This week Steve talked about the new Microsoft - or Ozziesoft as it's now being coined. Ozziesoft represents a world where products aren't announced and shown off until they are ready to go out of the door - apparently. Steve uses the example of Silverlight as a product that's "in store to buy now" but he's actually wrong. Silverlight is still in beta, and will be for a little while yet.
I can see why he does use this an example though. If you think about the way Longhorn was first shown off at the PDC in 2003 it's come a long way since then - it's here that people got to see Windows as a concept. The fact of the matter is there is that Silverlight is a miniscule project in comparison with Windows and you need longer cycles to collect feedback and make changes. This is why it seems like it's released, because it's a very solid beta which has been developed rapidly - but only because it's small relative to an OS.
It's the feedback that is incredibly important as Robert points out and he rightly expresses concern about Microsoft adhering to this new model of releases (if indeed we are moving down that road) of announcing products only when they're ready. I personally don't think it will happen - I think there are valid reasons for postponing PDC - namely we're pretty busy with all the product launches.
It's true we did take a lot of stick about features slipping out of the final Vista product, but it was constructive criticism that you can afford to take on board at that stage. Your shareholders won't care that the IT press gives you a bad rap for not delivering Avalon in the longhorn core, what they will care about is if your end customers, the people who actually buy the software don't like it. That's why Microsoft need the beta cycles and the feedback from the likes of Robert and Long so they can make the changes to Network and Sharing Center (as an example).
I guess you can drawn comparisons between this and the car industry where manufacturers build concept cars that have no resemblance whatsoever to their next version of car. In that respect I guess you could say that Microsoft does a better job of realising it's concepts? Perhaps.
The bad press you get for showing conceptual stuff and then changing plans pales in insignificance to the bad press you would get for delivering a shoddy product.
For all the razmataz and pop star level icon status of Jobs and his announcements, I would give it all up for happy customers. As IT pros and enthusiast we are of course interested in what Jobs has to say, but does the average person on the street? That's the acid test.