++ [among other things] on GNU tools. In fact the GNU tools are starting to be built with some of these optimizations now as was evident on my RedHat box at work. FreeBSD needs to try to do the same since its my main development platform [luckilly its a dual Athlon MP 1600 so *nothing* seems slow there :)]. There are also a few UI issues like the Author of this article suggested but I must say that people want a snappier [speedwise] desktop and don't want the power of an industrial strength server just to run their desktop. [note: I love KDE... I have committed code to KDE... this is as objective as I get :)]

I spend most of my time on Mac OS X. The concept of being able to run the Microsoft Office Suite [which I actually don't yet on my Mac] on a Unix environment with 75% or more of my favorite tools either in place or on their way is very attractive. Let's face it nothing does DOC like Word [thank god!] and for compatibility purposes with all of my coworkers there just isn't a real substitute for everything it does. We use the revision control built into Word and other things so please don't offer Abiword, StarOffice, OpenOffice or KWord as alternatives. You can suggest till your blue in the face but you can't make my company change its stance on what tools must be used.... Its a fight not worth fighting based on my experiences with the alternatives out there. [I write a lot of stuff in LaTeX now... then I cut n' paste to Word when I have to... Time consuming and stupid yes but I don't have Word for OS X yet... :)].

I never got around to experiencing BeOS first hand but I heard it was a thing of beauty... There has been a fair amount of talk about adding the BeOS file system to OpenDarwin's CVS but I don't think anyone has committed the time to it yet.

Advice to KDE: Please please please don't get too bloaty... [application duplication seems to be a bit of a problem there... Why does the standard source distribution have to include these things anyway?] I love IOSlaves and KParts and think they are uber-cool but the end user doesn't give a shit about any of that because it doesn't directly enhance their experience... just gives the developer a woody.

Advice to GNOME: As a developer I do not agree with C as the tool for doing Object Oriented Code... especially when the manner in which things are being wrapped up is very C++ like. GTKmm has a long way to go before it can do what Qt can last I checked so I think that if you code for GNOME and want full access you must use C [correct me if I am wrong please... its been a while and I want to be as fair as possible]. I have to agree with some of the Author's UI comments if his experience was authentic and correctly reflects the actual situation. I still think GNOME is prettier than the KDE defaults but there are very good things coming in that respect it seems from what I have been able to follow on the mailing lists. [again I am unfortunately biased due to my KDE involvement].

Advice for OS X... yes.. sometimes you just have to realize that indeed your shit can also stink. The only major boo-boo I remember was the iTunes installer clobbering some linux partitions... That was naughty but obviously not a test case for Apple 'in-house' or it would have been caught. Live and learn! I understand some people have trouble with the lookupd for OS X dropping out on them from time to time [though I haven't seen this myself yet.] but that's not really a UI comment is it? Hmm, I guess keep doing what you're doing and maybe think about allowing users to pick schemes other than Aqua or Graphite in the appearances menu. Don't rush it though... I love the quality thus far and can deal with a minimal set of choices when it keeps the UI simple and straighforward [yes I still use the single button mouse on OS X because its actually possible to do so due to a good UI design around simplicity.]

I'd invite comments and criticism if I didn't know already that I was in for it.... so go ahead and get your shot in... I don't care - its only slashdot :)

Dave

Re:Quick Analysis

[scotch]

I personally had trouble changing an Emacs icon to use Xemacs and ran around looking for a "property list" for it...

Right click on the icon, select "Properties". What's so fucking hard about that? And you gave up on Gnome because you couldn't figure this out?

Re:Quick Analysis

[Leimy]

It didn't work that way in 1.4... Go try it then come back and tell me how fucking easy it is... Seriously do it. There was no properties option. I walked through this step-by-step with a friend who also didn't believe me when I made this claim... He said "Oh... THAT!".

Maybe I was unclear what I was trying to do. That's not the sole reason I gave up on GNOME... there was like at least 2 different ways to do desktop switching by the window manager and by whatever the hell service GNOME was providing to do it. Nautilis was awfully pretty but slow and resource intensive.... The other reason I gave up on GNOME was I was only trying to use it for a month at the request of my friends to see whether I could use it day-to-day at work.

The answer is "yes of course" but being a huge C++ fan and diving into KDE internals in my spare time didn't imply that GNOME would be the desktop of choice for me.

Maybe it was personal limitations of seeing how the pieces of GNOME worked together... [it seemed quite disjoint at the time]... KDE had a more universal feel to it across *most* apps... I won't say it was perfect but it "felt" better.

Re:Quick Analysis

[tempest303]

We use the revision control built into Word and other things so please don't offer Abiword, StarOffice, OpenOffice or KWord as alternatives

Ok, I suggest Crossover Office then. :)

It's good stuff, runs Word flawlessly. (yes, flawlessly.)

Re:Quick Analysis

[Leimy]

Nautilis icon... right click... properties
3 tabs Basic, Emblems, Permissions.

Which one chances the program being executed? I just looked less than 5 minutes ago and still couldn't find it.

Re:Quick Analysis

[foobar104]

I understand some people have trouble with the lookupd for OS X dropping out on them from time to time [though I haven't seen this myself yet.]

And you probably won't. I think that problem was fixed around 10.1. It was a pain while it lasted, though.

Re:Quick Analysis

[scotch]

Copy the nautilus icon as a button on your panel. Then you can change the app run ("Command" on the basic panel). You can't change they system menu entries, because they are system menu entries - not owned by you.

You might be able to modify the this menu behavior, I don't know (and wouldn't advise it if you could). Or maybe I don't understand your problem?

I use OS X

[johnjones]

I use OS X as my client now

why ?
it has a port of X (windows)
and it has THE best terminal app IMHO

other things are that fonts are far nicer than anything that I could get under X even when using the sgi font server

oh and ssh is in the box (alright then a update )

regards

john jones

Re:Is it possible to Cut'N'Paste yet ?

[proj_2501]

It's not a problem with X. X Window has a consistent clipboard interface. The problem is not your X server. The problem is that application developers do NOT understand interface design AND there are two schools of people crashing together to make programs.

I hope people does not totally trust this review

[GauteL]

Because it is on some points totally wrong. For instance, the speed-issue and the "no central place for configuration issue". Everyone else reports a speed-increase, unlike this reviewer.

The central place is just wrong. The dialogs the reviewers seem to suggest is kept "all over the place" is in reality in ONE place. No, there is no unified control-panel GUI for GNOME 2.0, like gnomecc in GNOME 1.x or the KDE-panel. This was changed because almost everyone hated the unified dialog, and actually it has some pretty large usability issues as well.

In GNOME 2.0 the configuration dialogs are seperate windows, much like in Windows. But the dialogs are ALL reachable from a centralized place (Like Windows 2000 and 9x, unlike Windows XP)

Secondly. GNOME has taken a very far step towards KISS (Keep it simple stupid) unlike some comments on here seem to suggest. Some of the comments seem to be based on the review, and not from actual usage.

The reviewer tries to make himself out as a GUI-expert, something he doesn't seem to be at all.

There are ACTUAL GUI-experts and usability exports working on GNOME. Of course there are still lots of little mistakes and bloopers in the GUI. But some comments here, and from the reviewers seem to suggest that this isn't thought of AT ALL. Which isn't the case.

When it comes to Galeon running. The reviewer states that he does not have GNOME 1.x libs installed, which could be why Galeon (which currently is a GNOME 1.x app) won't run. Even if he does there were several issues with earlier versions of Galeon with GNOME 2.x, which can be solved by upgrading Galeon. The reviewer doesn't state what version of Galeon he uses. This is thus most likely a Galeon issue, rather than a GNOME 2.0 issue.

The reviewer does have some valid points though. Especially a shortage on help-files.. though it isn't as bad as the reviewer seems to make it out.

One of the worst parts though is the notion that in GNOME 1.x you could turn off Nautilus for speed, but in GNOME 2.x you're left with a naked desktop if you do.

First. Turning off Nautilus for speed should be rather unnecessary except for people really short on memory.

Second. Of course turning off Nautilus gives you a naked desktop. Nautilus is the desktop-manager. Turning it off removes the desktop (apart from the background-image). This also happened in GNOME 1.x, except some GNOME 1.x installations was totally screwed up in the way that it ran BOTH Nautilus and gmc (the old GNOME file-manager) at the same time. And thus if you turned off Nautilus, the old gmc-desktop was shown. This meant wasted memory because you ran two desktop-managers at the same time. I'm a bit disappointed that there is actually an option in the GUI to turn off Nautilus, which will be difficult for Newbies to actually turn ON again.. but that is a seperate issue. People desperate to get rid of Nautilus, could do it via gnome-session-properties, and actually, as of GNOME 2.0 I don't see the point apart from feeling 31337.

GMC was never ported to GNOME 2.0 and probably never will, because it frankly made much more sense to just fix Nautilus speed-wise. Which has been done, and will continue.

NO menu editor?!!?!?!

[MicroBerto]

I installed Gnome 2 with Garnome and here's my reactions:

THERE ARE NO MENU EDITING CAPABILITIES

How in the *blue fuck* do you release a window manager without the ability to change the menus! That's AWFUL, and there is absolutely no excuse for this. Gnome 2.0 should not have been released.

Yeah yeah, the speed is great and Nautilus is now usable. But expect a lot of shaking-up for Gnome in the next 6 months, becuase the UI blows and the configs are impossible and you've seen all of the other [correct] posts about how the devolopers need a REAL ui expert.

Re:NO menu editor?!!?!?!

[Bandman]

hey, at least it was intentional? Could have been worse.... "oops, well damn"
kinda like "hey Bob, great design, where are the bathrooms???"

Re:NO menu editor?!!?!?!

