Slashdot Mirror

While Microsoft should be applauded for releasing information about the Office document formats, their manoeuvres around the standards process demonstrate that they are not pursuing standardisation as a platform for innovation for the entire industry. Indeed, Microsoft continues to behave in the abusive manner of an unreformed, convicted monopolist with no passion for true industry collaboration in the interests of users.

• GNOME (and Novell) do not support the standardisation of OOXML. They are both members of the ODF alliance, both use it as the default file format, and if it was even remotely realistic to have a decent office product without OOXML support (where the Windows desktop is unfortunately in such an insane over-dominance currently), then they would of course be all for it.
• The implementation of OOXML is all about interoperability. I don't see anyone (wrongly) trashing Samba as a project, and yet its existence and the effort to implement OOXML support is virtually identical in terms of free software.
• You like software freedom and hate the software patent system? Great, so do I. Free implementations of proprietary solutions, though, are a good thing; not a single one of my friends are going to be using Linux if they can't submit their assignments to their lecturers. We need interoperability, to ease the transition for people coming from the proprietary world.
• The KDE/Koffice developers issued a statement basically saying they didn't have the resources or the time to implement OOXML, and suddenly a lot of silly talk gets thrown at GNOME. If I volunteered to implement OOXML support in Koffice I doubt (i) that they would object, and for sure that (ii) any distribution would not include it.
• Even if you dislike Jeff Waugh, it's pretty tough to find a rational basis for criticising him based on the podcast or his approach to the problem other than (i) not getting the GNOME statement (again, which you really can't fault) out soon enough, or (ii) giving Roy the publicity he wants.
• The itwire article plays Roy as some sort of victim in the podcast talk. That is ridiculous. Unfortunately -- and to the detriment of the FLOSS community -- Roy is an incredibly prolific, poisonous person willing to do or say anything that might cook up some self-publicity, and with an irrational hatred of Novell. And in fact on the contrary, Roy skipped around every question that was directly asked to him; instead opting to just give background on Microsoft's "evil" nature and talking about how bad OOXML is (both of which we palpably know).
• Finally, even if you decide to ignore all the other above facts, please tell me why you're not also staging wide protests against OpenOffice.org or your distribution for including OOXML support, as well.

FWIW, Gnome is pretty consistent now: more than Vista, maybe less than OS X.

The Gnome HIG (Human Interface Guidelines), http://library.gnome.org/devel/hig-book/stable/ , is followed very, very closely. If your app violates the HIG you'll get a stream of annoying bugs filed about it. If you start up Ubuntu it's quite hard to find any non-HIG programs.

KDE have something similar I think, though I don't follow KDE development closely.

http://www.gnome.org/press/releases/ecma-tc45-statement.html

BTW, has anyone ever read and tried to understand this:

http://www.gnome.org/press/releases/ecma-tc45-statement.html

I think having read this most of your points just do not stand anymore.

Try a different reader like Foxit under Windows or Xpdf or Evince under Linux. Acrobat is getting bloated and slower with every release, but looking at how the folks at Adobe ruined Cool Edit Pro (now Adobe Audition) when they acquired it, that doesn't surprise me at all.

The killer had also 5 toes on his foot. This rules out Gnome.

https://lists.linux-foundation.org/pipermail/desktop_architects/2007-February/001127.html

"I did something better than any Gnome user has apparently ever done: I actually wrote the code to fix the thing." - LT

Does he really think he's that much smarter than every Gnome user and developer?

"I want something very simple: I want to configure my mouse button window events. That doesn't sound so bad, does it? Everybody else can do it, gnome does not." - LT

You already have that with KDE. Gnome developers can do what they want. It's really a design choice. "And I find it *offensive* how Gnome people can never just admit that they can't do something." - LT

How absurd. Sounds like a really simple UI feature. Does Linus really think Gnome developers are incapable or is he just trying to taunt them?

Torvalds posted on the GNOME-usability list that "I personally just encourage people to switch to KDE."

http://mail.gnome.org/archives/usability/2005-December/msg00021.html

This, I find kind of odd. If he feels so strongly that KDE is the best, why attack Gnome incessantly?

Why should he try to make Gnome into KDE when KDE already exists? I guess this is a turf battle and he's pissed that Gnome has more market-share. He seems kind of ambitious beyond the scope of merely making his Kernel the best it can be.

Not that my opinion matters, but I think a lot of really talented people are wasting their time getting pulled between OOXML and ODF. Right from Jody Goldberg and a lot of others are spending a lot of time supporting both (and debating why).

