CNet open source commentator not knowing history

Don’t usually bother commenting in my blog about wrong statements made in the mainstream trade press. But this being a slow period between Christmas and New Years I figured I point out why I think Matt Asay seems to have no clue in this blog post. The quote in his article that made my facepalm was this one:

The "right" model seems to hearken back to an experiment Trolltech made at the end of last millennium. Trolltech used the GPL license for its code, but added a clause requiring commercial users to pay. This, of course, wasn’t open source according to the OSI Definition, and Trolltech took heat for its position. But eight years later this is essentially the model followed by SugarCRM, Zimbra, MySQL, and others (including Red Hat, if you look closely at its business), though each company uses different means to get to this result.

First of all Trolltech did not require commercial users to pay, cause if they had done that then they had been in direct violation of the GPL. What Troll Tech did was offer the code under more standard commercial license in addition to the GPL, a license for which you had to pay. So only if you found the GPL unsuitable for your needs the option to pay to get another license came into play. That was absolutely open source according to the OSI definition and I don’t think I ever heard anyone claiming otherwise. As for Trolltech taking heat for its position, I can’t say I remember that happening either, although there was/is of course a lot of debate about using Qt for core parts of open source infrastructure due to it only being available under the GPL. To the degree that debate could be considered taking heat is was if anything some of the projects using Qt taking it (most notably KDE), not Troll Tech themselves, unless one is of the (weird) opinion that considering a library unsuitable/non-optimal for certain things due to its licensing its giving the library developer ‘heat’.

I can only assume that Asay confuses the issue and is thinking about Trolltech’s license precending their adoption of the GPL, which was a quasi open source license allowing cost free open source development on top of Qt, but which was not GPL compatible or OSI approved in any way or form.


#1 Alex on 12.29.08 at 17:21

The original author may also be mixing elements of the old QPL situation in there.

#2 Joe Buck on 12.29.08 at 17:37

Yes, it’s a reference to the old QPL situation, where TrollTech said that people using the GPL could use their code, which KDE did, despite the fact that combining QPL and GPL code violates the GPL. So Matt Asay is closer to right about the history than Christian is, even if he botches it.

#3 ozamosi on 12.29.08 at 18:18

I love how two people have commented this blog post (that are approved), and neither of them have read the final paragraph…

#4 Chris Hubick on 12.29.08 at 19:16

Matt is from Alfresco, which we use at work, and from my experience, they play a lot of word games and don’t really Get It when it comes to Free Software and community.

For all his trumpeting of Open Source and Alfresco, you can’t get the source for stable point releases of Alfresco (only trunk). You can see plenty of their customers coming to this realization in the Alfresco Licensing forum:

#5 stormy on 12.29.08 at 22:05

You can charge for GPL software, However, it’s really hard to get anyone to pay you for the software alone!

#6 Christian on 12.30.08 at 12:58

@Alex, Joe: I felt my last paragraph covered the pre-GPL story of Qt :)

@stormy: no disagreement, but it is still a highly inaccurate statement from someone portending to be the open source expert at CNet

#7 M. Bolton on 12.30.08 at 20:59

Asay gets more attention than he deserves. He’s part of the wave of business hangers-on that showed up in the wake of Free Software’s success in the mainstream.

Also, I happened to notice that his blog is called “The Øpen Røad”. I wonder if he’s been bit by a møøse?

#8 Daeng Bo on 01.01.09 at 07:49

“Until version 1.45, source code for Qt was released under the FreeQt license — which was viewed as not compliant with the open source principle by the Open Source Initiative and the free software definition by Free Software Foundation, because while the source was available it did not allow the redistribution of modified versions.”