[An Onerous Coward]

According to the Gnome folks, it was better to rip it out than release it with severe bugs. They've promised a fix by 2.0.1, and in the meantime just edit them by hand. And don't give me any of this "My grandma could never figure out how to do that" crap. What are you thinking subjecting the poor woman to a .0 release in the first place?

Dropping buggy features is a good thing. Releasing a product promptly is a good thing. Releasing a product without an important usability feature is a bad thing. Two out of three ain't bad. Even if you think it's an obvious blunder, try to keep your criticisms polite and constructive. It's a volunteer effort.

Re:NO menu editor?!!?!?!

[MicroBerto]

I don't recall mentioning my gramma in my post. I consider myself a smart guy and setting up gnome2 was still a pain in the ass. Not impossible, but not friendly.

Re:NO menu editor?!!?!?!

[MicroBerto]

Umm, no.... I am talking about general configuration of gnome.

First off there is no unified setup anymore. Second, half of the preference tools segfault on crash for me (yes, i have submitted bug reports).

Like gnome 1.0, this is probably going to take till the 53rd release before it's nice.

the usual whining

[g4dget]

Every UI designer will laugh at this

UI designers laugh at lots of things, most of them completely irrelevant. In this case, the author is complaining about some baroque scheme for the theming UI. But theming is optional--you don't need it. People play around with themes when they are bored; it might even be bad if the theme configuration UI is too slick.

Gnome 2 does not come without its problems. I do not have sounds on my Gnome 2. I think that Gnome 2 assumes that you have Gnome 1.4 installed,

That's an issue with packaging, not Gnome2 itself. The same goes for many of the other grips that the author has.

The new version removes the flexibility found on Gnome 1.x and it does not introduce anything really new or spectacularly interesting in its UI design.

If the translation of this is "it has fewer options to confuse users and it didn't change its look or feel significantly so that people don't need retraining", then that sounds perfectly reasonable to me.

Overall, I didn't see a single substantive or informed criticism in the article. There probably are plenty of things wrong with Gnome2, but we'll have to wait for a more careful write-up of those.

Also, you can't expect too much from any desktop that follows current paradigms. Windows and MacOS have plenty of warts and problems, too. Overall, in my experience, Gnome and KDE are no worse.

Re:the usual whining

[rjstanford]

From the review:

Gnome 2 does not come without its problems. I do not have sounds on my Gnome 2. I think that Gnome 2 assumes that you have Gnome 1.4 installed

From the previous post:

That's an issue with packaging, not Gnome2 itself. The same goes for many of the other grips that the author has.

And a more perfect example of the kind of geek-superiority that people accuse the Linux community of I don't think I've seen in a long time.

Here's a newsflash for you: people don't care about technical quibbles like this. Imagine going to a new car dealership and finding out that the car you're looking at doesn't come with a Stereo. Upon commenting on this fact, people tell you, "Its still a good car, that's a packaging issue -- check out the engine timing! Besides, look at the huge hole in the dash into which you could install your own stereo (or even two or three of them) -- most of the wiring is even there for you!"

This review spent a lot of time talking about usability. If somebody does the "normal" things when obtaining a package, especially one designed (as a desktop is) to hide complexity and produce a more usable system than before it was installed, and it doesn't work as it could be reasonably expected to -- then there is a problem.

Sure, the internal code may be fine. But from a user's perspective, Gnome is Gnome. You could have the best algorithms on the planet, but if they're not enabled, or not included until you do some steps that only the developers know about, nobody will care. Okay, this is a packaging issue. You know what? This was a review of the whole package -- code, help, defaults, et ceteara. Including packaging.

Then again, far be it for anyone to offer a rational objection to a favored OSS project... That might lead to open discourse or, worse yet, improvements. Shame on all reviewers.

-Richard

Re:the usual whining

[g4dget]

Here's a newsflash for you: people don't care about technical quibbles like this. Imagine going to a new car dealership and finding out that the car you're looking at doesn't come with a Stereo.

Here's a newsflash for you: Gnome2 is the radio, not the car. We know the radio works because we can hook it up to power and it plays music. If you get the radio with your new car and the radio doesn't work, it was installed incorrectly. Sensible people complain to their car dealer about that.

But from a user's perspective, Gnome is Gnome.

There is nothing in the world Gnome developers can do if the Linux distributors screw up the installation. Sensible people complain to the people with the power to fix a problem; venting to some hapless bystander is pointless and aggravating.

And a more perfect example of the kind of geek-superiority

Actually, you just gave us another example of populist cluelessness.

Re:the usual whining

[eddy the lip]

I haven't tried Gnome2 out yet (and I'm not likely to use it, given my preference for blackbox), but based on this review, I'd at least look at it to recommend to others.

Any review of a desktop environment that complains how "unintuitive" it is (read: I wouldn't do it that way), declares broadly what UI designers will think of it (when the writer obviously isn't one, or he wouldn't be complaining about lack of configuration options or new gee-whiz features) is wholly untrustworthy, and probably 180 degress from what I'll think of it. So now I'm curious.

Re:the usual whining

[g4dget]

People who read the review now know of the packaging issue.

No, they don't, because he doesn't even tell us what distribution is running. If I install Gnome2 when it becomes part of the standard RedHat or Debian install, this problem will almost certainly not be present.

Should he not have mentioned a serious problem he encountered in his review?

He should do a review competently. That includes starting out with a clear statement of what he was testing and how he was testing it.

Re:the usual whining

[Nailer]

That's an issue with packaging, not Gnome2 itself. The same goes for many of the other grips that the author has.

That packaging is considered not a part of Gnome 2 is an issue.

Re:Gnome and KDE are more or less the same these d

[Avakado]

I'll try to be a bit more constructive in my criticism this time:

Aqua's keyboard navigability: It's well known that keyboard shortcuts will improve your efficiency when using a GUI. Every single part of a GUI should be accessible via the keyboard, so that experienced users can be as effective as possible, using these. These shortcuts must act consistently throughout the entire GUI, and properly marked (like underlining the character that is part of a shortcut in menu items). Moving from one widget to the next, scrolling, opening menus, starting applications et.c. should all be possible via the keyboard. Text widgets would also benefit from having more shortcut keys, like ^U for "kill line", ^W for "erase word" et.c. In many of the applications of MacOS X, most of this functionality is non-existant.

Multiple desktops: it's obviousely an advantage to be able to have multiple workspaces running at once. Users not wanting this feature can easily refrain from using it or disable it (or not enable it). Aqua does not provide this feature at all.

Configurable look (themes): if you for some reason can't stand the default look of Aqua, or want any other color than blue or gray, you are out of luck. As far as I've been able to tell, there's no way to change the appearance of GUI widgets (beyond the colors blue and gray), as opposed to virtually all open source widget sets I've seen. You might argue that themeability slows down the GUI, but that can easily be resolved by providing a binary interface (i.e. styles are dynamically loadable libraries) like KDE does (and Mosfet Liquid and Keramik use).

Scriptability: You mention AppleScript, and claims it is like having shellscript for GUI. No it isn't: you are bound to use that specific language. They could easily have supplied a network protocol (like KDE's DCOP) or any other more generic interface. Since they didn't, everything has to go to this dreadful language. Any experienced programmer would instantly fear "an easy-to-use, approachable, English-like language".

Stupid messages: "You need to click here to continue" (why not just friggin' do whatever needs to be done, instead of requiring user interaction at every possible step?), "An error has occured" (did you ever hear about strerror()?) and similar. While many of these aren't severly obstructive, they are nevertheless very annoying signs of sloppy programming and interface design.

Widget usefulness: in certain applications (most notably the QuickTime player), completely untraditional widgets are used for the sake of visual appearance. Many of these widgets seem like they're designed to be handled with a physical hand, and not with a pointing device and keyboard (like knobs and switches).

Re:Gnome and KDE are more or less the same these d

[The Faywood Assassin]

Well, I would tend to agree with you, but Apple loves to send out their lawyers everytime they even THINK some one is copying their GUI design (remember the AQUA theme fiasco?). I don't think any developers are willing to invest time and effort to incorporte "OS X-like" ideas into their work just to have Apple's lawyers tell them that they have to scrap the whole thing under threat of "look and feel" violations.

GNOME or KDE does not mean much to me now

[r6144]

As a user who have about 3 years of Linux experience, what I need is just X, some window manager that let me have 10 virtual desktops and switch between them easily (quite a few does now), and some pretty widget libraries (gtk or qt does their jobs, although tk and others are okay too). So gnome (or kde) has never meant anything to me other than a lot of potentially useful libraries. The included applications are of little use. If the configuration is difficult to use, I configure my window manager only once, anyway.

Re:Gnome and KDE are more or less the same these d

[shadow303]

Well, you just can't please everybody. If you copy windows, people complain that you are not innovating. If you do something original, then people say it is non-intuitive and hold it against you.

If it goes on like this...

[idletask]

GNU/Linux, BSD et al will simply never make it to the desktop. Both KDE and GNOME have constantly failed in designing a good GUI (in case you'd wonder, yes I've used them both, alongside with Win9x and MacOS 9 - the latter got the lead in usability, but that shouldn't be a surprise). Keep in mind that the guy who wrote this review is not even an average user. A real Joe user wouldn't even have bothered to write a review given the poor shape of the thing.

Designing a good GUI requires everything that a geek doesn't have: notions of ergonomy (this goes for graphics too - GNOME icons may look very cool, but their ergonomy is disastrous) and psychology, the ability to sit on behalf of the user, and most of all, not the slightest care about how it works behind the hood.

Before KDE/GNOME can reach the "usable" qualification, both need a team of GUI designers whom the programmers will *listen to*. The rare persons who have some qualifications in this respect are constantly being bashed by coders who say "That's not how it's supposed to work [in the code]". Said coders therefore don't understand that even if their code is well written and would deserve some note in software engineering courses, it fails at its primary mission: meeting the users' needs.

