Slashdot Mirror

Not really true. If you bring up a Gnome file dialog and just start typing a file name Gnome will open a text box and allow you to enter the file name with tab completion.
It's a very slick example of what Gnome needs to do more of. Gnome has focused its efforts on simplifing the interface for the masses. They've made good progress but the masses seem unimpressed.

It's time to think about finding elegant ways to put that power back in while keeping it transparent to the masses.

Wrong. It's completely hidden functionality that's actually less functional that the previous incarnation. There's no way to know that Shift-DoubleBuckey-CokeBottle pops up the "Go" window or an indication of what the subwindow does. Most importantly, typing a path doesn't actually open the file, but rather simply changes you to the directory, so you still have to hunt and click.

It goes against 30 years of GUI development, and it's broken by design.

Havoc Pennington can suck my cock, but I imagine he'd find that too complicated too.

http://bugzilla.gnome.org/show_bug.cgi?id=136541

(And yes, I am a UI expert.)

The problem with that "Linus is a user" argument is that Gnome is NOT designed for that kind of users, and thus Linus is not a proper judge for the design decisions taken in Gnome.

KDE is the desktop designed for tweak-it-yourself power users. Thats why KDE is much better for Linus; but it DOESNT mean that Gnome is bad for Gnomes intended user base, wich is the i-dont-care-how-it-works-just-dont-waste-my-time kind. This response from Havoc Pennington explains it fairly well.

now that somebody has mentioned gimp, i will write my little rant here.
gimp is the gtk app i use the most. i immediately noticed changed open/save window. it seemed pretty nice overall, but lack of address bar was driving me nuts.
then somebody mentioned that typing "/" would allow to enter path directly. this was pretty nice, but there are two things that make this dialogue so irritating i prefer clicking instead of writing.
first, if i start typing with ~, this doesn't work.
second, if autocomplete kicks in, it works _completely different from any other app_ and BLOODY AWKWARDS.

i have screamed at my monitor how much i hate it.
let's say, i have a directory "/mnt/net" i want to get to by typing it. what i get is "/mnt/net/t/net". wtf ?

turns out, if autocomplete kicks in and it has only one suggestion, my further typing _is not_ replacing the suggestion, it is appended to it. if this is not a bug, somebody has seriously screwed up.

basically, if i type a path in, i type it pretty fast. current implementation basically forces me to pause after each bloody character to see wether i will be able to continue my writing or something has been autocompleted.

this implementation has so many problems i am surprised it was pushed in this state, especially given all these usability zealots :)

see http://mail.gnome.org/archives/usability/2005-Dece mber/msg00028.html for some examples (including starting with ~)

read the reply with all the issues that are wrong with the file dialog, and why its still is broken:

http://mail.gnome.org/archives/usability/2005-Dece mber/msg00028.html

GNOME is made by people who try to be like Apple,

You must be using a different Gnome than the one that is at gnome.org. None of the screenshots there look even remotely like any MacOS nor OSX. What it does have remarkable similarity to, however, is windos.

Well, for what it's worth, the usability study conducted by Sun seemed to be fairly thorough to me. Then again, I'm not a design guru of any sort, so what do I know?

Frederic Crozat did cite this reason, yet he did not make the decision. Contrast with the answer from Murray Cumming

Whether or not what Linus has to say is relevant, here is least is the other side of the coin. Read here. Make up your own minds about what is useful and helps you stay productive. But for the record, I prefer GNOME and my parents prefer KDE. Figure that one out.

Two is that within those who contributed the the UI discussion there was a surprising lack not only of experience in the HCI field (ok, I had just started out there myself) but also a strong resistance to pick up the vast literature available or trust in actual end-user studies.

List of references considered for the Gnome Human Interface Guidelines.
Usability Study from Sun Microsystems.
Usability testing videos from Novell

Somehow, the current situation looks quite different from what you describe.

Two is that within those who contributed the the UI discussion there was a surprising lack not only of experience in the HCI field (ok, I had just started out there myself) but also a strong resistance to pick up the vast literature available or trust in actual end-user studies.

