Please be a good OSS denizen and file bugs at BGO or (preferably) SourceForge. Some of what you mention is possible to do within the GTK theming/settings framework, and some is simply out of the theme-writer's hands presently.
If people identify and file problems, I know what's bugging them and can work to resolve them. But I won't track down and resolve issues on newsgroups and message boards like Slashdot.
Finally, GTK-Wimp's look will "fall back" fairly nicely on Win32 systems that don't have UXTheme installed (2k,9x) or on XP/2k3 where it isn't running. In those cases, it's basically the Raleigh theme engine with Win32's color palette and system settings in effect.
As Bryce pointed out in TFA, the Inkscape guys (as well all of the Gnome community) follow the Human Interface Guidelines, which lead to more "standard" programs. Bryce even says that Inkscape follows that standard even though he disagrees with parts of it. Bryce also talks about the initiatives going on over @ freedesktop.org and how promising many of them look.
When people talk about "standards" in the OSS world, they often toss off references like your spreadsheet cut + paste example. Those kinds of categorizations are a bit disengenuous - in the Windows world, there is a standard - Microsoft's way or the highway.
The root problem, IMHO, is not one of egos but of organization. Windows has one vendor, one monarch, and his name is Bill Gates. Under him is a hierarchy of managers, engineers, UI folks,... In this schema, there is a good standards enforcement policy - do it this way, or you're fired. There's no question about who is in charge.
The fact that projects the size of Gnome and KDE have guidelines that their developers follow is amazing. The fact that we're creeping toward a set of guidelines that cross desktops is even moreso. These facts alone invalidate your first paragraph and its assumptions.
What's frustrating is that people beg for freedom of choice across the board - freedom to choose an app, protocol, file format, library, language, whatever... to get the job done. But then they complain when two apps don't approach the same problem from the same angle because of "ego" or that they now have N "redundant" choices or X dependencies. You can't have it both ways.
Here, I have to disagree. Putting up billboards across the street is not the same thing. This is more like putting a phone tap on the line, and setting it up so that each time someone calls Geico, they get an intercept telling them how wonderful State Farm is.
Importantly, nothing gets intercepted, because the main search results still go through to Geico. What's added is a small annotation saying that if you're interested in Geico, you might want to check out State Farm instead.
If rulings like this stand, I think the next rung of lawsuits against search engines happens when Geico isn't the first search hit for "Geico" or if "State Farm" shows up second on the search results page. [tongue-in-cheek]Brand Confusion, I tell you![/tongue-in-cheek] At least in the current case, "State Farm" is marked as an advertisement and displayed *distinctly* from the main results.
What else would rulings like this ban? Miller ads that say - "half the carbs of Bud light" Or "Aleve lasts twice as long as Tylenol"? Comparitive advertisement is a cornerstone of the US marketplace. I understand that it is much less so (if not altogether verboten) in Europe.
In any case, whatever you want to call this, it ain't TM infringement.
IMHO, this is a pretty bogus ruling. The ads are distinct from the regular results and marked as such, so there's little grounds for any of the typical trademark "confusion" statues to apply. I don't think that this case would pass muster in a US court.
In short, I fail to see how Google selling Louis Vuitton adwords to LV's competitors is any different than State Farm putting up billboards across the street from Geico offices or Viagra buying an ad on the page after a Cialis article in Men's magazine. It's nothing more than good business sense. It isn't TM infringement.
Some browsers (KHTML, the rendering engine that KDE's Konqueror and Apple's Safari browsers use, comes to mind) do in fact strive for pretty-much bug-for-bug compatibility with MSIE.
There's no arguing here - what "Netscape" ended up with here is a UI abomination. I'm just saying that I can understand why a browser developer in general would want "IE compatibility".
And, quite frankly, the reason that MSIE exports all of those COM controls/interfaces is because you *shouldn't* have to do any real work to use a browser widget. Like them or not, Microsoft got that point right. And since they're interfaces, some clever person can (and indeed, has) re-implement those interfaces in terms of Mozilla.
