Tim Bray’s posting on the Sun Microsoft interop demo is fascinating on many levels. It certainly adds to his reputation as an independent voice that while sympathetic to Sun’s goals is not entirely in the thrall of its message discipline.
He gives some validation to my hypothesis that the customer demand is one of the key drivers here when he writes “there’s no doubt that when the customers tell you to interoperate, then you bloody well interoperate.”
He makes the point I’d expect of a principle of institutionalized standardization to make “WS-Federation is yet another WS-backroom spec that might change (or go away) any time the people in the backroom want it to; not something I’d advise betting on.”
That the WS specs lack legitimacy from the perspective of a professional standards craftsman is absolutely true, and I suspect that customers were demanding that much more than they were asking to interoperate. An interop bridge is a poor substitute.
Getting a stamp of approval from a real standards body isn’t relevant if your solution wins in the marketplace. Microsoft’s installed base on both desktop and browser make that outcome entirely plausible.
Predicting how the market for enterprise identity management will standardize and how many vendors will get a portion of the market share remains a tough call. Both WS and Liberty are such complex specs that they aren’t well positioned to get rapid adoption. That is good for Microsoft and a problem for Liberty. Liberty is available in the market today, with implementations, real legitimacy from a credible standards body, a choice of vendors, and broad industry support. So that’s good for Liberty and a problem for Microsoft.
I don’t know what’s going to happen. Does the recent interop between Sun and Microsoft signal an end to the battle in this area, or does it just buy Microsoft some more time. It certainly does tend to let CIO/CTO decision makers defer deciding.