This was asked by Neil, and I think I can answer it. Clearly, the publication is way more important than the actual product, sadly so.
The paper (or letter, comment, correspondence, whatever it is) that he points out is a complete joke, the website is a complete joke, as the most complete page on it is the CV of each member in the project. It seems that in this case there’s no peer review involved, but I bet the “authors” will dutifully used the reference in their CV.
I can say from experience that even with peer review, the actual product is irrelevant, at least in some of the cases. Sometimes software that I published were completely dissected by the reviewers and it was clear that they installed the application and tested. In other cases, usually the rejected ones, the software wasn’t even described in the review. It didn’t mean that we included a good documentation for the end-user, it didn’t mean that we worked on portability or an efficient code.
Also from experience, I can tell many stories of published software (by others) that is broken, doesn’t compile, has no documentation and fails to deliver what is supposed to be delivered.
Sometime to publish software, the authors’ names are the only thing that matters.