List of references considered for the Gnome Human Interface Guidelines.
Usability Study from Sun Microsystems.
Usability testing videos from Novell

Somehow, the current situation looks quite different from what you describe.

Yes, some GNOME developers are self-appointed control freak antifeature
nazis who've stripped functionality in pursuit of some theoretical "non
geek" user who does not exist, thereby crippling their software.

And probably some KDE developers are feature sluts who never saw a
checkbox they didn't love, exposing users to all kinds of broken
features.

On Mon, 2005-12-12 at 17:46 -0800, Linus Torvalds wrote:
> I personally just encourage people to switch to KDE.

Everyone on this list knows the Linux desktop is in a "pick your poison" state right now.

Anyone who's used Linux for a year has experienced this, whatever choices they've made of desktop environment, settings, etc.

We can snipe at each other all day long. (Linus, every time I copy large files between devices on my Linux system my mouse pointer skips. It works fine on my Mac). That's not productive.

Usability is important. Usability encompasses multiple things: functionality, robustness, performance, sensible user interface design. We all need to do a better job of this (insert usability testing/betterdesktop.org plug here).

Yes, some GNOME developers are self-appointed control freak antifeature nazis who've stripped functionality in pursuit of some theoretical "non geek" user who does not exist, thereby crippling their software.

And probably some KDE developers are feature sluts who never saw a checkbox they didn't love, exposing users to all kinds of broken features.

Follow either of these ideas to their logical extremes and we won't have a useful desktop for a large user base.

We need Linux to grow up if we're going to make Linux on the desktop a success. Let's have a grown-up discussion. If I worked for Microsoft I'd be very happy to see you throwing pejoratives around like that on this list.

So, yes, usability is important and Linus being able to bind his mouse buttons to whatever he wants is important, I guess. But it's probably not what's stopping Linux from dominating the desktop market. What's holding Linux back on the desktop? Applications, device support. Time, also. The printing dialog? I don't know.

(By the way, on my GNOME machine at home, there is code running that parses the options from the PPD file and makes a GUI out of them. Maybe this ships in SUSE but not in whatever distro Till is using?)

Nat

http://mail.gnome.org/archives/usability/2005-Dece mber/msg00027.html Just a sample: We're not aiming for "powerfully extensible". We're aiming for "Just Works". Some people will hate that. Some will love it. Personally, I'd rather have passionate users, lovers and haters, than be than average and ignored, and I think you'll find most GNOME developers feel the same way. Personally, I think Linus ought to know better by now than put out a self-centered post like that. There are more users in the world than just geeks. Most aren't geeks, in fact. For Linux on the Desktop to survive beyond the lifespan of its proponents, it needs to acknowledge that, not fall trap of intestinal power struggles.

Gnome developer Havoc Pennington's response points out that "reducing complexity" was not, in fact, the reason the particular dialog in question doesn't have all the options Linux wanted:

"Just for the record, since I made this decision I can tell you that 'might confuse people' was not the reason. More evidence for my point that 'might confuse people' is the reason made up by others, not the reason given by the decision makers."

Which is not to say that Linus is wrong (in the e-mail he writes that "If this was a one-off, I'd buy it. But I've heard it too damn many times. And only ever from Gnome.") -- I'm not a big fan of Gnome's lack of features (at least as compared to KDE), but it's not like anyone on Slashdot really conforms to the "average computer user" concept. And Linus surely doesn't either. Maybe Gnome is better for Mom and Grandpa. I'll stick with KDE, myself.

The lack of strong Bidi editor (yes, I do use gedit), hurts advancing the use of free Hebrew-supported software. For Emacs addict users (like me), it is difficult to document in Hebrew usage examples, or write simple notes in order to encourage the use of general free software by Hebrew speaking people.

The screenshot is an exact copy of Excel 97 (and I think that was the stated intention).

Concerning the pivot tables: it's in the make. Concerning macros: you are wrong, Gnumeric IS scriptable (Programming Gnumeric using Python)!

