I got a notification from the Web Services Journal about a new comment on the "Why WSDL Is Not Yet Another Object IDL" that Jim and I wrote some time ago. The comment by Leo Golubovsky is really nice.

After googling for a link to the original article, I run across other posts on the subject that I wasn't aware before. In one of them ("WSDL, IDL and CDL"), Mike Taulty refers to our article and quotes one of the sentences that was completely misunderstood (our fault of course). In retrospect, I agree that we shouldn't have included the part about in-memory representations because we didn't really explain our intentions well (if I remember well, I am to blame for that but anyway... what's done is done). However, I think that Mike, as some others, missed the intention of our article which was to suggest that WSDL should not be used as an IDL to describe objects. Yes, you can use IDL to describe services but our thesis has always been that WSDL should not be used as an object IDL.

Perhaps we should have been more explicit about the difference we were identifying in our article.

Hopefully, this is going to be the last entry about that article 🙂