OSNews

So, are you pro-FOSS or just anti-Nvidia? "

That is a very good question.

If the source code had been available for study (ie open), it would have been possible to examine the differences between the last version that worked and the newer versions that introduced the regression. From that study, it should have been possible to fix the bug far faster than two years.

However, Nvidia have the perfect right to keep their own source code to themselves. I have no problem with that at all.

The question then becomes ... why didn't nvidia examine their own source code and fix the bug? They kept a bug in their code ... for Linux only ... for over two years, and wouldn't agree to fix it or even agree that it existed (despite published benchmarks showing that it did).

That is the primary problem as I see it with closed source from a single supplier. There is no possibility to fix a problem that the vendor won't fix ... or doesn't want fixed.

I can't say if this was the case with nvidia or not with this bug ... but frankly, the circumstances around its history don't make a good case at all for nvidia's good intentions.

This is especially so when you consider that this bug is very good at crippling KDE4 but doesn't have anywhere near as much impact anywhere else, that it appeared just as KDE4 was due out, and that it doesn't exist at all in the Windows driver.

Couple all this with the observation that GNOME is slowly but surely succumbing to dependency on Mono, and KDE4 is potentially huge competition ... and one can almost feel the beginning of a conspiracy theory starting to germinate here.

Departing from the back and forth about NVidia... Conspiracy theorists see evil plots and schemes lurking in their breakfast cereal. That way lies insanity, and I see no reason to go down that road.

Regarding Mono in Gnome, I disagree that is happening. But it would probably be helpful for me to briefly clarify my position on Mono.

In a way, I think of it a lot like I think of Samba. I wish we didn't need it but I'm glad we have it. It helps us integrate into this Windows-oriented world, but I prefer that Unix/Linux admins/devs reserve it for when compatibility is needed, and do not use it unnecessarily. Please don't start new FOSS projects based upon Mono.

I think that as long as we stick to the standardized parts of it, it's probably safe enough. That said, I'm still generally nervous about playing anywhere *near* MS's IP playground. Plus, as a user and administrator, I think that Mono apps suck, and badly.

So... I immediately remove Mono from my machines. This keeps me alert to current Mono dependencies in the distos I use and deploy. And there is not much that needs it. F-Spot? Neither my clients nor I care about photos. Does anyone actually *need* F-Spot? Tomboy? I refuse to allocate 30 or 40 MB or ram to a freakin' sticky note application, and a better, lighter, and faster one comes with gnome. Beagle was beginning to become a problem... until everyone realized how sucky it was and switched to tracker. (I expect other Mono-based apps to fall in the face of superior competition in the future.) That's about it. And with the demise of Beagle, the list is actually getting *shorter*. Note that none of these are core Gnome apps. Now, if you said that Novell and OpenSuse are slowly but surely becoming Mono dependent, you might have a better case. And careless statements like this one from a recent distrowatch review of OpenSuse 11.1 don't help:

"I tried out Banshee, GNOME's default audio player and found it quite usable."

Banshee is *hardly* Gnome's default audio player; It's OpenSuse's default audio player for *their* Gnome desktop. They also use that horrid start menu of theirs.

So basically:

1. I don't think that Gnome is becoming dependent upon Mono.

2. I do think that if it did, quality would be more of a concern than IP issues.

3. I'm not the one to talk to about any wild Gnome/Nvidia conspiracy theories.