The screenshot is an exact copy of Excel 97 (and I think that was the stated intention).

Concerning the pivot tables: it's in the make. Concerning macros: you are wrong, Gnumeric IS scriptable (Programming Gnumeric using Python)!

If the MS Office model is so flawed, then why does every open source office application look exactly like its Microsoft counterpart?

No, they don't. Have, for instance, a look at Gnumeric. It offers features currently not present in Excel.

OOo is mainly an MS Office clone. Innovation happens in independent projects like Gnumeric, AbiWord, etc.

• OpenOffice.org Hackers - note that the domain is ooo.ximian.com.
• Hacker's guide says CVS checkout is based at :pserver:anonymous@anoncvs.gnome.org:/cvs/gnome.
• Planet OOO features a few faces I remember from Planet GNOME, in particular, Michael Meeks.

• OpenOffice.org Hackers - note that the domain is ooo.ximian.com.
• Hacker's guide says CVS checkout is based at :pserver:anonymous@anoncvs.gnome.org:/cvs/gnome.
• Planet OOO features a few faces I remember from Planet GNOME, in particular, Michael Meeks.

There are at least two other open source suites out there. Its too bad that they only run well on Linux, but they are being ported to Windows and OSX.

If you are going to make comments on Open Source ability to make office software, you would need to comment on those. Especially since they are both open source right from the beginning, unlike OO which has commercial origins.

In my experience all the components of Koffice work really well. Gnumeric has many advanced features and continues to be intensively developed. AbiWord still needs some polishing IMO.

Most of Gnome's clipboard issues were fixed in the 2.12 release back in September (as included in Ubuntu Breezy). From the release notes :

GNOME now remembers data that you copy, even when you close the window from which it was copied.

See also: A Prerelease Tour of GNOME 2.12 .

Most of Gnome's clipboard issues were fixed in the 2.12 release back in September (as included in Ubuntu Breezy). From the release notes :

GNOME now remembers data that you copy, even when you close the window from which it was copied.

See also: A Prerelease Tour of GNOME 2.12 .

Please please please please use the GNOME Logo from this century on Slashdot:
http://live.gnome.org/LogoGuidelines

Has Gnome has some kind of epiphany or something?

Yes it does, but it's just a web browser. ;)

I've never heard of GLUI and never run into a program that used it; a more established library is GtkGLExt, and the most common GTK GL interface is gtkglarea, which has become a part of Gnome.

Both desktop environments appear to have very good internationalization.

For Gnome: http://www.gnome.org/start/2.12/notes/en/rni18.htm l and http://www.gnome.org/i18n/

For KDE: http://i18n.kde.org/stats/gui/stable/toplist.php

So, currently Gnome supports* 43 languages, and KDE supports 23 languages.**

It is not at all obvious to me how KDE's internationalization is so superior. If you could explain your rather blanket statement, I would appreciate it. Otherwise, it seems to me that both desktops have excellent internationalization. Kudos to both KDE and Gnome.

* "supports" defined as at least 80% of strings translated.
** Note: I'm sure KDE will support more languages as their 3.5.1 release comes out: the x.y.1 usually has a lot of attention devoted to translations.

Both desktop environments appear to have very good internationalization.

For Gnome: http://www.gnome.org/start/2.12/notes/en/rni18.htm l and http://www.gnome.org/i18n/

For KDE: http://i18n.kde.org/stats/gui/stable/toplist.php

So, currently Gnome supports* 43 languages, and KDE supports 23 languages.**

It is not at all obvious to me how KDE's internationalization is so superior. If you could explain your rather blanket statement, I would appreciate it. Otherwise, it seems to me that both desktops have excellent internationalization. Kudos to both KDE and Gnome.

* "supports" defined as at least 80% of strings translated.
** Note: I'm sure KDE will support more languages as their 3.5.1 release comes out: the x.y.1 usually has a lot of attention devoted to translations.

ALT+ should stick to activating buttons and menus. In other languages, there are menus that should have the accelerator D, and there are more of these examples of these collisions. I'm pretty sure the designers of IE has had plenty of reason to smack themselves in the head for not thinking this through beforehand.