Like it or not, a lot of corporations have at least 1 browser-based ActiveX control that their employees must use. Allowing Mozilla to run these programs would eliminate a major barrier to entry.
The point in supporting IE rendering is that a large number of pages just don't work with Mozilla or refuse to render "correctly". For this reason, some browsers strive for bug-for-bug compatibility with MSIE.
Of course, that isn't to say that these are necessarily good or bad decisions, or that there aren't better solutions out there. Just that it's easy to understand these decisions and the motivations behind them even if you don't necessarily agree with them.
If you're on Unix and have librsvg installed, you can have librsvg act as a Mozilla SVG plugin. It integrates well with both Mozilla/Firefox and Gnome. The last time I checked, it is more standards-compliant and featureful than Mozilla's own SVG implementation.
From the email threads and writeups on KernelTrap, it seems as though Linus (and his Lieutenants) have some issues with the invasiveness and maintainability of the patch, which are reasonable concerns from the maintainers.
Ingo Molnar - a RedHat employee/kernel hacker - has some patches that are similar in scope but different (and most likely preferable from a performance and maintainability viewpoint) in approach.
Reiserfs will apparently soon have what you're looking for. Already, all primitive operations are atomic, but they plan on exporting a user-space transaction interface soon.
http://www.namesys.com/benchmarks.html
"V4 is a fully atomic filesystem, keep in mind that these performance numbers are with every FS operation performed as a fully atomic transaction. We are the first to make that performance effective to do. Look for a user space transactions interface to come out soon....
Finally, remember that reiser4 is more space efficient than V3, the df measurements are there for looking at....;-) "
Those 2 women in the richest 10 are the wife and daugter of Sam Walton, founder of WalMart. They didn't get rich on their own - they inherited 1/5 of daddy's money when he died. The family (including these 2 women) has no control over the store's operation. They merely own 38% of the stock.
This fact doesn't validate or invalidate your point, but it's worth knowing.
http://www.forbes.com/maserati/billionaires2004/ bi ll04land.html
*) Gaim does ship with my theme. As does dropline GTK. As does the Gimp's GTK.
*) The GTK 1.2 port for OSX does suck visually, but you mentioned "portability", so I thought I'd throw that in just for kicks. I wouldn't suggest using it in a production setting. They are, however, making some progress using the OSX Appearances Manager API:
http://gtk-osx.sourceforge.net/pix/gtktestthing. aq ua.buttons.jpg
*) The QT3 OSX theme also uses the Appearances Manager APIs. It's not perfect, but I don't think it's as bad as you think it is. Also, doing your evaluation on KDE's "Ranger Rick Ported" software probably isn't a good test. Adobe's is probably a better benchmark. It will also be interesting to see what comes of this: http://www.trolltech.com/campaign/maccontest.html? cid=11
XUL, XPCOM, NSPR, NSS, etc... are promising - XUL especially so when using the Firefox toolkit. I'd love to use them in my OSS and commercial apps. Some things (NSPR, NSS) are available as separate libraries right now. But many things essential to 3rd-party developers are progressing at a snail's pace, making the Mozilla "platform" anything but one.
1) Get GTK 2.2.4-2 from http://dropline.net/gtk/. It includes my native Win32 theme. 2) Get Gimp 2.0 from http://www2.arnes.si/~sopjsimo/gimp/stable.html 3 ) Uninstall any old GTK, Gaim, and Gimps, just to be safe. 4) Install GTK 2.2.4-2 5) Open regedit 6) Change HKEY_CURRENT_USER/Software/GTK/2.0/Version to read "2.2.4-20040124" 7) Change HKEY_LOCAL_MACHINE/Software/GTK/2.0/Version to read "2.2.4-20040124" 8) Install the Gimp 9) Enjoy
"GTK is highly portable, but it is however horrible when it comes to any platform but linux. It runs in windows, without a windows theme to match. it runs in os x, but only via x11."