And looks like I'm not the only one who thinks that - quoted from Jdub's email to gnome-lists.

> [9] What is your positioning with respect to the issue of OOXML?

An exasperating waste of time -- on both sides of the debate -- that will
ultimately harm international technology standards more than it will ever
help Microsoft's bottom line or harm the absolutely inevitable success of
Software Freedom.

I've already shouted down MooXML, but I think I'm done talking about this, if I'm not going to do anything in particular (say, does the Koffice ODF guys need some help?).

Brasero is my weapon of choice. Now that I've helped you, why don't you find me the native KDE equivalents of Evolution (PLEASE don't say krashy KMail)

Thought so. I just filed a bug about this one.

If so... bug has been filed.

Figured I should file it: http://bugzilla.gnome.org/show_bug.cgi?id=493964

True. s/Abiword/KOffice/ or you can use Abiword with Agnubis and Gnumeric

True. s/Abiword/KOffice/ or you can use Abiword with Agnubis and Gnumeric

Red Hat is doing something close to this through their Mugshot project. It has progressed quite a bit since that Ars Technica write up and is an important component of GNOME's Online Desktop project.

If everyone likes C# so much, then we should take matters into our own hands and implement a language with the features we like that is under our control!

While working for Novell I was able to join both OASIS for ODF and ECMA for OOX. After leaving the OO.o development team at Novell to return to other work I lost both memberships and had to scramble to rejoin. Maintaing Gnumeric is a hobby, and my current employer is not involved in the standardization process. Paying out the membership costs of either OASIS or ECMA was not going to happen at the personal or corporate level. Thankfully there was a non-profit tier available, and the GNOME Foundation generously sponsored me to re-join ECMA and TC45 to continue to participate in the the specification process. After spending the last 8 years playing proctologist to every spreadsheet format around, and complaining loudly at the poor quality of documentation for XLS it seems ridiculous to pass up the opportunity to engage MS, and ensure that the spec of their new format was more detailed than previous efforts.

My personal opinion (not speaking for the GNOME foundation or past or present employers) is that both specs should be standards
      http://www.gnome.org/~jody/files/2007-ON-Linux-Beyond-ISO-Dome.pdf

The FLOSS community is going to need to implement importers for both formats to help our users, and I'll be happiest when both OOX and ODF are significantly clearer. 5700 pages of OOX is too _short_, Likewise the 700 + 300 (Open formula) in ODF is far too short. Lets double the size of OOX (although with better formatting the number of pages would likely be unchanged), and lets quadruple the level of detail in ODF to get it into a useful state. I only wish that ODF had undergone a fraction of the review that OOX has seen.

This is not about GNOME endorsing OOX, it's about GNOME doing the work necessary for users. There should be reps from Sun's OO.o team on the ECMA TC, and MS reps in the ODF meetings. The goal of this process is to produce useful documentation, and it takes an implementor to know where the really important details are. It hardly seems in the best interest of the FLOSS community to leave the standardization efforts up to corporate interests at Microsoft, Sun, or IBM.

The GNOME Foundation does not support ISO standardisation of OOXML. But whether or not that happens, we're still going to have to support Microsoft document formats, just like everyone else. Should we let Microsoft shove OOXML through ECMA without challenge? Hell no. That's why we have one of our best hackers in there, holding their feet to the fire.

But whether or not that happens, we're still going to have to support Microsoft document formats, just like everyone else.

I looked at the gnumeric developpement version and nothing is done to support ODF but everything is done to support Microsoft OpenXML. It's a shame this software was a great one!

Having Gnome team members promoting the agenda of its main opponent, however, is not only counter-productive but also reflects negatively on the project and its credibility.

For example, one high profile team member can cause a lot of trouble for Gnome, especially when promoting proprietary technologies in opposition to open source and open standards. Quotes like, Time to play with C#, ASP.NET and some nifty toys (you can make almost Windows feel like Linux now) seem to be promoting themes advanced by bloggers at Gnomes (and open standards) main antagonist, Microsoft.

For your peeve #1, the solution is simple. Go to the Preferences dialog and uncheck "Activate the focused image" in the Window Management section. The gimprc manual page explains this setting:

        (activate-on-focus yes)

            When enabled, an image will become the active image when its
            image window receives the focus. This is useful for window man
            agers using "click to focus". Possible values are yes and no.

Peeve #4 is also simple to adjust to. Just change the Move tool to "Move the active layer" instead of "Pick a layer". This can be conveniently toggled with the Shift key (as shown in the tool options).

