First off, this article is not designed to bash anyone, especially WPC. I was motivated to post this entry to identify what could be the start of a disturbing trend – the concept of “open source” XML schemas (complete with an open source license).
Licensing XML schemas is nothing new – many products and standards organizations have already taken this approach (albeit in an open manner to encourage mass adoption). Thinking of XML schemas as “open source”, however, is quite different (and not necessarily in a good way).
XML is not code.
Go back and re-read that line until reality sets in. Go ahead, I’ll wait.
XML is nothing more than text. Source code is also nothing more than text, however source code in commercial products tends to be distributed in a compiled format (since most users would rather use an application to be productive instead of spending time tweaking it). XML schemas cannot be distributed in a “compiled” (DOM?) manner simply because XML schemas are nothing more than text.
Most people (including yours truly) think of open source as “compilable” source code (C#, Java, Ruby, COBOL, whatever). After all, the word “source” in open source is short for “source code”, right? Not anymore, apparently. Today the Washington Publishing Company announced the release of open source HIPAA transaction schemas complete with one of those wacky GNU licenses.
If this concept catches on I can claim this to be an “open source” blog entry (good thing for you I’m using a royalty free licensing model).
I fear this may be the beginning of a disturbing trend. Let’s hope that I’m wrong about this.
What does a closed source XML schema looks like? How would you use it? Are you aware of any other “open source” XML schemas?