Mark, > It got worse the last years. Specs, or at least draft specs would be > published publicly without having any click-through license to which > people have to consent. There are also some nice counter examples though > of expert groups doing everything publicly (JSR133 about the memory > model, JSR166 about concurrency util classes).
Even those "open" JSR's ultimately have a click-through license on the final spec. All of Sun's API specifications are protected by such licenses. I'm not a lawyer so I don't know how far a book can go in describing an API before it either needs to re-assert the original license terms, or else finds itself in violation of copyright. Not that I would expect Sun to persue such issues because it is in Java's best interests to be written about. But it does concern me that such unofficial sources would be preferred over the actual specification. Further, given Classpath's goals, I don't see how it could ever claim to be what it is, without requiring compliance with Sun's licenses regarding independent implementations. David Holmes _______________________________________________ Classpath mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/classpath