There's a patch for #2 (Editing at image boundaries) in the bug-tracker (bug #362915). It has issues but I very much hope that we can fix them and get it in for 2.6.

I'm partial to the Geany code editor. It has all of the features I need, seems to support just about every language you'd like (both in terms of highlighting and running), and still seems very light-weight.

For designing graphical interfaces, it doesn't get any easier than GLADE. No auto-generated code, just an XML file that can be loaded by any language that has a port of libglade, and like that, you have a GUI for your programme.

Both can be found in the package manager, but here are their websites if you're interested.

Geany: http://uanews.org/node/16426
Glade: http://glade.gnome.org/

We've just tried this one out as soon as it was released, and there was quite some differences in installation on our modern laptop hardware compared to 7.04 at least. Proprietary graphics driver installation couldn't really be much easier from what I can see -- besides by making it automatic, but I suppose there are reasons other than technological ones behind that.

After the few guided clicks to get that done, a reboot later and suddenly Compiz was also activated without any user actions needed. Hmm, so how do you configure those 3D effects then? No way we could find, but from reading an online computer magazine, we found out that the Compiz Config Settings Manager wasn't included. We installed that one, and it then integrated nicely into the Desktop Settings as a new "Custom effects" option. Why that one wasn't part of the distro by default is still unknown to us though. It seemed like an obvious choice to let the user customize the window effects?

Otherwise, I think Compiz didn't lag or anything even once when maximizing windows or rotating the desktop, etc, and this was on a laptop without a *that* hot graphics card. So we were impressed about how smooth the UI was. No interruptions from some service suddenly kicking in to work a bit like every user of Vista has no doubt grown used to take for granted by now with the SuperFetch, System Restore, Search Indexer etc services. They seem to kick in at the most inappropriate times -- not even when the computer is idle! Come on! Maybe Ubuntu's new desktop search indexer make it suffer too, but nothing we could see anyway.

After doing this, we unplugged the network card, and voila, it automatically discovered our WLAN. We didn't have to do anything, really.

So let's try open the (already mounted and ready) NTFS drive with Windows Vista on it? Oh, we can simply drag a file there now too -- cool! NTFS-3g apparently installed and ready.

We seemed to have to install Windows Media Audio support though and as we're still quite some Linux amateurs, we have still not got around that part as the work day is over. It's been fun experimenting though, and getting up to date with what a modern "desktop Linux" distro can offer. Looking at the feature list of Ubuntu 7.10, and summing that one up with the new features of GNOME 2.20 gives one a mighty impressive list of new features compared to just 6 months ago.

Linux desktop development (GNOME, KDE, desktop distros, ...) really seem to be picking up some pace lately. And we're just months away from KDE 4. This is exciting times to follow for sure, and for the first time I'm starting to become a believer in "Linux on the desktop".

I have some pretty high demands of novice usability though, which makes me hesitate still as for some distros. E.g. SUSE Linux 10.3 had a few quirks on my home stationary computer. Its NVIDIA driver install having me to use the command prompt and special "SUSE for NVIDIA" instructions is unacceptable for amateur usage IMHO, although I finally got it done. It also even failed to install the distro to the hard drive the first time around, because it couldn't mount the SATA drive it had just formatted (??). A reboot, and then it could do it like it was no problem at all. *shrug* That also gave an early feeling of "still aimed for geeks" that I'd so much like it to see it move away from.

But back to Ubuntu 7.10 -- so far no problems here, and I was left with an excitement to play with it more after the day. :-)

I'm actually really excited about this. We've got a demo running here. We installed it on a two year old notebook and everything just worked. Pointed Evolution to our Exchange server, and it just worked. Which IMHO is key, I love to hack things just as much as the next guy, but if I have to hack things just to get them to work the first time, its a major turn off.

It's got a slick UI and the package manager is well done.

Add in support from Dell.

All that is missing now is a really awesome developer environment.

Of course, this kind of thing never happens on a Linux Desktop.

• GNOME http://library.gnome.org/devel/glib/stable/glib-Perl-compatible-regular-expressions.html
• Qt http://doc.trolltech.com/4.3/qregexp.html
• WxWindows http://www.wxwidgets.org/manuals/stable/wx_wxregex.html
• .Net http://msdn2.microsoft.com/en-us/library/system.text.regularexpressions.aspx
• Java http://java.sun.com/j2se/1.5.0/docs/api/java/util/regex/package-summary.html

The only one i've heard of (for Windows) is Foxit PDF reader, which is about 2mb - never tried it myself though. On linux, Evince works great, and had no issues with everything i've thrown at it.