Epiphany for instance refuses to follow the conventions on this one, and while it is a bit annoying the first period of time, I do think they are making the right descision. CTRL+L still works, of course, and in the same smooth way as Firefox. Being consistent and catering to all users of all nationalities are both worthwhile goals IMO. Copying an old mistake from a mostly obsolete dinosaur is not.

I'd start with the following:

• The man pages and info documentation included with any Linux system.
• The Gnome and KDE developer sites. If you're developing commercial software, develop for Gnome and your app will work fine on KDE.
• For specific applications like graphics and audio, you may need to use other libraries (i.e. libsdl and wxWidgets)
• If you have any questions, there is 24/7 live human technical support for application developers on IRC. Don't undervalue this - being able to ask a real human and get an immediate answer is huge. (try asking in irc.freenode.net #linux or #debian for what channel to ask specific questions if you can't figure it out)

most of what the writer of parent wrote (at least in the first part before he started rambling a little) is absolutely correct.

No, it's bullshit. You can't say something is correct when it's just an opinion.

It is more of a patchwork than a coherent desktop.

Opinion.

Most of the work since GNOME 2 has been trying to smooth over the rough edges because so many of the projects that make up GNOME make Windows 95 DLL hell look like fun.

How so? You mean because you have to install 20 libs? If you take a look at KDE, the same is true - except they simply bundle those 20 libs into one gigantic KDE-libs package. The same thing could easily be done for GNOME, simply make a new directory with a new configure.in and a new Makefile.am with all the other lib src trees as subdirs. How does this make anything better? What if there's a security hole in one of the smaller libs included in the massive bundle? As a user, that means I have to upgrade MassiveLibBundle instead of updating LittleLib.

Maybe you define this as "dependency hell" because you compile everything by hand. Why the hell would you do that? There exist good package managers out there TODAY that resolve dependencies for you. All of the GNOME libs also follow strict API/ABI compatability, so there's no real "hell" here either.

The simple fact is that there really are huge gaping holes in providing anything near a useful set of reusable objects

O RLY? Please, good sir, pray tell. Oh wait, you don't know what you're talking about. Nevermind. Please continue your il-informed rant about something of which you know nothing about...

The biggest strength that KDE has is Qt, and GTK and the million or so other libraries that are hacked together with it to make up the base of GNOME, are not even close to equivalent as a logical, consistent framework to develop real applications.

More hot air? This just makes me laugh... as a GNOME developer there are very few things that feel this way. I can only assume you're rattling this off because some random clueless Joe spouted this and you read it on Slashdot and so It Must Be True(tm).

All the effort is put into making the desktop look polished without fixing the underlying issues that make GNOME development a freaking nightmare, especially compared to any other environment, like KDE, Cocoa, Windows, or even Java, that provides the developer with things like reusable toolbar widgets (to use the author of parent's example).

Wow. So much cluelessness in a single sentence. Where to begin? As someone working on this very aspect of GNOME, I can tell you that this is laughably fictitious. Oh, and btw - there is a reusable toolbar widget. It's called GtkToolbar, but I suppose it's too much to ask for someone to actually have a clue before spouting off like you did.

Apologies for coming off as a flame, but you asked for it with your ranting.

I'd rather let Nomachine NX deal with the optimizations of X protocol transfers.

A day after re-reading this essay on unmaintainable code, I was hit by a nice example of what is described as "Code That Masquerades As Comments and Vice Versa" in the chapter "Camouflage". Here is the code (sorry for the missing indentation - blame Slashdot):

<%
String submit = request.getParameter("submit");
// if (submit == null) {
// /* if the user did not confirm, go to the exit page */
// %><jsp:forward page="./SomeExitPage.jsp" /><%
// }
do.something(useful);
%>

Guess what? The scope of the "//" comments is probably not what you expect. I have written a blog entry about this funny case.

Has MS sued anyone over Mono patents? No.

However look here and here.