The second biggest problem is the existence of GNOME itself. While it was kind of justified given the licensing problems with Qt at first, it has been obsolete from the day when Qt got GPLed. But geeks have their pride. Too much pride. Result: code duplication, well designed toolkits of all sorts but still no consistent GUI on either side.

In 5 years, Microsoft went from the somewhat clunky but usable win3.x series to the very usable (even apart from the "but-it-comes-preloaded-everywhere" argument - that's one I don't buy, sorry) Win9x series. In 5 years, the MacOS GUI has evolved very little - an evidence that it was built from the ground up with usability in mind (MacOS X is another matter, but I bet the guy who made it is certainly not the one who did the Mac0S 9- GUI). In 5 years, what has most evolved in our two main contenders are the toolkits. Who cares?

Re:If it goes on like this...

[The Asmodeus]

It sounds as if you've got a good handle on usability. I look forward to seeing YOUR contributions in Gnome 2.1.

Oh, you're not going to be contributing?

One note for all you self-appointed GUI experts out there, if you know so much, help out. If not, then shut up because your bitching is on the level of a person who didn't vote then complaining about the President.

Re:If it goes on like this...

[Phoenix Rising]

Designing a good GUI requires everything that a geek doesn't have: notions of ergonomy (this goes for graphics too - GNOME icons may look very cool, but their ergonomy is disastrous) and psychology, the ability to sit on behalf of the user, and most of all, not the slightest care about how it works behind the hood.

Um... The folks at Eazel, who did the initial Nautilus work, were the same people who did the GUI for the original Macintosh. I think they have some clue and background in ergonomics and Human Interface Design. Additionally, Sun has been providing resources for HID in Gnome2.

Just because you don't like it, doesn't mean someone hasn't taken time to do the research. Gnome has formal Human Interface Guidelines and a team devoted to seeing them implemented. And it's a team with experience, not just a bunch of hackers who think they know something.

Perhaps you're a minimalist?

Re: KDE 3.0 slow...

[Leimy]

as a KDE supporter and very-very-very part time developer I am pleased to hear this experience was positive for you.

Great quote

[swagr]

From the review:
which is which and what each does? Good question.

Is that really a good question?

Un less you *use* your computer

[TomorrowPlusX]

What's with all the "my WM starts faster than yours" hoo-ha I'm always hearing?

Doesn't anybody actually *use* their computers? When I start (kde) I leave it up for days (on a laptop, no less) and I *use* it. I write code, lot's of it. I write mail. Etc etc.

However, I can't help but wonder, what do you types do? Do you just grab a stopwatch and repeatedly time how long it takes to start different WMs? Is the whole goal of modern computing to provide the most obscure functionality as fast as possible? As nice as fluxbox and windowmaker are, I'll take KDEs rock solid APIs and frameworks any day, even if they take ~30 seconds to start on my little thinkpad. But of course, *using* my computer isn't very l33t of me, is it?

Re:Un less you *use* your computer

[nagora]

I do use my computer all day for work. The speed of WindowMaker is indicative of its efficiency rather than a goal in itself.

KDE as an environment is bloated and slow and contributes nothing to productivity that WM doesn't do better.

The longer you have KDE (the very epitomy of "obscure functionality") running the more time you've wasted waiting for it.

TWW

Re:Is it possible to Cut'N'Paste yet ?

[Erik Fish]

This is my question as well -- and to the people saying "it's not a problem" well, it is. I'm using Ximian with all the latest updates and I can't even copy and paste between two GNOME apps (Terminal and GEdit) let alone from a GNOME app to Mozilla or Licq.

These days copying and pasting text is such a fundamental GUI feature that it should be unthinkable to release with it broken -- like releasing without the ability to minimize windows.

Yeah, right.

[cluening]

This sounds more like somebody that doesn't know much about writing (learn to use commas correctly!) who performed a bad install on a Redhat 7.x machine. Some of the problems described are obviously just RH's Gnome 1.4 setup conflicting, which is no big surprise. And many problems sound like he just didn't bother thinking while using the system. Whoever came up with this idea that "we shouldn't have to think to be able to use a computer" should be beaten with something large. You have to think to use any other machine, right? Even I have to spend a moment thinking about what darkness setting I want my toaster set at. Part if living is thinking about things, people!

sawfish 2.0

[Ender Ryan]

My experience so far has been a mixed bag. I like some of the simplification, they've tried (with only some success) to use saner defaults and make everything easier to configure and use.

The panels work ok, but there are some serious quirks. If you make a panel with no menu, and you remove the hide buttons... you can no longer configure that panel. Only way I found to fix it is to add a menu to another panel, drag it to the menuless panel, then you can use the menu for configuring the panel. A pretty large oversight if you ask me.

The other extremely annoying panel problem is... on logout/login, the panel completely forgets the order of the launchers on it! If that's not a huge oversight, I don't know what is.

Now for my biggest gripe. Sawfish 2.0. Someone was smoking some serious crack. I don't mean to be mean, but it has absolutely been destroyed. It is completely useless. It plain sucks, terribly. First of all, it's crashy, very crashy. See the bugzilla database on gnome.org, serious crash bugs in sawfish 2.0, definately NOT release material. Second, sawfish was designed with extreme configurability in mind, every aspect of sawfish is meant to be configurable, but now they have completely removed 90% of the configuration options. They supposedly tried to choose sane defaults, but with something as configurable as sawfish, that's simply not going to happen. There are some serious problems with the default settings. The new sawfish control panel... what can I say, it plain sucks. The tabs are across the top now, and you have to use the dumb little arrow buttons to scroll across the stupid things. This makes it an extreme pain to search for settings.

No favorites menu. I always found this very useful, I always put all the little utilities I often use in there. It's gone, and there is no equivalent replacement. Now your stuck browsing through the damn apps menu. A very poor decision in my opinion.

Now those problems are all extremely annoying, time draining, and basically make gnome 2.0, simply put, not ready for prime time. It's simply not release quality at this point, not even close.

There are some positive aspects though, quite positive actually. Fonts, gnome2.0's font rendering is really, really great. Fonts are rendered very cleanly, not blurry looking, and not jaggy, they look very good. Speed, despite what the reviewer was saying, gnome 2.0 is pretty speedy, speedier than 1.4. It loads up really quick, probably 4 or 5 seconds on a reasonably fast machine. The menus are much less cluttered by default, a plus in my book, they were simply full of junk before. GTK+ 2.x is much better. The default theme actually looks pretty good, file selectors work better, save dialogs don't wack the filename when you change directories(!).

All in all, I have to say that I'm pretty disapointed. It's not a lost cause, but it seems to me that gnome may be heading in the wrong direction.

And that's all I have to say about that.

Re:sawfish 2.0

[qweqwe]

The main reason Sawfish 2.0 sucks is that no-one is working on it right now. It's based on the old GTK+ architecture and despirately needs a rewrite. Metacity is what GNOME (at least Sun GNOME) will ultimately use. It's currently more limitted than Sawfish, but it's really great. Try it out!

As for the 2 panel quirks, please report the bug to either GNOME or Ximian (who's going to release Ximian GNOME 2.0 soon). It should be *really* easy to fix. It sounds like a bug that no-one noticed. If you're quick, it might end up in the next Ximian or GARGNOME update.

Re:Gnome and KDE are more or less the same these d

[Dixie_Flatline]

I agree with you on some of your points, but others are pretty touch and go.

1) While multiple desktops are handy, saying that they're 'obviously' an advantage is abusing the term 'obvious'. Obvious to who? You? Me? The average X11 user? Joe Sixpack with his iMac? Your grandmother and her iBook? Adding an extra UI 'feature' like that (by default) is just confusing to the average person. The Mac tries to present a simple, friendly interface, and such a thing would be decidedly confusing to anyone that thinks Nascar is a sport. People that want it will find a way to get it. Such a UI enhancement is under development by independent developers right now.

2) Themes are not actually useful. Anything other than purely aesthetic themeability (ie. the theme changes nothing other than some colours) is bad, in terms of UI design. The reason why everyone copies Windows' UI is because it's familiar. Uniformity of interface is a BIG DEAL.

If you're just talking about colours, is it really that big a deal? I'm just reading my mail and ssh'ing to my mail server. I don't care what the window dressing looks like, that much.

3) You think that programmers only like hard-to-use, unapproachable, syntactically impenetrable languages? I would argue that Smalltalk is easy-to-use, approachable and occasionally 'English-like', and I don't have any problems with it. I've never used Applescript, but as an experienced programmer, I don't think you should be making generalizations like that.

4) I agree that modal panels are foolish, but Apple has sort of met the user half-way. Ideally, what Apple would do is USE that fancy alpha-blending UI, and drop a translucent panel down explaining the situation while it did the right thing. The panel wouldn't change the focus of anything, and the user could easily ignore the panel while it hung around, and work right through it. However, if you ever talk to an ordinary user, they hate having their machine do things without telling them. They LIKE feeling a bit involved. If you pop up 20 modal panels with an 'okay' button on them and nothing else, they'll get irritated, but they want to feel like they're in charge. If the machine starts going off without them, they start to resent it.

More or less, I agree with your assessment, like I said. I didn't see the parent, but I'd assume that it was trying to defend Aqua.
Aqua is a fine interface, and it's clear that a fair amount of design went into it. Personally, I think that THAT is the real lesson that we should take away from it. It doesn't do everything perfectly, fine, but at least it wasn't just thrown down by a programmer that was too lazy to actually read some interface books, which are what the Windows, KDE and GNOME interfaces feel like to me. I use (and like!) GNOME, but it's clear that almost none of it is thought out to any greater extent than 'Windows does it this way, and X11 does things this way. Let's go!'

Says it all

[minkwe]

and I compiled it with -03 and -march=i686 using gcc 3.1.1-CVS on my Mandrake Cooker
Using a buggy compiler on a buggy distribution to compile gnome, and then going on to rant about the result like this, I'll say he/she has an agenda here which I dare not mention.