BS. I guess that I didn't write the Gtk-wimp theme engine, then. And I guess I'm not running Gimp, Gaim, Xchat, Dia, Inscape, and Workrave on Win32 right now... The gtk-wimp screenshots are out of date, btw, but are good enough to get the general idea across.
GTK 1.2 has also been ported to Apple's Carbon framework. I have plans to do something similar for GTK 2.x.
"Qt's one of the better-off toolkits, but it suffers from gtk's problems too. It's only good on linux. half-assed os x port (at least it's not x11 dependent) and the windows port is totally unmaintained (yay 2.0!)."
Huh? Trolltech gets a lot of money from licensing QT 3.3 on Win32. I've licensed it. It's maintained, and works great. The OSX port isn't as half-assed as you'd like to think. It's not FOSS on Win32, like it is on OSX and *NIX, but that's Trolltech's decision, not mine.
You're right, XUL may be horribly under-appreciated. But realize that's largely Mozilla's own doing. They've been promising a standalone GRE for about 5 years now, and have yet to follow through. Much of the Mozilla platform is meant for eventual public consumption, but it's not ready for that today. When I can 'gcc -o myapp -lMozillaGRE -lMozillaJScript', come tell me.
Definitely. I wonder if he's aware that the latest SuSe 2.4.x kernel has no fewer than 2400 patches, many of which were backported from the 2.6 series...
Hello stoplights, hello tolls. Hello $30/day Manhattan parking lots, and hello to those half-hours wasted circling the street looking for an open spot. Say hello to the pedestrians and bikers, darting out in front of you. Hello traffic jams, honking horns, and cursing, irratic drivers. Hello noxious fumes and single-digit speeds on urban highways.
Say goodbye to reading the newspaper on the way to work. Goodbye to the half-hour nap you took on the train each morning. Goodbye to your stress-free commute.
I posted this on OSNews, but I'll post it again here...
I'm not sure why so many people think that SVG is slow. It doesn't have to be, even without hardware acceleration. I've done tests of librsvg vs. libpng:
Given a SVG image $s. Transform it into a PNG image $p using librsvg, Batik, or something similar. Run "gdk_pixbuf_new_from_file()" on both $p and $s. This will turn $s and $p into identical RGBA images. Time this operation. Lather, rinse, repeat.
Generally, it is no slower (if not faster!) to render $s than $p. This surprised me and quieted many "Vector graphics are too slow for the desktop" pundits.
Of course, once you start using some of the more advanced features (like certain filters), the rendering time is likely to go up. It all depends on what features you use and how you use them.
If this case is not hypothetical, it is a serious matter, and you should do one or more of the following:
1) Contact a lawyer. Have him/her write a Cease and Desist order to $company. Contemplate further legal action against $company for copyright violation. You may also be able to hit them with something akin to "profitting from stolen goods". IANAL.
2) Contact the FSF. They've dealt with these matters before. Bradley Kuhn and Eben Moglen will talk to you over email, phone, and in person. There are good guys and will defend your rights, tooth-and-nail.
I've been through this situation before. Eben Moglen said something like the following to me:
"People always ask me, 'Is the GPL enforcable? After all, it has never been proven in court.' I always respond that it has never gone to court because the infringers don't have a leg to stand on. It is always in their best interest to settle on our terms."
There's quite a bit of inter-operability work going on at freedesktop.org. There's a lot of shared specifications and software there. Plus there are software libraries that both DEs use that aren't listed on FDO, like libxml2.
The KDE folks have also worked on some Qt-GTK toolkit inter-operability stuff. See also:
Actually, the biggies (Direct3D, DirectDraw, DirectInput, DirectSound) are each about 80% done. Specifically, these APIs are what most game writers use. The reality here is that you can have high game-coverage while having relatively low API-coverage. This is because most games use the same subset of APIs and avoid odd or esoteric features. This ultimately means that great number of DirectX <= 8.1 applications work well under Wine.
Transgaming is not Wine. Transgaming is a company that has built a business model around WineX - a fork of the Wine project which is not entirely licensed under the LGPL. Contrary to your assertion, Transgaming does occasionally contribute code back upstream to the Wine project and several of their modules are licensed under the LGPL, though DirectX is not one of them.
