Warner Onstine wrote: > Ahhhh, someone from Sun on the list (hides head in shame ;-).
Apologies for the sneak attack -- no way anyone could know I lurk here. And, we all find Dilbert funny for a reason -- companies and organizations do sometimes do things that at least look silly even if in fact they aren't (though of course, sometimes they are too -- which is why the letter was circulated a few years ago that was ostensibly from Sun telling the island of Java that it had to change its name and which we all found funny.) [No, Sun never wrote such a letter. It did and does write other letters in protecting the Java trademark, which is why the fake letter was funny instead of nonsensical -- there was an underlying truth it played on.] > Yeah, I know too much about Copyright, Trademark and Service Mark than > I should (for not being a lawyer ;-). The legal system forces us all to know more about these things and patents than anyone should have to. > > That said, I don't know of any other (active) binary standard in the > > industry. > > When you say binary, what exactly do you mean here? A standard that covers products, rather than the technologies that comprise products. UNIX, for example (or its relatives, POSIX) is a source standard, describing the standardization of the technology that comprises a product claiming to be a UNIX implementation. The UNIX standard doesn't say anything about the products yielded from it, e.g., Solaris, which is a projection of the UNIX standard into a product space that is unique to the SPARC/Solaris combination (or PA-RISC/HP-UX for HP, RS*/AIX for IBM, etc.) Having a program being "UNIX portable" is not sufficient for there to be a product (i.e., a binary) that runs on all UNIXs. The best you can do is have a set of source code that the development environment for a given UNIX system can transform into a product for that system. Portable technology (source code), not portable products (executables). The Java space of course includes source programming, but most importantly talks about the product realization of it. It's why the term WORA, while not perfectly technically accurate, is nonetheless fairly compelling to people -- one Java .class file does generally run on all things claiming to implement Java. It's why there's 3M+ Java developers in a relatively short period but (probably) less than 1M developers for UNIX 30 years on. Java's really covering two distinct things -- the binary standard (JVM, class files) and also a set of expressions of how those are used in the form of the Java language and evolving set of class definitions. > Just FYI, you can patent a business process - it doesn't have to be a > program, it can be an idea of how to do something that no one (and this > means no one who has patented it before) has thought of before. They > have a prior art rule, however I haven't seen this enforced greatly at > the USPTO. You're right, and in theory the JCP contains elements that might have been patentable at one time, but many of them descend from earlier atttempts at creating and managing binary standards and so have been long ago publicly disclosed (and thus their eligibility for patent protection has lapsed.) A few of them were unique to the JCP but we'd be more interested in seeing other groups steal and use them rather than keeping them unique. As it happens, Sun was involved in creating some of those elements but although we may trademark everything we don't patent everything :-). --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