You don't need to sue someone so stifle progress as evidenced by the fact their Mono patents are currently stifling progress by the risk of lawsuits where Microsoft could easily remove that threat.

What type of diagram are you doing in OO.o?
For "Visio"-like stuff, I've been using Dia. Even on an old Win98 box it still presents well when importing into other presentation methods (Office XP docs, ppt / OO.o writer, impress).

Regarding your comment that accessibility should be built into the toolkit by default, you're described the GNOME accessibility architecture:

See, e.g., http://developer.gnome.org/doc/API/2.0/atk/index.h tml

See also http://developer.gnome.org/projects/gap/tech-docs/ at-spi-docs/book1.html

Here's the home page for the GNOME accessibility project (GAP): http://developer.gnome.org/projects/gap/

It's strange how the underlying article mentions GNOME accessibility over and over again as being even superior to Microsoft ones to date, and yet all the comments completely ignore GNOME's framework.

Regarding your comment that accessibility should be built into the toolkit by default, you're described the GNOME accessibility architecture:

See, e.g., http://developer.gnome.org/doc/API/2.0/atk/index.h tml

See also http://developer.gnome.org/projects/gap/tech-docs/ at-spi-docs/book1.html

Here's the home page for the GNOME accessibility project (GAP): http://developer.gnome.org/projects/gap/

It's strange how the underlying article mentions GNOME accessibility over and over again as being even superior to Microsoft ones to date, and yet all the comments completely ignore GNOME's framework.

Regarding your comment that accessibility should be built into the toolkit by default, you're described the GNOME accessibility architecture:

See, e.g., http://developer.gnome.org/doc/API/2.0/atk/index.h tml

See also http://developer.gnome.org/projects/gap/tech-docs/ at-spi-docs/book1.html

Here's the home page for the GNOME accessibility project (GAP): http://developer.gnome.org/projects/gap/

It's strange how the underlying article mentions GNOME accessibility over and over again as being even superior to Microsoft ones to date, and yet all the comments completely ignore GNOME's framework.

Now, if this goober had coded up a new manager which integrated all the functions he talked about, or had an extensble base manager to replace the native file system, with a defined api for plugins that would allow you to customize the environment, that would be news.

Now, if this goober had coded up a new manager which integrated all the functions he talked about, or had an extensble base manager to replace the native file system, with a defined api for plugins that would allow you to customize the environment, that would be news.

http://blogs.gnome.org/view/tko/2005/06/01/2
http://www.advogato.org/person/pycage/diary.html?s tart=31

Disappointing to see Dapper will still include those awful, tired (6 yrs?) old icons.

Who really thinks of an old life-rescue ring when seeking help? When one wants to engage with an office productivity suite, do we think of an old typewriter? Scissors and Right Angle rule for 'Accessories'?? Nostalgia aside, it's time Ubuntu revisited 'polish' within a contemporary and aesthetic context.

Placement of icons are also still ugly: look at the 'help' and WWW icons in the menubar of this screenshot: they are several pixels closer to the top of the bar than the base. The icon in the middle is faded out to the point of being a waterstain. Why not replace those menubar icons with words and do away with these bizarre, misplaced symbols. Better still, why not draw upon the abilities of those contributing to Art dot Gnome dot Org or better still, the Tango Project.

Ancient. soggy icons that are poorly placed only impoverish the otherwise striking, and singular, visual field of Ubuntu as a whole.

I know that for you, as a Windows user, Firefox does not use the GNOME file selector dialogs. I don't dispute that.

Under Linux, however, Firefox 1.0.7 does use the GNOME file selection dialogs. Trying to save a file using Galeon brings up the exact same dialog as when trying to save a file using Firefox. This is with GNOME 2.12, mind you, the most recent stable release. I repeat, Firefox does not use its own file selection dialogs.

I'll show you a picture:
http://www.gnome.org/~seth/designs/filechooser-spe c/save-as.png

It's that piece of shit that is used by both.