For a list of games and applications that work well under Wine proper, I suggest checking out:
IBM could do as you suggest. Then again, they have already written their own Java compiler (Jikes), at least one of their own JVMs, their own servlet container (Jakarta), etc...
I'd recommend looking at this page for more info on IBM + Java + OSS.
IBM has already written at least one high-quality JVM implementation which is not OSS because of contracts that IBM has with Sun. Of course, suggesting that IBM work on GCJ and Classpath has some merit in and of itself. But realize that IBM has sunk untold man-hours and dolars into developing its own JVM - resources that they now wish to contribute to the community at-large as OSS. I personally can't blame them if they didn't wish to spend a similar amount of resources on GCJ and Classpath when what they've got works.
Perhaps with this Open Letter IBM is looking for permission to open up the code. Perhaps they are looking to collaborate with Sun to create an even better project. Perhaps this is all just marketing/PR bs. Time will tell.
Tk is a horrible example of a GUI toolkit tied to a language. Even if you wanted to write using Perl::TK or TK in C, you needed a TCL interpreter installed.
QT and GTK+ have a fabulous assortment of high-quality language bindings. You aren't tied to using C or C++, respectively. I'm familiar with the Python, C++ and C# GTK+ bindings, and can say that they are superb.
Slashdot Mirror
Please be a good OSS denizen and file bugs at BGO or (preferably) SourceForge. Some of what you mention is possible to do within the GTK theming/settings framework, and some is simply out of the theme-writer's hands presently.
If people identify and file problems, I know what's bugging them and can work to resolve them. But I won't track down and resolve issues on newsgroups and message boards like Slashdot.
Finally, GTK-Wimp's look will "fall back" fairly nicely on Win32 systems that don't have UXTheme installed (2k,9x) or on XP/2k3 where it isn't running. In those cases, it's basically the Raleigh theme engine with Win32's color palette and system settings in effect.
Best,
Dom Lachowicz, GTK-Wimp maintainer
As Bryce pointed out in TFA, the Inkscape guys (as well all of the Gnome community) follow the Human Interface Guidelines, which lead to more "standard" programs. Bryce even says that Inkscape follows that standard even though he disagrees with parts of it. Bryce also talks about the initiatives going on over @ freedesktop.org and how promising many of them look.
... In this schema, there is a good standards enforcement policy - do it this way, or you're fired. There's no question about who is in charge.
... to get the job done. But then they complain when two apps don't approach the same problem from the same angle because of "ego" or that they now have N "redundant" choices or X dependencies. You can't have it both ways.
When people talk about "standards" in the OSS world, they often toss off references like your spreadsheet cut + paste example. Those kinds of categorizations are a bit disengenuous - in the Windows world, there is a standard - Microsoft's way or the highway.
The root problem, IMHO, is not one of egos but of organization. Windows has one vendor, one monarch, and his name is Bill Gates. Under him is a hierarchy of managers, engineers, UI folks,
The fact that projects the size of Gnome and KDE have guidelines that their developers follow is amazing. The fact that we're creeping toward a set of guidelines that cross desktops is even moreso. These facts alone invalidate your first paragraph and its assumptions.
What's frustrating is that people beg for freedom of choice across the board - freedom to choose an app, protocol, file format, library, language, whatever
Here, I have to disagree. Putting up billboards across the street is not the same thing. This is more like putting a phone tap on the line, and setting it up so that each time someone calls Geico, they get an intercept telling them how wonderful State Farm is.
Importantly, nothing gets intercepted, because the main search results still go through to Geico. What's added is a small annotation saying that if you're interested in Geico, you might want to check out State Farm instead.
If rulings like this stand, I think the next rung of lawsuits against search engines happens when Geico isn't the first search hit for "Geico" or if "State Farm" shows up second on the search results page. [tongue-in-cheek]Brand Confusion, I tell you![/tongue-in-cheek] At least in the current case, "State Farm" is marked as an advertisement and displayed *distinctly* from the main results.
