How have the bewildering array of WS-* specs been faring as of late? Does anybody care?
Steve Jones cares a lot, and didn't mince any words in his latest critique of the WS-splats, separating the good from the bad and ugly.
Steve said that some basic fundamentals, such as WSDL, WS-Addressing, and WS-Policy are all good ideas that are shown to be working within Web services/SOA implementations.
He also has kind words for WS-BPEL 2.0, which has been knocked around as of late, noting that the spec just needs some human workflow. (The debate over BPEL was covered here a couple of months back.)
Among the ideas that could break either way (good and dumb) are:
Semantic Web Services may not be deployable on a practical level, Steve said. Decent front-end tools may may these level of services a reality.
UDDI "never realised its grand vision," Steve points out. "Sure, its still going inside some decent products but it clearly isn't the standard it was cracked up to be."
WSRF - Resource management threatens to make "resource state into some complex beast."
Steve considers the following to be DOA, or Dead on Arrival:
WS-Choreography "sounded so good, but just doesn't seem to have the weight behind it."Web Service Quality Model also "sounded good... but has it gone anywhere?"
WS-Reliability et al sas "killed off by the better WS-RM standards."
WS-Contract and WS-SLA are both "missing in action."
And finally, Steve labels these WS-splats as "dangerous Ideas":
WS-TX introduces two-phase commit via Web services, and as a result, "people will expose this via the Internet and then wonder why they are having massive performance problems. If something is so closely bound in that it needs two-phase commit, then build it tightly bound, not using Web services."
WS-BusinessActivity "shifts logic around into the ether."