Awww ok then.

Here is a picture I took that you might find interesting:

http://art.gnome.org/backgrounds/nature/2481

In my opinion, Linux distros must provide a means of doing away with text configuration files, but still retain the ability to access them for those who wish.

Well, GNOME did exactly that. Try Ubuntu and take a look.

The GUI exposes the most common preferences (look at System / Preferences or System / Administration). For the more obscure preferences, there is something remarkably similar to Registry Editor; it's called gconf-editor. (It's a measure of the success of Ubuntu that I haven't really needed to use gconf-editor for anything in years. The standard preferences are doing it for me.)

The GNOME guys were heavily flamed on Slashdot for making something that looks like the Windows Registry but I think it's a good idea. If two processes both try to write a text file, there is a possible race condition where the first process updates the file and the second process clobbers the update from the first one. The GConf system manages the updates, so that doesn't happen. Yet the back-end storage of data is still plain text files, so if you have to boot in single-user mode to recover after a disaster, you can still just use your favorite text editor to tweak the settings.

steveha

Well hold on--there are two separate classes of windows that we are talking about here. The document windows, and the toolbox and palette windows.

If I raise a document window, I already do get the toolbox and palette windows raised. So I assumed you were talking about the other image document windows being raised -- not the toolbox and palettes.

However after replying I did some thinking and decided that I would find it useful if all the GIMP's document windows were raised together. I filed a bug http://bugzilla.gnome.org/show_bug.cgi?id=483159 but it was promptly RESOLVED NOTABUG. Oh well.

Oh, I don't think they're unrelated. Novell's strategy is clearly to be a parasite on the "Open Source" community. Microsoft can use them to reap revenue /and/ simultaneously undermine the threat, so they win either way. Novell's only hope of surviving as a distro is to make sure that: 1) as few projects as possible adopt GPLv3; 2) they don't make any copyright assignments which allow projects to later relicense the work as GPLv3.

We've seen Greg Kroah-Hartmann's recent announcement of a GPLv2 only driver project (with copyright assignments to them, thus precluding relicensing later as GPLv3), now we see them ensuring that they have an office suite. This is a good make-or-break time to put pressure on sun and hopefully sway enough contributors/developers to go with the Novell fork, because Sun let's face it, are not particulary nice trustworthy guys either. If they want OO.o to be an actual Free project then they need to set up a non-profit foundation like Mozilla, but Sun can't afford to take that risk because it's the last thing keeping the possibility of an OpenSolaris gaining traction in the enterprise desktop.

MichaelMeeks admits in his own blog posts that the purpose of the JointCopyrightAssignment is to allow the holder to change the licensing: Novell's fear is that sun will license it as GPLv3 instead of LGPL.

This is all Novell trying to make sure that they harvest as much community IP as possible before people start pulling up the drawbridges and dropping the portcullises.

'course, if the judge who walked in with the seeing-eye dog was an Orca user, he'd of known all about Ubuntu as it's the place lots of blind folks use their Orca screen reader. Then there wouldn't have even been that $50 fine...

Don't forget Dia to replace that unactivated Visio for free. For windows, or linux, or macs.

Dia is a OSS replacement option that runs on Unix and MS Windows. http://live.gnome.org/Dia

It's back now. :)

You are joking, right? Xpdf lacks all kinds of features useful in the corporate world. Forms that can be filled out is one. PDF is an open format, and Adobe publishes the standard for your convenience, but even after years of work Xpdf and offshoots like libpoppler still can't support much more than they did years ago.

The following picture I took might make you happy:

http://art.gnome.org/backgrounds/nature/2481

Yes, of course, "everyone" knows it is true. I'd like, just once, even one example of this.

Yes and I would like, just once, an example of the common myth that water is wet.

Yet, I am loathe to let you wallow in your ignorance, so I've done a quick search for you. I follow gnome development only from a distance, so I'm sure I've missed a lot here.

The menu editor, removed somewhere in the 2.x cycle, not replaced until years later: http://www.linux.com/articles/57088
Sawfish replaced with Metacity, losing tons of features/configurability. http://mail.gnome.org/archives/usability/2002-December/msg00069.html
Galeon 1.2 replaced with Galeon 1.3, losing features, and then later replaced with Epiphany, losing more features.
http://wouterverhelst.livejournal.com/46098.html
xscreensaver replaced with gnome screensaver, which has no options at all https://bugs.launchpad.net/ubuntu/+source/gnome-screensaver/+bug/22007

And an example where important features are intentionally not implemented for usability reasons:
https://lists.linux-foundation.org/pipermail/desktop_architects/2005-December/001587.html

There are many more, this list was just the product of a quick google search

True enough, it takes a certain amount of effort, but then again, so does complaining about it on slashdot.

