I think it sucks when this means they didn't get the right input into product features. You couldn't have developed to .NET framework without being open from the beginning. MS are a platform company, and platforms need to be built with input from people that use them.
XBOX is a closed platform (for the most part) - so you can get away with it. Zune is not a platform - although I would argue some of the features dropped in v.2 were a victim of not enough feedback.

The question is - does the big bang surprise release equal bigger sales ?

These are good comments though I am not sure what the right answer is. You look at features that really impact IT and users (not necessarily developers) and introducing WinFS (et al.) may have been a mistake. The feature was needed, but they still didn't have a implementation solution. So in that case keeping it under wraps would have been the better approach. Fail in private, succeed in public?

Tough position for MSFT, the 'competition' outside the desktop and corporate IT worlds is nearly all open-source where the expectations are generally so low that gestation and maturity are allowed to develop organically. A project can start with little more than an idea and college spirit, and reinvent itself into a useful product well outside the intent and scope it started with. Its like failure is viewed as a chance to learn and improve, a completely positive aspect. The surviving companies and surviving features seem to be the only lingering memories, pain from setbacks is forgotten.

I think the real negative with this is that MSFT is put into a position where they are cutoff from that growth period for anyone from an aspiring programmer to a small business managing their people to a simple ecommerce website. Its like they are in a position similar to the movie industry that typically only films with high-production values on 35mm rolls with cranes and other special gear instead of a handheld video camera where again the expectations are much lower and success easier to attain if something truly is a great idea.

Microsoft has great platforms, initiatives, and standards, but they aren't always in touch with current events or fast enough for today. I think a real solution would be to have an incubator company, like MbetaS or something more creative, that could freely do whatever the feedback asked for in a community-sourced way. Its mission would be as simple as to build something that some strongly opinionated and motivated people wanted to build, and either continue to advance it or let it fade away. The expectations would be low, and failures would occur frequently. It would allow the Microsoft name to remain a clean standard, while also fulfilling that human exploration and curiousity need without tarnishing anything. I think it would be cleaner than making Microsoft Research more public-facing and provide an outlet to have a free v1.0 that everyone could try and then turn into a viable commercial product.

Instead of fail in public, it would be more like grow in public view. Success could be as simple as converting products that are succeeding in the Darwinistic world and making them a 'shrink-wrap' deliverable product.

Personally, then I'll finally get a media player that can do playlists like winamp with a simple file format (or backup/restore), will keep an organized inventory of my library, will sync the lists to my Sansa and completely rid me of any ipod/itunes envy. :-) Or give me a Zune SDK so if I go to the store and if I don't like something about it, I know I can go home and make it work the way I want it to work. Alpha or beta is plenty good enough, its not mission critical, flexibility beats reliability.

I think it could scale to the point of creating an upgradeable version of things like SalesForce that could be a transition product to allow a company to grow while subscribing to a hosted solution and then transition with a Gold Partner into a better solution hosted on an internal Windows server later on and so forth. Because from my perspective, even Windows SBS edition is too big a leap and leaves a big market well underserved. Man is this ever a tall soapbox.....