Microsoft's pseudo-open source: open trap for open-source developers?

If you believe some of the headlines, Microsoft just open sourced a bunch of software related to its .Net libraries. Don't be fooled. The definition of open source is very clear. This is not open source. Not even a little bit. In fact, this may actually be an insidious trap (more on that below).

Will Hurley captures the move accurately:

Is .NET open source now?...The license indicates that developers can "see" the source code, but Microsoft's not providing any means of copying it. If a developer finds a bug in the code, rather than fixing it themselves and submitting a patch to the community they'll be encouraged to submit feedback via the product feedback center. They're showing us the man behind the curtain, but we're not allowed to speak to him in person just yet. We're still stuck with the giant, disembodied green head. And since community involvement is essential to most open source efforts, well....

In other words, it's not open source. But is it good for developers, anyway?

Steven J. Vaughan-Nichols of eWeek is not nearly so polite, claiming that Microsoft's pseudo-open source is a real, legal danger to any open-source developer foolhardy enough to look at it:

Let's say a year from now, Microsoft does a SCO. They claim that Mono contains code that was stolen from the .NET Framework reference source code. They point at their code, they point at the license, and sure enough, there's similar code. After all, both projects are implementing .NET; there will almost certainly be lines of code that looks alike.

Better still, from Microsoft's point of view, all they need to do is find one Mono programmer who has signed the license to look at the .NET Framework reference source code. With that "proof," they'll claim they've found their smoking gun. SCO failed in its attempts because it never did have any evidence that there was Unix code in Linux.

Microsoft, however, is baiting its trap for Mono programmers with .NET cheese. They'll claim, come that day, about how open it was in letting people look, but not touch, their code. With the combination of "proof" that some Mono code has been stolen from Microsoft and its attempt to muddy the waters about what open source really means, it can look forward to having a much better chance of killing off an open-source project than SCO ever had with Linux.

Far-fetched? Well, not if you look at Microsoft's history. Microsoft is playing to win, and it seems to believe the only way to win is if open source loses. This is myopic, but the company refuses to get LASIK.

Maybe Microsoft deserves some credit for the move. After all, I'm willing to bet that Bill Hilf and team had to fight hard to get this done. But if there is even the slightest intention of going down the path that Steven lays out, the road to Microsoft patent Hell will indeed be paved with good intentions.

[mattflaschen]

Assume good faith

Make no mistake. This license isn't open source. But I don't think it's a trap either. I also don't see how it's that different from Rotor (http://en.wikipedia.org/wiki/Rotor_%28software_project%29). It's obviously risky to rely on this code when implementing an alternative (even if you try to refrain from copying). But I'd rather see it as a step on Java's path (remember that Java had been available under a read-only license for a long time before it went GPL) than a trap.

[PACSferret]

Perhaps a challenge for version control guys......

Hey Matt.. in your scenario - it will only take one Mono developer (e.g.) to sign the license to give MS a way in. But what if the VC system (SVN in Mono's case, I believe) has a legally watertight (difficult, but not impossible) mechanism to show that that developer hadn't touched the element of code under question? I'm very much a casual SVN user and certainly no legal expert so I'm very much outside my comfort zone here.

In the world of patents say in parmaceuticals, is such non-contestable record keeping a cornerstone of patent activity?

[mattflaschen]

Not really proof

That wouldn't really be proof against a copyright infringement charge. Microsoft could just claim the developer that read the .NET source secretly sent the patch to the developer that committed it. Revision control systems can't do anything about this. Also, patents are a totally separate issue. Mono may already infringe Microsoft patents (/some/ of which are licensed through ECMA). That doesn't require that the code is similar, only that it operates the same way as a patent specifies.

[bloke12]

SCO Failed?

You write that SCO eventually failed. You are wrong. The SCO debacle, as poorly organized and as pathetic it was, was unbelievably successful for the enemies of OSS. SCO, using FUD, turned corporations away from Linux, won ("extorted") money from unknowledgeable users, and was a thorn in the movement for years.