Re:Says it all

[Cardhore]

I'm using Mandrake Cooker with gcc 3.1.1. GNOME2 compiles and works. In fact, everthing included, minus the software that is under development, like the Mandrake tools, works fine with the new compiler. This included getty, bash, linux (the kernel), Xfree, libraries, galeon, mozilla, gnome and kde.

Depends on your usage pattern

[EnglishTim]

I don't leave my computer at home on 24/7 because

a) It's noisy
b) It seems like a waste of elecricity to leave it on when I'm not using it

Therefore, speed of start-up is quite important to me*. I leave my computer at work on all the time because it gets put on the render farm when I'm not using it.

* One of the things I love about WinXP is the hibernation feature - turn the computer off - it saves the memory and powers down - turn it on, it restores it to exactly what you were doing before you turned it off.. lovely!

Re:Depends on your usage pattern

[Junta]

I do that with linux too, its called the swsusp patch, a google search can show the way, very convenient, especially for laptops without BIOS-managed suspend modes.

Re:Depends on your usage pattern

[Eil]

On all of the computers I've tried it on, hibernation took only marginally less time than a cold boot to start up and about 5x the time to shut down.

Probably has something to do with writing 200MB (or more, depending on how much memory you have and what's running) of data to disk on every shutdown. Also, I'm fearful that reading and writing hundreds of MBs of data to my disk on every shutdown and boot would cause extra wear and tear on my disks, which I can't afford to replace right now. (Data-wise and money-wise.)

I would imagine hibernation is useful in the rare scenario where you're in the middle of something important and *have* to shut down and don't want to lose your place. But beyond that, I'd rather boot the old-fashioned way.

Re:Depends on your usage pattern

[Eil]

b) It seems like a waste of elecricity to leave it on when I'm not using it

P.S.: Your computer's not wasting electriticy if it's researching a cure for cancer. :)

Re:Depends on your usage pattern

[EnglishTim]

Hmmm... well, on my 512Mb system, it seems pretty quick to start up - the thing I like is that it also goes straight into that 'everything loaded up stage'. As for shutting down, I have the power button on my computer configured to hibernate, so it's just a case of hitting the off switch and then walking away. It hibernates, powers down and that it! love it ;)

Say, (and this isn't a troll/bait, just curious) does Linux have that kind of thing avaliable?

Re:Depends on your usage pattern

[Eil]

Say, (and this isn't a troll/bait, just curious) does Linux have that kind of thing avaliable?

There is reportedly a third-party kernel patch, but I don't recall if it's even stable or active. I seem to remember it being targeted towards certain laptops too, so maybe this patch I'm thinking of is just to make particular laptop BIOSes not screw up on APM standby or whatever. (Like mine always did with Linux.)

Doesn't seem like a generic "hibernate" patch would be all that hard for a moderately skilled kernel hacker.

Give it some time

[jaaron]

Look, I'll admit that I really like GNOME as far as desktops go. More often than not I end up just using blackbox or evolution or windowMaker, but I do like the GNOME desktop and I've been looking forward to the 2.0 release. Anyways, I'd like to offer the thought that it's too soon to be judging GNOME 2.0. A lot of the apps aren't ported to it yet. Distributions aren't shipping it yet. A project like Gnome isn't like Mozilla where you expect everything in one package. There's a lot of other projects, not officially part of Gnome that go together to make it. When all these parts have been put together and companies like Ximian and RedHat start shipping a complete Gnome 2.0 product, then I'll start getting critical with it. Until then, I think it's too early to pass judgement.

Re:Gnome and KDE are more or less the same these d

[pldms]

Ok, that was quite a list. I'll bet someone has replied as I type, but here are some replies.

Keyboard. Hmm, try turning on 'full keyboard access' in System Prefs. You can now hop around the UI using just the keyboard. As for delete word etc. try emacs equivalents (work in all cocoa apps at least). There are alternatives as well. I just use those since I devote brain space to the damn things.

Multiple desktops. space (http://space.sourceforge.net) does a little of what you want. However I agree, Apple should add it themselves.

Themes. Colors are ok, but I generally against the ui makeover that some apps seem to delight in. They usually just cover for faults in the original UI. (not an original pov, I should add).

Scriptability. I think you should look at scripting again. There are many languages for scripting, including (IIRC) javascript. They just hook into AppleEvents which provides the underlying functionality. They can also go over the network (see sharing - allow remote apple events).

Stupid messages. I'm surprised by this, since Apple are generally pretty good at this. But they aren't perfect. I think asking whether it is ok to continue is fine, but I don't know the details of the case you cite.

Widget usefulness. Let's hear it for the volume dial on QT 4 (was that the one?). Terrible. They seem to have been fixing these (QT has certainly improved). The worst offenders seem to be Apple's own media apps, which is pretty bad.

I think some of your points are valid, but OS X is generally pretty good. They seem to have half an idea about this stuff.

Driven from gnome to kde...

[mjh]

I'm not trying to start a flame war here. I just think that it's time for me to express my opinion on this matter.

I've been tinkering with gnome and kde since pre kde 1.0 days. I have always preferred gnome to kde. Not because I thought gnome was prettier, but because I could get the functionality that I wanted out of gnome and couldn't get it out of kde.

With the advent of kde3 and gnome2, I will be switching from gnome to kde. Is kde3 slower? It doesn't feel slower to me than gnome1.4. Is kde3 prettier? I think mosfet's liquid is stunning. Can I get kde3 to do what I'm used to doing in gnome1? Not 100% but closer (maybe 90%). Can I get gnome2 to do what I'm used to doing in gnome1? No. I'd say about 50%.

So, from a functionality point of view, gnome1 wins and kde3 is a close 2nd, with gnome2 a distant 3rd. From an aesthetic point of view, kde3 wins, and flip a coin between gnome1/2.

So I'm switching to kde. IMHO, gnome is just not going in a direction that I like.

Remember, this is my opinion. I'm not trying to incite a flame war. I'm just a lone user letting the gnome developers know that they just lost me.

Re:Driven from gnome to kde...

[Pierre]

I haven't tried Gnome2 yet.

When you say that you can do 90% (KDE) and 50% (GNome2) of what you normally do what do you mean?

Can you give an example?

For me I with a little fiddling I can do 100% either way, but maybe my needs are that complex.

Is it desktop setup or applications that you can't do?

Re:Driven from gnome to kde...

[Pierre]

that should have been that my needs are not that complex

Re:Users not changing thier Desktop?

[mccalli]

What planet is this individual from?

Planet Corporate. In these environments, users have locked down desktops, each checked for consistency with a central server via logon scripts, and are rarely able to make many decisions for themselves.

For example, where I am it insists on sticking MS Office icons on my desktop even though I mainly just use Exceed. Delete the icons? Certainly sir, but then they reappear the next time you log in.

The reviewer is absolutely correct on this one.

Cheers,
Ian

Re:Gnome and KDE are more or less the same these d

[TRACK-YOUR-POSITION]

It appears obvious to me that people claiming the MacOS X GUI is intuitive have either not really tried it themselves, or never tried anything else.

As someone who uses Mac OS X extensively after much Windows and X experience, it appears obvious to me that anyone complaining about OS X's GUI was too attached to the horror that was OS 9. The animations can be turned off, later versions of the OS will be faster, and you're simply speaking nonsense about it being obstructive or non-intuitive.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

Before using Mac OS X, I used Linux for a couple years, and then found NeXTSTEP and OpenStep. Mac OS X hasn't annoyed me since DP4. I used OS X as my primary OS for about a year, until I recently switched back to Linux for a number of reasons, none of which had to do with OS X sucking or falling short.

Most Linux GUIs do suck. I've used Mac OS Classic, OS X, Linux, and *STEP extensively. Naturally, NeXTSTEP 3.3 is the epitome of all that is good (except POSIX compliance). Mac OS X is in the second place. BeOS wouldn't probably be up there if I ever had used it for anything.

If all these GNOME and KDE people are set on ripping something off, we'd all be better off it they were ripping off OS X than Windows. However, poorly copied OS X features could be quite disasterous. Far more than poorly copied Windows features.

Re:Gnome and KDE are more or less the same these d

[civilizedINTENSITY]

Multiple desktops are such an obvious advantage that I can't believe they aren't as prevalent as overlapping windows. Its all about being able to categorize when I organize. I am amazed that this wouldn't be considered obvious.

Themes are useful to people who spend alot of time in front of their computer. Changing the appearance without breaking the pattern of functionality is stimulating. It prevents a form of "highway hypnosis". Its fun. Its pretty.

Your parent posts' keyboard shortcut concern is also of concern to me. Keyboard shortcuts are essential.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

I've tried them all. OS X was my main OS for a year. As a desktop, I'd rank it #1 over the other available desktop options. I didn't find it a real pain to use, not 10.1 at least. DP4 was. I found it visually appealing, yet productive. Nothing wrong with that.

You may not like OS X. That is swell. However, it doesn't mean that it's cumbersome and annoying for everyone. Simiarliarily, there are many of us, including myself, who find GNOME, KDE and Windows a helluva pain in the ass to get anything done with. I've used them. They suck. Doesn't mean they can work well for you.

Re:Gnome and KDE are more or less the same these d

[civilizedINTENSITY]

Agreed: I was saving money to buy my first apple because of OSX, then bought an Athlon in disgust over the AQUA theme fiasco. It was the attitudes of online apple chat enthusists that drove me away as much as apple's stance. I decided they weren't something I wanted join.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

Huh? In one paragraph, you say that Mac OS X isn't intuitive. Then you say it is.

