Bob Muglia said it at DevCon (and then left to explore other possibilities). For Sinofsky to say the same things 8 months after that, and 5 months after Muglia took a walk, means that the battle within Microsoft has been won, as I see it.

The good news: Microsoft has always (on the 3rd try or more), created great developer tools. If, as the article says, MS is building tools to work in this next UI context, that is very good news.

An interesting thing happened, btw, when MS ditched the dynamic languages last September (same timeframe as DevCon): the lead developer (Dino) for the IronPython IDE in VS said he would do what he could in his spare time at MS. And then ended up writing a new, open source version that IDE that works with Python -- that's non-.Net Python as well as works with IronPython. That's not something doable in "spare time." Something changed between last September (same timeframe as DevCon, when Mr. Muglia sealed his fate) and now.

This is all just reading the tea leaves, of course, but there is a coherence in the way the tea leaves are aligned.

The back-channel word seems to be that .Net will still be used on the backend. But of course anything else can be used on the backend, also. I had to write an http web service receiving and sending JSON messages a couple of months ago. I had never used the package (CherryPy) before: I had it taking and sending messages in 12 lines of code which, due to my inexperience, took me an hour to get working. Part of the time was spent learning that JSON messages map exactly to Python dictionaries: it took me a while to accept that it was that easy. Once the UI is freed from .Net, we'll have a choice as to what to use on the backend. That's a good thing: the best-for-the-purpose language will win, as it should.

Hank

>http://www.theregister.co.uk/2011/06/06/windows_tablets_without_silverlight_dot_net/
>
>Let the rumours begin...
>
>How feverish are things going to get before "September"?