One of the things I haven't seen discussed is how exactly this happened. It just seems incredibly sloppy. Apparently the OpenBSD developer intended to start with a working driver and reimplement it piecewise until he had replaced all the code. But is that kosher? That would strike me as almost the very definition of a "derived work", no?
Now, surely there are people who would disagree* with this analysis and say that it's fine. But shouldn't the developer have asked first? Something like "Is this OK to go into CVS as it is, or should I do development somewhere else?" Did any conversation like ever take place? Why not?
Even if this is an one-time process failure, the general lack of interest in enforement makes one wonder if, perhaps, this isn't the only such license violation in the tree.
* The original AT&T/BSD settlement involved a similar piecewise replacement of the Bell Labs code with "pure" Berkeley code, for example. So maybe this is just a tacit assumption in the BSD world.
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds