Thursday, March 16, 2006

WS-Eventing, WS-Transfer and WS-Enumeration to W3C

Whilst it is good to see WS-Eventing, WS-Transfer and WS-Enumeration finally go to a standards body, it not necessarily a good thing. As Paul points out on Savas' blog, these specifications have been used in yet another political game of chess, in the same way as WS-ReliableMessaging/WS-ReliableExchange or WS-CAF/WS-TX. One good thing is that this time IBM are on the receiving end, despite what they may say.

Ignoring the technical merits (or lack thereof) of these specifications, one thing that simply cannot be ignored is the fact that this will now cause more fractures within the WS-* architecture, and not less. Two competing specifications, within two different (and often competing) standards bodies, will mean yet more FUD and delay as far as vendors and customers are concerned. At least when these current specifications were purely vendor specific, customers who wanted a standard approach to events (for example), had only one choice. Now they don't have any: they've got to wait until something is sorted out.

Now don't get me wrong: I'm not disagreeing that these things shouldn't be in a standards body. Quite the contrary: I believe strongly that they should be. I just wish that the likes of IBM and Microsoft could have gotten together 2 years ago and sorted this out, so we had a single standard now! I think that the world of Web Services can learn a lot from the OMG in terms of how to work collaboratively. It's not perfect by any means, but it's inclusive, which the world of Web Services standards simply is not. There's a lot of rubberstamping going on.

2 comments:

Mark Little said...

http://www.tbray.org/ongoing/When/200x/2006/03/24/WS-Evolving

Mark Little said...

http://h20325.www2.hp.com/blogs/vambenepe/archive/2006/03/24/827.html