What else would rulings like this ban? Miller ads that say - "half the carbs of Bud light" Or "Aleve lasts twice as long as Tylenol"? Comparitive advertisement is a cornerstone of the US marketplace. I understand that it is much less so (if not altogether verboten) in Europe.
In any case, whatever you want to call this, it ain't TM infringement.
IMHO, this is a pretty bogus ruling. The ads are distinct from the regular results and marked as such, so there's little grounds for any of the typical trademark "confusion" statues to apply. I don't think that this case would pass muster in a US court.
In short, I fail to see how Google selling Louis Vuitton adwords to LV's competitors is any different than State Farm putting up billboards across the street from Geico offices or Viagra buying an ad on the page after a Cialis article in Men's magazine. It's nothing more than good business sense. It isn't TM infringement.
Some browsers (KHTML, the rendering engine that KDE's Konqueror and Apple's Safari browsers use, comes to mind) do in fact strive for pretty-much bug-for-bug compatibility with MSIE.
There's no arguing here - what "Netscape" ended up with here is a UI abomination. I'm just saying that I can understand why a browser developer in general would want "IE compatibility".
And, quite frankly, the reason that MSIE exports all of those COM controls/interfaces is because you *shouldn't* have to do any real work to use a browser widget. Like them or not, Microsoft got that point right. And since they're interfaces, some clever person can (and indeed, has) re-implement those interfaces in terms of Mozilla.
While this may not be exactly the same thing AOL is using, it's interesting and topical nonetheless:
Mozilla ActiveX Project
Mozilla ActiveX Control
Like it or not, a lot of corporations have at least 1 browser-based ActiveX control that their employees must use. Allowing Mozilla to run these programs would eliminate a major barrier to entry.
The point in supporting IE rendering is that a large number of pages just don't work with Mozilla or refuse to render "correctly". For this reason, some browsers strive for bug-for-bug compatibility with MSIE.
Of course, that isn't to say that these are necessarily good or bad decisions, or that there aren't better solutions out there. Just that it's easy to understand these decisions and the motivations behind them even if you don't necessarily agree with them.
If you're on Unix and have librsvg installed, you can have librsvg act as a Mozilla SVG plugin. It integrates well with both Mozilla/Firefox and Gnome. The last time I checked, it is more standards-compliant and featureful than Mozilla's own SVG implementation.
Link here: http://librsvg.sf.net
Dom Lachowicz
From the email threads and writeups on KernelTrap, it seems as though Linus (and his Lieutenants) have some issues with the invasiveness and maintainability of the patch, which are reasonable concerns from the maintainers.
Ingo Molnar - a RedHat employee/kernel hacker - has some patches that are similar in scope but different (and most likely preferable from a performance and maintainability viewpoint) in approach.
Read about them here and form your own opinion:
Linux: Real Time Kernel Prototype
Linux: Realtime Preemption
Reiserfs will apparently soon have what you're looking for. Already, all primitive operations are atomic, but they plan on exporting a user-space transaction interface soon.
http://www.namesys.com/benchmarks.html
"V4 is a fully atomic filesystem, keep in mind that these performance numbers are with every FS operation performed as a fully atomic transaction. We are the first to make that performance effective to do. Look for a user space transactions interface to come out soon....
Finally, remember that reiser4 is more space efficient than V3, the df measurements are there for looking at....;-) "
Those 2 women in the richest 10 are the wife and daugter of Sam Walton, founder of WalMart. They didn't get rich on their own - they inherited 1/5 of daddy's money when he died. The family (including these 2 women) has no control over the store's operation. They merely own 38% of the stock.
/ bi ll04land.html
This fact doesn't validate or invalidate your point, but it's worth knowing.
http://www.forbes.com/maserati/billionaires2004
Yeah, just to re-iterate, Glade works great on Windows.
http://gladewin32.sourceforge.net/
GTK# ships with Glade#, so yes, there is a libglade for Mono.
*) Gaim does ship with my theme. As does dropline GTK. As does the Gimp's GTK.
