File under ground

WinFS is no more. The wonderful icing on the file-system cake that would unify all Windows data types in one superior, searchable, magical way has been canned. Ten years of promise have been wrapped up with various bits hived off to existing projects — but WinFS as a concept has gone.

There are layers of irony here too deep to mine without breathing apparatus. The biggest problem — and the biggest opportunity — in the world of IT is finding the right data in the right time. Some of the most exciting ideas online are to do with data, metadata and the tools to handle them: when Tim Berners-Lee uses a public forum to ask Frank Mantek of Google to implement a SPARQL interface, take note. This may not be the right answer, but it's answering the right questions.

Unlike WinFS. What was the point of a database-cum-filing system that would not scale, and what was the point of making it scale if it were then to come into direct conflict with SQL Server? And how can you put a sensible user interface on a filing system that relies on user-generated metadata when so many users are barely comfortable with the idea of files in the first place? And who's going to get excited about a closed, proprietary filing system that's going to take hard work and NDAs to plug into anything else?

And then there's the manner of WinFS's passing. Is it dead? "Yes and no", says the WinFS team blog — which, as even the rawest marketing recruit knows, decodes to "Yes, yes, dear God yes". In 2004, it was one of the three pillars of Longhorn. Two years later, it's not even a dirt-track off the road map, but if you take that blog posting at face value it's just a natural part of the product's evolution. "It represents a change to our original delivery strategy, but it's a change that we think that you'll like based on the feedback that we've received."

Microsoft got it wrong. It took people's money on that basis — how much of the Software Assurance pitch from two years ago stands up now? — and now it's saying that all is well. Apart from that, Mrs Lincoln, how did you like the play?

If Microsoft is going to communicate major strategic changes through a marketing filter that tries to minimise their importance, then people will believe the worst, and rightly so. Whatever the solution is to the company's present malaise, it won't find it in the wrong sort of search — or the wrong sort of answer.