Indeed, I don't find Mac OS X's GUI to be much like Mac OS 9s. Ask any old-school Machead, and they'll tell you that it's nothing like OS 9, it's too much like NeXTSTEP. Ask any ex-NeXTie like myself, and we'll tell you it's too much like OS 9. It's somewhere in between. If it just stuck to pure NeXTSTEP or pure OS 9, more people would probably like it. OS 9's memory management is hell, but it's quite intuitive.

I used the original Mac 128

[HiThere]

Mind you, I'd seen the Lisa, so I had some prior exposure. The OS took almost no time to learn how to use. It that's not what an "intuitive interface" is, then I don't know what the words mean to you.

OTOH, it was limited. The power included was amazing. It was also amazing how long it took me to figure out why cut and paste was a good idea. And I can't even remember what I used to do instead.

It must have take hours to learn how to use that early MacPaint. But that was during the same day that I first used a mouse.

Still, as intuitive as it was, it was limited. To make it easier to learn the mouse was reduced to a one button mouse. But this limited it's flexibility (not it's power, but rather the number of steps that it took to do something).

As the versions rolled by, the Mac interface got slightly more complex. But only slightly. After version 7.5 I pretty much stopped using it, so I can't talk about the more recent versions. Perhaps OS X isn't as intuitive.

However: There's always a trade-off in OS designs. (Well, several trade-offs.) If you design a maximally simple OS GUI, then it will take a lot of work to do something that wasn't one of the builtins, and it won't be customizable. If you build a maximally customizable GUI, then it will take time to learn how to use it, and different machines will do things differently (on the surface). Probably both end points are bad choices. The Mac leans toward the simple side. It's fast to learn, but limited in flexibility. The Linux GUIs lean toward the complex. They're flexible, customizeable, and ... well, they take time to learn.

Not note that I'm not claiming that there aren't subsets of the two GUIs that aren't equivalent. But those are subsets. Even the cut and paste on the Linux GUI is more complex than that on the Mac. And more flexible. And less automatic (this is being worked on ... and progress has been made, but to some degree it's probably inherrent.)

Both of these choices are defensible. More people come to Linux already knowing the basics of what a GUI is than came to the original Mac, so there's less pressure toward maximal simplicity. Except that there's an idealogical pressure that says "Any GUI command should be translateable into a shell command." The Mac didn't have this constraint. (And it showed when they tried to back-patch in a scripting language [AppleScript].)

Original??? It can be good to be original. More often it means making a new series of mistakes. E.g., I prefer to use tried and true sorting algorithms. I don't go around inventing new ones. I'm not being original. But I can use libraries, and numerical analysis routines, and run-time estimations (well, it's a sort of around 50 cases, so I could use a bubble sort, but it's easier to just use the one built into the languae).

Originality is vastly overrated. It's something that should be used when it is needed, but one should always remember that using it is expensive. It takes longer to write the code, it takes longer to debug the code, it take longer to explain the code, it takes longer to.....

User-friendly is a much better concept than originality. The catch is, if you don't know who your use is, then you don't know what friendly is. And remember, the user base will change. Always. So it's also a concept of limited use. (N.B.: limited doesn't mean none. It means it only handles certain situations, so you need to be careful how and when you use it.)

Re:Is it possible to Cut'N'Paste yet ?

[Shelled]

Highlight the text with the left mouse button, centre-click to place the selected text. It's been this way forever in X. Jumping between Windows at work and Linux at home on a daily basis, I find the X solution far quicker and more intuitive.

And this is a surprise how?

[marm]

GNOME 2.0 has been rushed out of the door, just like GNOME 1.0 was.

The 1.4->2.0 development cycle has been a lot longer than originally anticipated, due to a big influx of developers (Thanks Sun!) and lots of core systems changing quite radically, coupled with some pretty piss-poor project management (now where have I heard THAT before?). In the mean time, KDE has been gaining a very large amount of traction as the most popular Linux desktop, and Sun has been wanting to push Solaris 9 out the door ASAP.

So GNOME had to release now, really, if they had any hope of keeping the users they have and for Sun to get Solaris 9 out approximately on schedule.

KDE underwent a similar change about 2 years ago, in the 1.1.2->2.0 transition, and not everyone was convinced then that KDE would survive, but it did, and look where they are now. Of course, KDE had the advantage of doing it first - although KDE 2.0 was far from perfect UI-wise, it had a considerable lead on GNOME in changing to a component-based architecture, so there was a very big Unique Selling Point for it at the time which GNOME 2.0 does not now have.

It took KDE 2 further major releases to turn the framework they built into a really nice desktop, and I suspect it will be similar for GNOME. The big question is whether the framework that was built for GNOME 2.0 will be good enough for their future plans... time will tell.

Personally I'm sticking with KDE3 for now. There are certainly issues with KDE, mostly in terms of speed and size, which themselves mostly stem from the choice of C++ rather than C, but these are being fixed one by one. KDE3 is now quite snappy, actually quite a bit faster than GNOME 2.0 on my Debian machine once you've got past logging in (all those double-buffered GTK+ 2.0 widgets are smooth and dandy, but they sure as hell ain't fast). Also, right now, KDE absolutely has the edge on both functionality and usability. Konqueror in particular is way out in front - indeed, for me at least, it's the best file manager on ANY platform. Nautilus is good, don't get me wrong, but Konq is breathtaking.

I'll reassess the situation when GNOME 2.2 is out. 2.2 should be the first mature release of the new framework, then we'll really get to see whether it's good enough to compete. I'm hopeful, a lot of the new framework looks good but either needs loose ends tidying up or needs someone to use it properly. Let's keep our fingers crossed. KDE is a class act to go up against though - they crank out the releases on time every 6 months, they seem to have a consistent vision of where they're going, they know where their flaws are, and they have yet to make a serious error. GNOME can't afford any more releases like this one if it wants to stay in the game.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

Scriptability: You mention AppleScript, and claims it is like having shellscript for GUI. No it isn't: you are bound to use that specific language. They could easily have supplied a network protocol (like KDE's DCOP) or any other more generic interface. Since they didn't, everything has to go to this dreadful language. Any experienced programmer would instantly fear "an easy-to-use, approachable, English-like language".

Way to do your research, lil buddy.

The AppleScript system is open. In fact, AppleScript just happens to be the default language Apple gives you to use within their "Open Script Architecture" (OSA).

For example, you could use JavaScript to tie into all the hooks AppleScript can. There is an older list of other OSA languages available as well.

As an experienced programmer, I find AppleScript useful. When I'm scripting a bunch of Mac apps, the english-ness and gimpy-ness of AppleScript has never bothered me. Why? Because I'm not doing any "real" work. If I'd like to do a combination of "real" work and scripting apps, I could easily use a language from the above list, or call the script events from C or a C module access by a real language.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

While I'm a user of hot-keys in Mac OS 9, X, and elsewhere and generally a keyboard navigation proponent, people can build up a muscle-memory for mousing as well. I'm not one, but I've known plenty of people, especially graphic designers (who seem to be more visual than myself) who work faster going to menus than I do by using hotkeys. Not because mousing is inherently faster, but because their muscle memory of using the mouse for certain operations is more developed.

Goes both ways.

Re:Menu choices [TOTALLY OT]

[tempest303]

This is UTTERLY, hopelessly off-topic, but I felt the need to comment...

Your .sig is a quote from chumbawamba, with the speaker asking that someone "give the anarchist a cigarette". While it IS possible to grow your own tobacco and roll your own smokes, aren't the vast majority of cigs made by gigantic, nasty, corporations? Does anyone else see the irony in this? :)

Re:Is it possible to Cut'N'Paste yet ?

[FooBarWidget]

Hello? Since when were you NOT able to Cut&Paste?

Open Mozilla and go to slashdot.org. Select some text and don't do anything else. Open gedit, *press the middle mousebutton*, and voila! The selected text is pasted.

Select some text in gedit, open kwrite or something, and press the middle mousebutton... and the text is pasted!

I can't understand why people say all the time that Cut&Paste don't work. It has ALWAYS worked for me.

Re:Gnome and KDE are more or less the same these d

[Anonymous Coward]

The animations can be turned off

Huh?

I use OS X. I use it because i like the functionality, i like the power it gives me, and i like the stability. I hate the interface. I hate it. And the only reason i hate it is that the animations and gooeyness can't be turned off.

Kindly explain to me how i can turn off real-time window resizing? Or even better, make it so that it's only realtime with a modifier key? That's the most sluggish, useless animation of all. It's useful sometimes, but the rest of the time it gets in the way and makes a basic UI task clumsy.

I have sight problems. I have *REAL ISSUES* trying to use an interface in which everything is blurry and antialiased and soft fuzzy gray. I get headaches, and my eyes start unfocusing uncontrollably, when i use Aqua for too long. It helps a lot that there are (technically illegal) third-party themes that i can download and install. Once i do that, 99% of my problems go away. However, i still have wierd, semi-dislexic moments where my eyes tend to go very slightly unfocused after a long time in an os x window, because of the borders between windows. The window borders in os x are very very faint gray lines with shadows. You can turn the shadows off, but there is no way to make the window borders thicker or darker. By any themeing method. At all. How can i turn that off? How can i get thicker, darkgray/black borders? OS9-style draggable borders would be best of all.

I like the transparent menus. But they just take too long to open. I want to be able to click on a menu and see it immediately. How can i turn off the menu transparency and windows, and make it so that they are a simple overlay rather than being rendered as seperate windows within quartz?

I access my computer over VNC a lot. VNC has this bug that causes it to screw up hardcore with some viewers when exposed to the throbbing OK button animations. How do i turn this OK button animation off, for my benefit when i am in VNC?