. aq ua.buttons.jpg
? cid=11
*) The GTK 1.2 port for OSX does suck visually, but you mentioned "portability", so I thought I'd throw that in just for kicks. I wouldn't suggest using it in a production setting. They are, however, making some progress using the OSX Appearances Manager API:
http://gtk-osx.sourceforge.net/pix/gtktestthing
*) The QT3 OSX theme also uses the Appearances Manager APIs. It's not perfect, but I don't think it's as bad as you think it is. Also, doing your evaluation on KDE's "Ranger Rick Ported" software probably isn't a good test. Adobe's is probably a better benchmark. It will also be interesting to see what comes of this: http://www.trolltech.com/campaign/maccontest.html
XUL, XPCOM, NSPR, NSS, etc... are promising - XUL especially so when using the Firefox toolkit. I'd love to use them in my OSS and commercial apps. Some things (NSPR, NSS) are available as separate libraries right now. But many things essential to 3rd-party developers are progressing at a snail's pace, making the Mozilla "platform" anything but one.
1) Get GTK 2.2.4-2 from http://dropline.net/gtk/. It includes my native Win32 theme.
3 ) Uninstall any old GTK, Gaim, and Gimps, just to be safe.
2) Get Gimp 2.0 from http://www2.arnes.si/~sopjsimo/gimp/stable.html
4) Install GTK 2.2.4-2
5) Open regedit
6) Change HKEY_CURRENT_USER/Software/GTK/2.0/Version to read "2.2.4-20040124"
7) Change HKEY_LOCAL_MACHINE/Software/GTK/2.0/Version to read "2.2.4-20040124"
8) Install the Gimp
9) Enjoy
"GTK is highly portable, but it is however horrible when it comes to any platform but linux. It runs in windows, without a windows theme to match. it runs in os x, but only via x11."
BS. I guess that I didn't write the Gtk-wimp theme engine, then. And I guess I'm not running Gimp, Gaim, Xchat, Dia, Inscape, and Workrave on Win32 right now... The gtk-wimp screenshots are out of date, btw, but are good enough to get the general idea across.
GTK 1.2 has also been ported to Apple's Carbon framework. I have plans to do something similar for GTK 2.x.
"Qt's one of the better-off toolkits, but it suffers from gtk's problems too. It's only good on linux. half-assed os x port (at least it's not x11 dependent) and the windows port is totally unmaintained (yay 2.0!)."
Huh? Trolltech gets a lot of money from licensing QT 3.3 on Win32. I've licensed it. It's maintained, and works great. The OSX port isn't as half-assed as you'd like to think. It's not FOSS on Win32, like it is on OSX and *NIX, but that's Trolltech's decision, not mine.
You're right, XUL may be horribly under-appreciated. But realize that's largely Mozilla's own doing. They've been promising a standalone GRE for about 5 years now, and have yet to follow through. Much of the Mozilla platform is meant for eventual public consumption, but it's not ready for that today. When I can 'gcc -o myapp -lMozillaGRE -lMozillaJScript', come tell me.
Definitely. I wonder if he's aware that the latest SuSe 2.4.x kernel has no fewer than 2400 patches, many of which were backported from the 2.6 series...
Hello stoplights, hello tolls. Hello $30/day Manhattan parking lots, and hello to those half-hours wasted circling the street looking for an open spot. Say hello to the pedestrians and bikers, darting out in front of you. Hello traffic jams, honking horns, and cursing, irratic drivers. Hello noxious fumes and single-digit speeds on urban highways.
Say goodbye to reading the newspaper on the way to work. Goodbye to the half-hour nap you took on the train each morning. Goodbye to your stress-free commute.
I posted this on OSNews, but I'll post it again here...
I'm not sure why so many people think that SVG is slow. It doesn't have to be, even without hardware acceleration. I've done tests of librsvg vs. libpng:
Given a SVG image $s. Transform it into a PNG image $p using librsvg, Batik, or something similar. Run "gdk_pixbuf_new_from_file()" on both $p and $s. This will turn $s and $p into identical RGBA images. Time this operation. Lather, rinse, repeat.