FWIW, someone had taken the effort, looks like http://bugzilla.gnome.org/show_bug.cgi?id=415675, it's already fixed and even backported to 2.18.x branch (happens in Feisty with 2.18.0 but not in F7 with 2.18.3), so better start looking for new tricks ;)

Could be related to this: http://bugzilla.gnome.org/show_bug.cgi?id=420713 The detailed changelog is: Fix a crash when dragging a window to somewhere like another workspace/viewport. The original drag source widget might have been destroyed. We're now more solid against this, and use another drag source widget which lives longer. Fix bug #420713. Anyway, if it's not fixed run the applet or the panel (whichever it is that is crashing) under gdb. For an applet you do this the following way: gdb /usr/libexec/wnck-applet (gdb) r Then add the pager to the panel via right click -> add to panel Reproduce the bug and type bt in gdb to get the backtrace. Chances are that we can find out whether it's really fixed or just harder to reproduce now.

Mono is required for the full official GNOME desktop release as described on the GNOME web site. Note the list stating what order to compile things in, and that tomboy is listed.

The fact that you can rip Mono out and still run most of the GNOME desktop does not mean that Mono is not part of GNOME.

Mono is the reason why I switched from GNOME back to KDE.

Huh, did you actually read through the whole thread that you linked? Here is a relevant post:

  Comment #48 from William Jon McCann (gnome-screensaver developer, points: 22)
2006-09-26 17:08 UTC [reply]

Matthew and others:

There is really no reason to be rude or to encourage forking. In the FAQ,
regarding creating or modifying themes, is says:
"There is work to be done to make this simpler. Ask how you can help."

Not a single person has ever asked how they can help.

In the FAQ entry mentioned in comment #10 I added some information about
technical issues that need to be addressed (ie. code written) before this can
even be possible.

There are also cross-desktop compatability issues to be considered.

Here are a couple of the issues:
http://bugzilla.gnome.org/show_bug.cgi?id=355488
http://bugzilla.gnome.org/show_bug.cgi?id=354811

There have been various misreadings of what I have said about this issue. This
probably has something to do with the brevity of some of my replies, due to
lack of time, and I'm sorry for that. I'll try to summarize and explain a
little better:

* My view is still that any screensaver theme that *requires* configuration is
inherently broken. The key word here is requires. It means that the defaults
should work in a reasonable way that no one should *have* to change them.

* I am not denying that some people may want to change the defaults and
settings. I am saying that simply adding configurability like xscreensaver has
is not a good solution and I've listed some reasons in the FAQ. Please notice
that the original reporter asked for the "ability to configure individual
screensavers like Xscreensaver". That simply won't work. We need a new way
and that is going to take some actual work.

* I am saying that this isn't my highest priority and that if it is very
important to someone they should help out.

Jeez, these comments are pretty disheartening to someone spending years of his
own time on...

Thank you.

Huh, did you actually read through the whole thread that you linked? Here is a relevant post:

  Comment #48 from William Jon McCann (gnome-screensaver developer, points: 22)
2006-09-26 17:08 UTC [reply]

Matthew and others:

There is really no reason to be rude or to encourage forking. In the FAQ,
regarding creating or modifying themes, is says:
"There is work to be done to make this simpler. Ask how you can help."

Not a single person has ever asked how they can help.

In the FAQ entry mentioned in comment #10 I added some information about
technical issues that need to be addressed (ie. code written) before this can
even be possible.

There are also cross-desktop compatability issues to be considered.

Here are a couple of the issues:
http://bugzilla.gnome.org/show_bug.cgi?id=355488
http://bugzilla.gnome.org/show_bug.cgi?id=354811

There have been various misreadings of what I have said about this issue. This
probably has something to do with the brevity of some of my replies, due to
lack of time, and I'm sorry for that. I'll try to summarize and explain a
little better:

* My view is still that any screensaver theme that *requires* configuration is
inherently broken. The key word here is requires. It means that the defaults
should work in a reasonable way that no one should *have* to change them.