Firefox 1.0.7 also uses the GNOME 2.12 "Open" dialog. It's not as horrible as the GNOME "Save As" dialog, but it is still nowhere as good as the one offered by KDE, Windows and OS X.

That's because it's not accepted as a valid theory in most circles. I think most /.ers believe in Evolution anyway.

This is not about preaching. This is not about pushing ideology. This is about developers maintaining creative control over the kernel, the kernel that they themselves have been building. Exactly what is your problem with this?

You say here that 'not only do users not care about "Free", but they will actively dislike "Free".' While regrettable, that is fine -- that is the users' prerogative. I sincerely expect they will think more highly of "Free" the tighter the IP and DRM restrictions become, but these related issues have been discussed to better effect in many other threads.

The trouble with your argument here is that the folks building and maintaining the Linux kernel (and most of the rest of userland Linux software too) are precisely the people who are concerned with this "Free" that you apparently couldn't give a rat's ass about. This has nothing to do with wanting 'to push their ideoligy [sic] on others', and everything to do with wanting to stay in control. As A nonymous Coward noted, building an unchanging driver API for the convenience of corporates does nothing for the kernel:

A "stable" binary API removes the possibility of keeping everything up to date and would dramatically show down the adoption of new features and general improvements.

Again, this has nothing to do with pushing ideology. This has to do with developers maintaining control over their own project, a project that has been provided to you as a courtesy out of the strong moral belief of many in the OSS community that the tools we use to get our work done should be freely available.

Furthermore, given the significant number of websites devoted to OSS usability concerns (over 17,000 at last count), I think we can safely regard as invalidated your claim that '"Free" software people ... don't care about makeing [sic] functional easy to use systems.' They would indeed seem to care, but specifically within the scope of making free software -- i.e., they are not interested in kowtowing to your whinging demands that they gut their principles solely to make something passably usable right this very moment, and shackle themselves with a rigid driver API that hampers kernel development far into the future and threatens to scuttle years of effort to make a fully-usable computer system that is not beholden to secret vested interests.

You, sir, appear to be crying out that consumer convenience is more important than freedom. This is much in tune with the prevailing cultural trends in the United States. I find this deeply ironic -- "Give me liberty or give me death" has been turned into "give me convenience or I shall whinge", while "the land of the free and the home of the brave" has become "the land of the sheep and the home of the enslaved". While I understand the frustration apparent in your posting regarding when things do not work, I cannot agree with your sentiment. Some things, sir, are more important than instant gratification. I am deeply sorry that you do not appear able to recognize this.

Case in point: KDE and Gnome both put a lot of work into eye candy, and justifiably so, but neither can give me a list of all the wireless networks around my computer in just 2 clicks in a default setup - but Windows can. I'd imagine OS X probably could too.

Desktop-Linux means KDE in Europe

Thanks for clarification.

You confirm JDS is a failure.
"Hence the major reshuffle, reorganisation and opening up of the JDS process, which you stupidly label as "abandonment"."
Outsourcing to open source. We had that before.

See:
http://www.gnome.org/~gman/blog/2005/Nov/01

I mean, everybody knew JDS was a failure once it was announced. Because it was not what users wanted. It was the idea of SUN people to let Gnome look like Win95 to step into the success of Win95 and bundle it with SUN fashion. 10 years too late.

When SUN ships Gnome with Solaris I am fine with it. But you cannot say solaris is a desktop Operating system. Chosing Gnome means that you do not want to target the desktop market because you will be incapable to fix the system to serve the desktop market needs. Others tried before.

But that's not surprising, considered that GNOME is mainly developed by American developers working for American companies, while KDE is a far more international effort

Mod parent flamebait. Or "Dumb, uniformed crap".

I don't need to explain to anyone that GNOME's a11y tools are far more mature and advanced than KDE's (kudos to Sun), do I? But you could have found that out yourself.

AmaroK music player

Banshee, developed by a Novell employee, is leaps and bounds ahead of any existing music app for linux.

DigiKam

F-spot, also by a Novell employee.

As far as all of the other applications you mentioned, each has a gnome equivalent that in many cases does a better job.