Microsoft Atlas - should I be worried?
For some time, I've been following the Atlas framework (now ASP.NET Ajax) with great interest. The ASP.NET team has proven to be one of the most agile at Microsoft lately, and with Atlas it seemed that they were continuing that trend. Regular CTP drops were being released, the open source-y goodness of the Atlas Control Toolkit was announced, and things seemed to be moving along well.
Since the release of the July CTP, however, things seem to have taken an ominous turn. No further CTPs have been released since July. The ASP.NET has been conspicuously absent from the Atlas forums, nor have they been doing much blogging. The bug reports having been piling up, with no fixes in sight.
In fact, for a long time, there was no clear answer on how to report bugs, period. The community even resorted to tracking their own unofficial bug list on the ASP.NET forums. The official word finally came down that Connect is the official channel, but there's still no Atlas category to post under. People have complained that bugs reported on Connect have been simply closed as "unreproducible" when they clearly are. I myself reported a bug there and haven't seen any update on it for over a month.
In the meantime, two major bits of news have been announced. One is that the Atlas framework is ungoing a major refactoring. Not much is known about it, except that it shrinks the minimum footprint of the client scripts, that it switches to using prototype-based classes instead of closure-based classes, and that it doesn't include all of the features of the current CTP. There have been hints that lots of the current bugs are fixed in the new version, and that we'll see it some time this month.
The second bit of news is that Microsoft intends to ship Atlas 1.0 by the end of this year. Assuming that means around the end of the month, that means we'll have November, and maybe parts of December if we're lucky, to play with the new bits before the product ships. I don't know about you, but that leaves me a bit uncomfortable.
Certainly everyone (including me) is excited for this technology to be generally available so we can start using it. But, given Microsoft's fanatical approach to backward compatibility, I want it to go out as fully baked as possible.
What do you think?