If i could just click a button and go back to the system 7 visual interface with OS X's power&stability, i would. But apple doesn't give me that option. Themeing may not be "useful" to most people, but for me it's NECESSARY to be able to turn off those omnipresent horizontal gray lines with text on top. It makes me angry that Apple takes such a paternalistic attitude toward this. No, people don't "need" complete interface control (though i need some). But when you get down to it, people don't "need" computers. We use computers because it's more enjoyable than using typewriters, pen&paper, etc. Apple is making that experience painful for some people by choosing to use an interface where whether you like it or not is a matter of opinion, and then not giving people the power of choice, when they have a perfectly good themeing architecture they refuse to release the specs to. At least with WinXP, you can go back to the non-bubbly interface style..

Re:Gnome and KDE are more or less the same these d

[zangdesign]

You're running up against the weight of history on the Mac GUI, which so far has proven pretty successful. Most of the common commands are accessible via keystroke and are the same in every program. Multiple desktops is not a huge requirement with Mac users, nor is the configurable look.

The one part I don't get is "Any experienced programmer would instantly fear 'an easy-to-use, approachable, English-like language'". Why? Because it's not arcane enough? Because people might actually understand what you're doing?

Go work on your own interface. Leave ours alone.

Re:Users not changing thier Desktop?

[EastCoastSurfer]

Any corporation that cannot even trust their employees to configure their own work environment is doomed.

It has very little to do with trust. It has to do with support. Corporation can't afford to support hundreds of users on PCs when everyone is installing random things off of the inet, mucking with fonts/colors/etc..., and bacially doing everything but working.

I know it is hard to think that the entire world is not just like you, but most corps are not made up solely of techies. In addition, in many of these lockdown corps the techs still get to do what they want with their desktop(since of course they control the desktop ;)

Re:Is it possible to Cut'N'Paste yet ?

[tialaramex]

All ICCCM compliant applications will work. We can't do any better than that, the ICCCM was published shortly after the last ice age, so anyone who hasn't read it, or didn't understand it is NOT OUR FAULT.

The GNOME developers cannot come to your house and fix the broken KDE 2.x install (which is the #1 reason for cut-copy-and-paste not working properly). Get KDE 3 and the problem goes away. That's not GNOME's fault.

The #2 most common complaint is caused by GNU Emacs, which has its own insane clipboard behaviour unique to that application. Use XEmacs instead if it bothers you, or complain to the Emacs maintainer. Not GNOME's fault.

The #3 source of problems is XChat, which does something moderately insane, I forget what exactly. If it bothers you, ask the XChat maintainer to fix it. Not GNOME's fault.

The #4 source of problems is Gnumeric 1.0, which has the same bug as KDE 2.x. This will be fixed in Gnumeric 1.2 even if I personally have to intervene.

If you find GNOME applications where Windows-style "Ctrl X/C/V" cut-n-paste doesn't work as expected, file a bug.

Re:Changing Emacs to XEmacs...

[Leimy]

Well two of us couldn't then... I will try it right now again in fact... I still have GNOME installed.

Re: KDE 3.0 slow...

[jonathan_ingram]

RedHat is probably the distro which is least likely to get you a decent KDE user experience -- mainly because they don't care about KDE. Indeed, lack of KDE was what lead to the creation of Mandrake (sadly, they've moved back toward GTK+ config apps in the newer releases).

After using RedHat and Mandrake for years, I finally tried out Debian, and was shocked at how quick KDE 2 felt -- subjectively at *least* twice as fast.

Ximian GNOME 2

[MAXOMENOS]

If people cant read release notes they should just pull down ximian RPMs

Funny, I tried to find them myself (on RPMFind, and I don't think they've been released yet. If they have been released, then they're not on RPMFind or, as of this morning, Red Carpet.

Does anyone know what's up with Ximian GNOME 2?

Re:Changing Emacs to XEmacs...

[Leimy]

I just tried again... Right click on the Nautilus icon gives me properties...

Properties has 3 tabs "Basic, Emblem, and Permissions"

None of the 3 tabs has a way to change the executed program... real intuitive.

Re:Gnome and KDE are more or less the same these d

[3Bees]

Aqua's keyboard navigability...Moving from one widget to the next, scrolling, opening menus, starting applications et.c. should all be possible via the keyboard. Text widgets would also benefit from having more shortcut keys, like ^U for "kill line", ^W for "erase word" et.c. In many of the applications of MacOS X, most of this functionality is non-existant.

Most of these do exist. Command arrow moves that direction to the largest extent possible (beginning/end of line, beginning/end of document). Holding shift at the same time selects the text. Option does the same movement, but by the word. Many developers choose not to use these shortcuts, but that is hardly the fault of the GUI. (indeed, Apple has gone out of there way to make Cocoa easy to use and design to their standards).

Scriptability: You mention AppleScript, and claims it is like having shellscript for GUI. No it isn't: you are bound to use that specific language

Wrong again. You do not need Applescript. You only need to use an OSA (Open Scripting Architecture) compatible script. Applescript just happens to be Apple's branded solution that they (duh) ship with the machine and support.

Stupid messages...

This whole complaint translates to "It doesn't do things the way that I'm used to *wah**wah*"

Many of your complaints seem justifed to me, i.e. themes and multiple desktops, but I think that on the others you should learn more about the OS (and the conventions/metaphors behind it) before you complain. Different doesn't mean worse

Re:I hope people does not totally trust this revie

[GauteL]

"Some people like to use their own programs to paint whatever they want on the root window".

If you're talking about stuff like "Xsnow" it works with the latest version of Xsnow.

Besides, I don't think cutesy hacks for 31337 people are that big of a deal.

Gaute

Criticism is Healthy

[DiscoBiscuit]

I admit, I havent seen Gnome 2.0 myself, so i'm not that qualified to comment, but i'm coming from a different angle.

Why be so quick to debunk criticism? We're all VERY quick to point at flaws in M$, and other evil empires' software.

As far as i'm concerned criticism is neccesary and healthy. It can be reviewed, considered, and if the result merits changes being made for the benefit of better software for us all, then I am all for it.

I for one am very keen to see Open Source software reach levels that surpass, in every aspect, commercial software. It's going to be a long journey, and if criticism is ignored, we'll never get there.

I don't believe the reviewer was 'having a go', just that they were genuinely dissapointed in a product they WANT to see succeed same as the rest of us.

Re:Gnome and KDE are more or less the same these d

[foobar104]

There have already been many replies to this, but here's my two cents anyway.

Every single part of a GUI should be accessible via the keyboard, so that experienced users can be as effective as possible, using these.

I don't accept this blanket statement. While there is a case to be made for full keyboard access to the UI for movement-impaired users (see "Full Keyboard Access" under the keyboard prefs pane), I think the jury is very much out on the subject of whether the keyboard is a good interface at all. There's no question that too much keyboarding is related to repetive stress injury. Being even more dependent on the keyboard than we already are could very well turn out to be a bad thing, not a good thing.

Multiple desktops: it's obviousely an advantage to be able to have multiple workspaces running at once.

You say "obviously an advantage" when what you mean is "I like it." Some people like having multiple workspaces or desktops. Personally, I don't. I prefer overlapping windows, so I can see what I'm doing without having to shuffle things around. So this issue boils down to personal taste.

I don't think "I like it better another way" is a very valid user interface design critique. And before you respond with "they should have given me the option," please remember that a good user interface is not one that gives the user every possible option. Simplicity is a virtue.

Configurable look (themes)

We'll argue about this forever. The bottom line is that lots of people spent a lot of time designing the Aqua user interface. They designed it to be easy to use and visually appealing. What possible motivation would Apple have for implementing an interface that lets little Jason from down the street make all of this windows black and purple and change the "File" menu to read "Zeppelin Rules!"

As a person of strong aesthetic opinions, I consider Apple's refusal to include an API for modifying the interface to be a good deed, worthy of praise.

All the rest of the comments in your post have been responded to elsewhere more or less as I would here, so I'll just skip to the end at this point.

Re:Gnome and KDE are more or less the same these d

[forevermore]

If you're just talking about colours, is it really that big a deal?

Yes, it is. I spend 10-12 hours each day sitting in front of a computer screen. Reading dark text set on a light background may be fine for paper, but when that light background is essentially a glowing white light bulb, your eyes REALLY have to work hard to keep that black text in focus. Personally, I enjoy not needing glasses, and so I use a theme (gtk) that uses a dark grey background with light-grey text (or when forced to use Windows at work, I at least tweak my editor to look similar). This has the effect of SERIOUSLY reducing my eye strain.

So yes, colors are quite important, and though Aqua may look nice for the average user, it's not such a good them for programmers (and I would do my best to change it if my mac could actually run OSX - as it stands, I'm running a light-on-dark theme in OS 9).

-xris

Re:Gnome and KDE are more or less the same these d

[Dixie_Flatline]

Well, I do the same. However, my window dressing is so entirely unimportant, it doesn't even phase me. My emacs has a dark background and a lighter foreground. That's what I'm staring at, so that's where the high contrast colours come into play. There's just not enough interface on the screen to stare at all day long for a programmer to care. The embedded editor in Project Manager can be changed to have whatever colours you want.

If you're telling me that not being able to change the titlebar and panel colours in OSX is causing you eyestrain, then I'd have to say that you're spending too much time staring at the 'compile' button.

Re:Gnome and KDE are more or less the same these d

[Phil Wilkins]

> You mention AppleScript, and claims it is like having shellscript for GUI. No it isn't: you are bound to use that specific language

It's actually much more powerfull than shellscript, as it can manipulate objects other than text files. The architecture is completely open and documented. There are numerous other languages available, only the basic object model is fixed. The whole fully factored and recordable cross application scripting thing blows shellscript out of the water.

Re:Gnome and KDE are more or less the same these d

[RevAaron]

Heh, OK. I thought that was the case, but thought I'd make sure. ;)

Re:Gnome and KDE are more or less the same these d

[Pierre]

Did you read the post?

The part about it depends on who your talking to.

Multiple desktops are not obvious to a newbie. The desktop is not obvious to a newbie.