Generally, it is no slower (if not faster!) to render $s than $p. This surprised me and quieted many "Vector graphics are too slow for the desktop" pundits.
Of course, once you start using some of the more advanced features (like certain filters), the rendering time is likely to go up. It all depends on what features you use and how you use them.
If this case is not hypothetical, it is a serious matter, and you should do one or more of the following:
1) Contact a lawyer. Have him/her write a Cease and Desist order to $company. Contemplate further legal action against $company for copyright violation. You may also be able to hit them with something akin to "profitting from stolen goods". IANAL.
2) Contact the FSF. They've dealt with these matters before. Bradley Kuhn and Eben Moglen will talk to you over email, phone, and in person. There are good guys and will defend your rights, tooth-and-nail.
I've been through this situation before. Eben Moglen said something like the following to me:
"People always ask me, 'Is the GPL enforcable? After all, it has never been proven in court.' I always respond that it has never gone to court because the infringers don't have a leg to stand on. It is always in their best interest to settle on our terms."
On Planters (tm) Peanuts cans and various jars of peanut butter I've seen:
"Warning: product contains nuts"
There's quite a bit of inter-operability work going on at freedesktop.org. There's a lot of shared specifications and software there. Plus there are software libraries that both DEs use that aren't listed on FDO, like libxml2.
The KDE folks have also worked on some Qt-GTK toolkit inter-operability stuff. See also:
GTK-Qt
Ditto
Glib/Qt main loop integration
amongst others.
Actually, the biggies (Direct3D, DirectDraw, DirectInput, DirectSound) are each about 80% done. Specifically, these APIs are what most game writers use. The reality here is that you can have high game-coverage while having relatively low API-coverage. This is because most games use the same subset of APIs and avoid odd or esoteric features. This ultimately means that great number of DirectX <= 8.1 applications work well under Wine.
r g/
Transgaming is not Wine. Transgaming is a company that has built a business model around WineX - a fork of the Wine project which is not entirely licensed under the LGPL. Contrary to your assertion, Transgaming does occasionally contribute code back upstream to the Wine project and several of their modules are licensed under the LGPL, though DirectX is not one of them.
For a list of games and applications that work well under Wine proper, I suggest checking out:
http://frankscorner.org/
http://appdb.winehq.o
http://www.winehq.com/site?ss=1
Dom
I think that you're mistaken. Wine does support DirectX and DirectSound to some large degree.
http://www.winehq.com/site/status_directx
Quite a few games work well under wine. In fact, a whole company or two is devoted to making DirectX games work under wine.
http://www.transgaming.com/
IBM could do as you suggest. Then again, they have already written their own Java compiler (Jikes), at least one of their own JVMs, their own servlet container (Jakarta), etc...
I'd recommend looking at this page for more info on IBM + Java + OSS.
IBM has already written at least one high-quality JVM implementation which is not OSS because of contracts that IBM has with Sun. Of course, suggesting that IBM work on GCJ and Classpath has some merit in and of itself. But realize that IBM has sunk untold man-hours and dolars into developing its own JVM - resources that they now wish to contribute to the community at-large as OSS. I personally can't blame them if they didn't wish to spend a similar amount of resources on GCJ and Classpath when what they've got works.
Perhaps with this Open Letter IBM is looking for permission to open up the code. Perhaps they are looking to collaborate with Sun to create an even better project. Perhaps this is all just marketing/PR bs. Time will tell.
Dom
Tk is a horrible example of a GUI toolkit tied to a language. Even if you wanted to write using Perl::TK or TK in C, you needed a TCL interpreter installed.
QT and GTK+ have a fabulous assortment of high-quality language bindings. You aren't tied to using C or C++, respectively. I'm familiar with the Python, C++ and C# GTK+ bindings, and can say that they are superb.
http://www.gtk.org/bindings.html
Actually, Bonobo is an optional dependency for both AbiWord and Gnumeric.
Dom Lachowicz, Gnome Office Maintainer