There's an excellent article at DrunkenBlog showing a damning pile of evidence that the commercial product CherryOS contains stolen code from the PearPC project and a number of other open-source projects. (Ed. note: Broken links in the previous sentence have been removed.) What's more, their marketing claims appear to differ so markedly from the reality of their product that they could be subject to a suit from their customers. Here's the background:
The performance issue is fairly simple -- the CherryOS folks claim their software runs at a certain speed, just about everyone who knows anything about the subject says it's impossible for their software to run that fast, and CherryOS has not yet proven their claims. I don't know how many $50 copies of their software have been sold, but so far not one customer has come forward to say his software works as advertised despite the controversy.
The "stolen code" issue needs a little more background. We'll be discussing the differences between two different types of software today -- open and closed source. Closed-source programs, also known as proprietary or commercial programs, are the type of software you're probably used to if you own a Windows or Macintosh computer. A single company writes it, hiring programmers, renting an office, and owning the code. They do not release the source code to the public -- it's considered a trade secret -- and they charge you money if you want to use their software. Think programs like Quicken, Microsoft Office, Windows, etc.
Open source software is different. It's written by groups of programmers spread all over the world, who communicate electronically and collaborate to build software projects that they let people use. Anyone can look at the source code, anyone can take and modify the source code. The only restriction the authors insist on is that if you make any changes to the code, you must release your changes to the public as well. (This is grossly oversimplified.) That way everybody benefits -- the original authors get improvements to programs that they use, without having to write them -- and everyone else gets the benefit of a better product as well. This requirement is part of a software license -- almost all programs have a license.
Programs like Microsoft word have licenses that say you can not make copies of the software and give them to other people -- and Open Source software has licenses that say "You can give this to other people or make changes to it as long as you do these other things that we want." Either type of license is enforceable in court. If you buy a copy of Microsoft Office and then give a copy to a friend, Microsoft could sue you in court for damages. Similarly, if a commercial company takes open source code and includes it in a commercial product without giving away their modifications, they can be sued by the original source code's authors.
The problem is that it's generally not big corporations releasing open source programs. It's individuals doing this in the evening from their dens or garages or bedrooms. They don't have the resources to sue another company.
At least, that's apparently what the makers of Cherry OS thought. If you read the DrunkenBlog article above you'll find a damning technical explanation (quite readable even if you're not that technical) of the similarities between Cherry OS and a number of open-source projects. The makers of Cherry OS continue to deny that any code was stolen, and maintain their accusers are mistaken. Based on the mounting evidence, Cherry OS has a lot of explaining to do. I hope they get to do it in court.