If somebody is unfamiliar with a computer - then I would say that multiple desktops are not obvious to them.

I am amazed that you are amazed at that.

Re:Gnome and KDE are more or less the same these d

[Dixie_Flatline]

The common mistake that you're making here is figuring that it isn't possible to make an interface that requires little to no customization, and yet suits the needs of everyone. What needs do you have over an above the average user? Why isn't it possible to give all users that interface without taking something away? What does the power user require that is MUTUALLY EXCLUSIVE from what the average user needs?

A big part of the problem with UIs is that we settle for too little. More design and more work on the part of the UI engineer could make a lot of problems go away.

For a better idea of what the devil I'm talking about, pick up Jef Raskin's book, "The Humane Interface".

Apple Events and AppleScript

[yerricde]

You mention AppleScript, and claims it is like having shellscript for GUI. No it isn't: you are bound to use that specific language.

Sorry, you're wrong. There was a product called "Frontier" that implemented AppleScript's functionality with syntax reminiscent of the C language.

They could easily have supplied a network protocol (like KDE's DCOP) or any other more generic interface.

They did. It's called Apple Events, and it's been in Mac OS since 7.0. The Open Scripting Architecture (which describes an app's object model to a scripting system) has been around since at least 7.5.

Re:Is it possible to Cut'N'Paste yet ?

[spitzak]

The seperate clipboards are necessary to get the behavior that Windows users expect. And it has one minor technical advantage: you can select the text to replace after you have selected the replacement.

The seperate clipboards are used by GTK and Motif, it was older Qt that did not do this (also true of a lot of other programs and toolkits, including fltk which I fixed a few months ago to match GTK). Newer Qt has been changed to match GTK.

Re:Is it possible to Cut'N'Paste yet ?

[spitzak]

Although there are problems with X, you can easily put the URL into Netscape from the terminal. Select it in the terminal, go to Netscape, and click the middle mouse button on the text field.

Re:"lock the toolbars"

[spitzak]

I would like to see some of these idiot GUI experts figure out that "menubar" and "toolbar" are the same thing (especially now that the toolbar buttons often bring up a menu) and get them on the SAME LINE!!! That would make me think that these designers (both at MicroSoft and at all other companies and in open source) are not just a bunch of sheep who refuse to make the slightest tiny changes because of a morbid fear that some poor stupid user will be "confused" by it.

Re:I hope people does not totally trust this revie

[spitzak]

That's right, there was new standards added so a program that wants to draw a background can communicate with the program that is drawing other things like icons on the window.

Older solutions required X to keep seperate the background area and the icons, which had to be done by making each icon a seperate shaped X window. This was tremendously slow and required the program drawing the icons to send a lot of information to the X server. And there was no antialiasing of the edges and serious limitations on what you could draw.

A program that just thinks it has a big window and can draw anything into it is *much* faster, and will always be faster no matter how many new features are added to X. So instead these programs can now read a desktop image stored on the server and merge their own display with it to make the window.

Older desktop image programs that do not know how to communicate this information to these new desktop programs will not work.

Re:Gnome and KDE are more or less the same these d

[Angst Badger]

You think that programmers only like hard-to-use, unapproachable, syntactically impenetrable languages? I would argue that Smalltalk is easy-to-use, approachable and occasionally 'English-like', and I don't have any problems with it. I've never used Applescript, but as an experienced programmer, I don't think you should be making generalizations like that.

I think the point that was being made here is that English and other human languages are great for communicating with humans, but they are just not structurally appropriate for writing programs. A programming language should be concise and clear, and must be absolutely unambiguous. Human languages are not, by nature, any of these things. The requirements that the two kinds of languages are designed to meet are completely different.

The problem with English-like programming languages such as AppleScript and COBOL -- both of which I have used -- is that they are designed to solve the wrong problem. It is assumed that end users cannot understand programming languages because they are syntactically obscure. This is, of course, crap. You can teach any person of average intelligence the syntax of C in a couple of days, possibly excluding the irrational mix of pre- and postfix pointer notation. Given a couple of months, that person could have a pretty good grasp of the ISO/ANSI specification for the language. But even if Joe Average could go on ANSI Jeopardy and answer, "Alex, what is dereferencing a NULL pointer?" he still could not write a complex application.

And that's really it -- knowing a programming language no more makes you a programmer than knowing English makes you a novelist. Having an English-like programming language like COBOL will not make it possible for non-technical management to look at source code and understand what's going on. All English-like syntax accomplishes is to very slightly shorten the initial learning curve for the syntax and the syntax alone, while it becomes a real pain in the ass later on once you understand algorithms and data structures and all of the meta-linguistic knowledge that is the real meat of writing software.

Re:I hope people does not totally trust this revie

[raistlinne]

People desperate to get rid of Nautilus, could do it via gnome-session-properties, and actually, as of GNOME 2.0 I don't see the point apart from feeling 31337.

Maybe because we absolutely hate icons sitting on the "desktop"? Btw, it's not a "desktop", it's THE BACKGROUND. It exists because there has to be something behind everything, not for any other reason. I don't want cute little icons or other stupid stuff taking up memory and doing no good besides annoying me.

I don't know where all these people who think that they're God's gift to the rest of us have come from to attach themselves to Gnome, but GET OVER YOURSELVES. Not everyone is the same. Hey, speaking of which, I just checked that link that you have. That site isn't in English. Why are you using some other language, to be 31337? Everyone speaks english, stop using all those other languages, they just cause confusion. Really, english is the most widely spoken language, by far. Everyone else who wants to cling to some other language is probably just a backwards, arrogant, snivelling pimple-faced socially inept pre-pubescent teenager who is of no consequence in the world and God hates them.

I hope tha tyou get my point.

~/.* is more than just four punctuation marks

[yerricde]

I really and truly haven't seen any OS that gives each user a unique configuration.

Both GNOME and KDE, desktop environments for UNIX compatible systems, store the users' desktop preferences in ~/.* (that is, hidden files in the users' home folder). Windows 2000 and XP do something similar: it stores settings in C:/Documents and Settings/$USER/.

Re:Changing Emacs to XEmacs...

[Leimy]

My bad then... I should have been more explicit/specific about what the situation was... as you can see the entry was already long though.... I over edited myself.

OT: NASCAR is a sport

[Rebel Patriot]

Don't believe me? Come get in a race car with me sometime and find out. True, I don't run NASCAR, but auto racing is allot more strenuous than most computer programmers think.

And while we're at it, stop the stereotypical comments. Most Southerners enjoy NASCAR; that does not make them stupid. The fact that I enjoy NASCAR doesn't mean that virtual desktops are confusing to me no more than raising hogs and chewing tobacco makes it more difficult on me to grep a source tree.

don't get me wrong, I still *really* like Gnome

[Ender Ryan]

I think my original message was a *bit* too harsh. Gnome 2.0 really isn't release material, but it's definately got *lots* of potential.

BTW, the panel bugs are already posted, hopefully I won't have to logout till the next panel release ;-)

I'm really looking forward to Metacity. It looks to be shaping up really quickly. I think screwing with sawfish for the gnome 2.0 release was a *BIG* mistake, they should have either held off on the release till metacity was ready, or left sawfish alone and released with that.

I expect Gnome 2.x to rock at around 2.4, just like 1.x did.

I should have mentioned in my other post, Nautilus has *really* shaped up nicely. I can't stress that enough, it's quick, extremely pretty, userfriendly... If you use Nautilus a lot, you'd probably do well to upgrade to gnome 2.x now or real soon, despite the quirks.

For anyone using gnome 2.0 now, you might want to stick with sawfish 1.x for the time being, 2.x is way to crashy.

Re:I hope people does not totally trust this revie

[raistlinne]

Have you ever heard of sarcasm? :-)

My entire point is that language is a perfect example of where the whole push for "one choice" falls so short, and is so obviously wrong.

And on the matter of programming languages, didn't you see the essay where Havoc Pennington criticized xchat for being scriptable in not just one, but four different programming languages?

Licenses that enforce usability

[Ilan Volow]

A lot of people in the linux community say:

"GUI design is a BS field of study, people who point out usability flaws that make UI's confusing are whining, the fine manual is perfectly understandable, the problem is people don't want to learn, and saying that fill-in-the-blank free software interface isn't ready for joe end-user is merely spreading M$ FUD."

And that's okay. They are welcome to feel this way.

As I am more than welcome to come up with new types of public licenses that enforce usability. For right now, take this as nothing more than a silly rant written by some random poster on slashdot. Take it with a grain of salt and a Big Gulp-sized jar of Malox. But, if my licenses ever do make slashdot and people starting yelling and bitching about "Anti User-Hostility Public Licenses", then those people need to understand that their attitudes created the necessity for such licenses.

Re:Gnome and KDE are more or less the same these d

[baka_boy]

Your comment echoes a number of standard arguments against the Aqua, and while I don't agree with all of them, there's no point in covering that territory again.

However, I think you missed a number of positive elements to the full OS X user experience that are missing from most other desktop environments, which might help to balance the evaluation somewhat.

First, we have the Dock, borrowed (though not without modification) from NeXTStep. Some people love it, some people tend to just ignore it, but personally, I find one feature it has to be *extremely* useful: dynamic icons for applications. The most commonly-seen example is the Mail app, which overlays a small red circle with the number of unread messages any time new mail arrives.

Second is my favorite "window dressing" feature of Aqua: the drop shadow applied to each window. It's a subtle thing, which I didn't realize the value of until I installed a hack that removed it -- suddenly, I lost what I had gradually come to rely on as a stable visual clue as to which window had focus.

Third is the standard design for toolbars in (Cocoa apps, anyway) that allows drag-and-drop addition or removal of commands, and a browser-like selection of displaying just the icons, just the text labels, or both. That means that since I like preserving screen real-estate, but still use a toolbar in some apps, I can switch things to only a text label, while my girlfriend, who uses OS X but isn't a hardcore techie, can leave all the icons in.