* I am not denying that some people may want to change the defaults and
settings. I am saying that simply adding configurability like xscreensaver has
is not a good solution and I've listed some reasons in the FAQ. Please notice
that the original reporter asked for the "ability to configure individual
screensavers like Xscreensaver". That simply won't work. We need a new way
and that is going to take some actual work.

* I am saying that this isn't my highest priority and that if it is very
important to someone they should help out.

Jeez, these comments are pretty disheartening to someone spending years of his
own time on...

Thank you.

Check out Vala. It's what C++ should have been.

-Mike

Another reason to follow the progress of the Vala project is that it would seriously jeopardize the future of Mono, by making its main drive irrelevant.

If you remember the history, Mono was introduced quickly and with some controversy. Miguel de Icaza and Ximian originally envisioned Mono as allowing an easy way to develop applications for the GNOME desktop, De Icaza's more famous project. The problem with Java at the time was that its license was not easily compatible with the GPL. While there were a few open source Java VMs out there, the projects themselves were problematic and mired in all kinds of problems, including legal uncertainty. Meanwhile, experimenting with C#, De Icaza managed to develop a quick-and-dirty C# compiler. The open source community didn't really care if it was C# or Java, as long as they had a virtual machine of their own. De Icaza had already proved himself capable, so he got the support he needed, and things moved very quickly.

In essence, though, the reason Mono exists is legal: Java's license. Mono definitely was not adopted as a way to make Windows .NET applications run on Linux. That was a "side" benefit.

Things have changed a lot since then. Sun's Java implementation is now moving quickly to GPL, making it a much more mature alternative to Mono for open source. This same opening up, at the same time, has made it possible to include Sun's original Java classes into GCJ, making GCJ much stronger than it was before. There are already trunk versions of GCJ using those classes. (Much of GCJ's weakness was the slow-moving progress of the CLASSPATH project, and open-source equivalent to Sun's Java classes.)

Meanwhile, at GNOME, there's been much controversy over using a virtual machine on the desktop in the first place. A few GNOME projects have been written in Mono, as feasibility studies, and while they worked, users found them sluggish and overblown. Users simply don't like virtual machines on their desktop, and there's no reason why they should.

The Vala project goes back cleanly and directly to the original impetus for developing Mono: allowing the productivity of Java and C# for the open source GNOME desktop. It does this, in fact, far, far better than Mono. GObject, the core of Vala, is also the core of GNOME. All those GNOME projects written in Mono could easily be ported to Vala, immediately making them first-class binary applications, fully interoperable with everything else in GNOME. Just like C# is

Another reason to follow the progress of the Vala project is that it would seriously jeopardize the future of Mono, by making its main drive irrelevant.

If you remember the history, Mono was introduced quickly and with some controversy. Miguel de Icaza and Ximian originally envisioned Mono as allowing an easy way to develop applications for the GNOME desktop, De Icaza's more famous project. The problem with Java at the time was that its license was not easily compatible with the GPL. While there were a few open source Java VMs out there, the projects themselves were problematic and mired in all kinds of problems, including legal uncertainty. Meanwhile, experimenting with C#, De Icaza managed to develop a quick-and-dirty C# compiler. The open source community didn't really care if it was C# or Java, as long as they had a virtual machine of their own. De Icaza had already proved himself capable, so he got the support he needed, and things moved very quickly.

In essence, though, the reason Mono exists is legal: Java's license. Mono definitely was not adopted as a way to make Windows .NET applications run on Linux. That was a "side" benefit.

Things have changed a lot since then. Sun's Java implementation is now moving quickly to GPL, making it a much more mature alternative to Mono for open source. This same opening up, at the same time, has made it possible to include Sun's original Java classes into GCJ, making GCJ much stronger than it was before. There are already trunk versions of GCJ using those classes. (Much of GCJ's weakness was the slow-moving progress of the CLASSPATH project, and open-source equivalent to Sun's Java classes.)

Meanwhile, at GNOME, there's been much controversy over using a virtual machine on the desktop in the first place. A few GNOME projects have been written in Mono, as feasibility studies, and while they worked, users found them sluggish and overblown. Users simply don't like virtual machines on their desktop, and there's no reason why they should.

The Vala project goes back cleanly and directly to the original impetus for developing Mono: allowing the productivity of Java and C# for the open source GNOME desktop. It does this, in fact, far, far better than Mono. GObject, the core of Vala, is also the core of GNOME. All those GNOME projects written in Mono could easily be ported to Vala, immediately making them first-class binary applications, fully interoperable with everything else in GNOME. Just like C# is