« Design Principles: guide rails for service design | Main | Application Platform Strategies on tour: SOA and Cloud Symposium: 22-23 October, Rotterdam »

September 16, 2009




I think REST is here also because WS-* has been a total mess. Now somebody from a vendor want to ride the hype again...
First Corba, then J2EE and finally WS-*.

The message is "you don't need to take architectural design decision, just buy our stuff and it will solve your problems". SOA took a bad name exactly because of this kind of messages, now is it the turn of REST?

Mark Little

Anne, I think you're missing the point or it's not being explained sufficiently well.

Hopefully the following will help:




Mark Little

BTW, as I keep telling you over the years, please don't equate transactions with 2PC :-) If you read the non-2PC protocol that Bill references you'll see what I mean. And as Bill keeps saying (and hopefully he'll post here too) these are simply starting points for discussion. We're hoping you and others will join the community effort, even if it's just to document agreed best practices and guidelines that others in the community can use.

Bill Burke

Anne, never said Atom wasn't restful, I just don't think it is appropriate for messaging especially for machine-to-machine communications. HTTP is a fine enough message format.

Also, we're committed to implementing this in an open source way. I think we have a great track record of innovating in the open source community and bringing these innovations to a standards body. REST-* is going to be run like an open source project.

Also, I'm glad people are skeptical. My hope is they keep us honest and keep us on track.

Bill Burke

Also, didn't reject RESTMS, I just didn't like their requirement for a envelope format.

The comments to this entry are closed.

  • Burton Group Free Resources Stay Connected Stay Connected Stay Connected Stay Connected

Blog powered by Typepad