It's exactly these kinds of details that take real usability testing, good design, and *time* to do well. When you come right down to it, the biggest advantage that Windows and the MacOS have over open source desktop environments is years of little tweaks and polishing.

Personally, I think that the KDE crew is on the right path: start with a simple desktop environment similar to what people already expect, and just pound away on the little stuff until the whole interface looks good, works consistently, and offers users in widely different experience levels a worthwhile experience.

*Metatheme is not a part of GNOME 2.0*

[nullity]

Most of his UI complaints seem to center around metatheme, the "Desktop theme editor" etc. Guess what? Metatheme was dropped from the GNOME 2.0 release about 4 months ago because it was deemed sufficiently unusable. While we plan to eventually have a single "themes" desktop preference page (that will, of course, replace the widget theme, etc, we're not going to be duplicating menu entries), we decided to for GNOME 2.0 because Metatheme's interface sucked so badly. It needs to be totally redone. I agree 100%.

It seems like 3/4 of his rant against GNOME usability is based on things installed by Metatheme. It is absurd to complain about GNOME's interface based on something that was dropped from GNOME 2.0 because we knew it sucked.

-Seth (GNOME Usability Project Lead)

Re:OT: NASCAR is a sport

[Dixie_Flatline]

F1 is a sport. Rally car racing is a sport. NASCAR is an exercise in turning left. I concede a small amount for NASCAR races held on non-oval tracks.

I'm a cyclist, myself. I watched Lance Armstrong climb through the alps last year. That's a sport.

Re:Is it possible to Cut'N'Paste yet ?

[Yorrike]

Too right.

Posting bug reports and features requests is one of the fantastic things about open source. If you don't like the way something works, or something doesn't work at all, tell the developers. If they're reasonable people, they'll listen and decide whether you have a valid suggestion or not.

People need to stop whining and complaining and damn well do something about it. You're not dealing with Microsoft here, your opinion matters to developers.

Re:Gnome and KDE are more or less the same these d

[kcbrown]

Uniformity of interface is a BIG DEAL.

It is. But the uniformity that matters is on the individual's desktop, not between systems as many seem to think.

In other words, what matters is internal consistency. Once someone starts using a machine, they should be able to customize it to their own tastes. The machine needs to be consistent in how it interacts with the user, even after the user customizes it.

Now, that said, it's certainly desirable for the default interface to be something that people can get used to easily, but this does not argue against user customization. Remember that computers are there to make our lives easier, so it is the computer that should be made to adapt to us, not vice versa. That, more than any other reason, is why a user interface should be customizable.

There are valid reasons (other than aesthetics) for people to be able to customize their desktops. Different people see differently, and some color schemes are much easier on some people's eyes than others. But even if there is significant agreement within a population on the benefits of one particular color scheme, there isn't universal agreement. That's why it's important for people to be able to customize their environment.

Re:Gnome and KDE are more or less the same these d

[mgv]

Multiple desktops are such an obvious advantage that I can't believe they aren't as prevalent as overlapping windows.

To my mind you might as well have your windows maximised all the time and click on the panel/taskbar. Its basically the same as having workspaces. What is the point in having multiple workspaces unless you use windows at less than full screen? I've got the powertoys for XP mod (4 workspaces) and I found I stopped using it - extra workspaces just waste my time switching. Any opera/Netscape 7 fan would understand - You just need a panel to click with one button per window.

Michael

Re: KDE 3.0 slow...

[Arandir]

I'm guessing this is because of build time optimizations?

Most likely. I use FreeBSD, and build everything from scratch optimized for my machine. At work, I am dual booting Win2K and FreeBSD/KDE, and there's no comparison. Win2K feels like someone dumped molasses in it. KDE doesn't feel slow at all.

Re:I hope people does not totally trust this revie

[Eil]

Secondly. GNOME has taken a very far step towards KISS (Keep it simple stupid) unlike some comments on here seem to suggest.

Yes, so simple they took out most of the features that made GNOME my favourite desktop!

The reviewer tries to make himself out as a GUI-expert, something he doesn't seem to be at all.

When I read the review, I had the immediate impressed that she was just a regular user. IIRC, she claimed right up front to not be too impressed with *NIX desktops in general. Much as I like Linux and Unix, I'd say I side with her.

There are ACTUAL GUI-experts and usability exports working on GNOME.

Where are they, then? Last I heard, Sun did some usability studies on GNOME and that was about it. Whether the results actually did anything for the resulting interface of GNOME 2 is debatable.

Turning off Nautilus for speed should be rather unnecessary except for people really short on memory.

...or have a sub-gigahertz processor.

People desperate to get rid of Nautilus, could do it via gnome-session-properties, and actually, as of GNOME 2.0 I don't see the point apart from feeling 31337.

Wanting to get rid of software I consider unnecessary is considered "31337"? (By which I take to mean "lame".) Please explain that logic to me. In that case, I'd have to call you "31337" as well if you've ever removed panels that display by default or disabled that damned annoying fish applet.

Let's not forget that one of the things that GNOME claims to be is modular which means you can remove, swap out, or write any component you wish and the rest of the components will not bitch. THAT INCLUDES NAUTILUS.

It's really fucking sad what's happened to GNOME. It really truly is. The GNOME developers and even the Slashdot audience, which up until now I've considered halfway clueful about this sort of thing, have completely forgot what made GNOME good in the first place: the fact that you could tell it how you wanted it to behave and it wouldn't argue. Now, with GNOME 2.0, we have a desktop with a couple of "sensible defaults" and a shitload of users who believe customizing your own environment is a fucking crime.

Re:Gnome and KDE are more or less the same these d

[el_chicano]

Go work on your own interface. Leave ours alone.

This is pretty funny when you consider this thread began with a Mac user implying that Gnome/KDE were not innovative and telling us to take a look at the "visual inventiveness of Mac OS X", a "GUI that's worthy of the 21st Century"!

Re:Gnome and KDE are more or less the same these d

[el_chicano]

As a person of strong aesthetic opinions, I consider Apple's refusal to include an API for modifying the interface to be a good deed, worthy of praise.

What about Apple's insistence on a one-button mouse? If humans had only two fingers and a thumb, then a one-button mouse would be best we could hope for.

I find it funny that your average cartoon character has a thumb and three fingers and could probably operate a two-button mouse while apparently your average Mac user can't!

So much for "ergonomics" and "human design"...

Re:Gnome and KDE are more or less the same these d

[el_chicano]

However, poorly copied OS X features could be quite disasterous.

Please, if you are going to copy OS X, don't copy the spellchecker!!!

Misconceptions about OpenOffice/StarOffice

[Moritz Moeller - Her]

Let's face it nothing does DOC like Word [thank god!] and for compatibility purposes with all of my coworkers there just isn't a real substitute for everything it does. We use the revision control built into Word and other things so please don't offer Abiword, StarOffice, OpenOffice or KWord as alternatives.

Well staroffice does word processing better than MS Office. Also it has revision control compatible to MS Office in the doc format. You have not really used staroffice obviously or you would not talk about it like you do.

Finally I would propose crossover office, if you really must use MS Office.

Re:Is it possible to Cut'N'Paste yet ?

[spitzak]

Actually the selection buffer and the clipboard I am talking about are the same thing. There can be an infinite number of selection buffers with different names, but the only ones in any common use are "SELECTION", "CLIPBOARD", and "SECONDARY_SELECTION". "SELECTION" was supposed to be the highlighted text and "CLIPBOARD" was intended to be used for cut/copy/paste commands. "SECONDARY" was for some weird and obsolete idea involving the ability to swap two pieces of data with the mouse and nobody supports it.

Perhaps unfortunately the middle-mouse paste of SELECTION turned out to be so incredibly convienent that everybody forgot about the CLIPBOARD, and most people implementing copy/paste commands put it into the SELECTION because a vast number of X programs could only paste with that middle mouse.

Only with the need to make Windows users happy has the original design of the CLIPBOARD reappeared.

The selection buffers have "type" information so it is in theory possible to cut any kind of data, just like Windows. What X lacks is any standard for much other than text, or a library that can draw the object on the screen or store it in a file without much thought from the programmer. This is the primary reason that anything other than text does not cut/paste. This is also true on Windows as soon as you get to more complex programs that want to analyze the pasted data and not just draw it on the screen, suddenly a lot of things don't paste (go ahead and try to paste that image into your IDE...)

Re:Gnome and KDE are more or less the same these d

[zangdesign]

Yeah, I probably should have left that out. Personally, I prefer the Mac interface over KDE, which I prefer over Gnome. Both of the latter look and feel enough like Windows that it bugs me.

I just wish someone could come up with something really different to kind of kickstart ideas.

Enforced consistency would also be nice.

Re:Gnome and KDE are more or less the same these d

[foobar104]

What about Apple's insistence on a one-button mouse?

I have two points.

First, a significant fraction of the population at large has problems with multiple-button mice. Young children, older people, and handicapped people all have trouble with manual dexterity. Even otherwise average people of middle age can have problems with multiple-button mice if they suffer from arthritis or other degenerative joint or muscle maladies.

So first, we have about 50% of the total population that would object to the required use of a multiple-button mouse.

Second, Apple's user interface supports but does not require a second (or third, or whatever) mouse button. While you can use multiple-button mice on a Mac-- and I do, on one of mine, with no special drivers or software-- it's not necessary in the slightest. The second (or third, or whatever) button is a user-configurable shortcut. For users that prefer them, multiple-button mice are available from third party vendors at very low prices.

The conclusion: Apple is right to design their computers around a one-button mouse, while building support for multiple-button mice into their software.

This is a basic principle that the designers of software like Gnome have thus far failed to implement in any comprehensive way: design first for everyone, and only then provide unobtrusive features for those who want or need them.