This post originated as a response to an email in the vwnc mailing list. The premise of that email was that our decision on Widgetry - and the stated justification - implied that there would be virtually no changes to the existing UI framework in VW. That's not true, and I thought it would be worth making a public explanation of the point. So:
The need for customers to make a significant change to their code was a major issue in the decision to cancel Widgetry. It was not the only one though. Widgetry was under development for 6 years before any effort was made to build tools or applications using it. This was a large mistake on our part, and one that won't be made again.
It is possible to make significant improvements to the UI frameworks without a complete migration to something new. You'll start to see improvements fairly quickly. Bear in mind that no real changes have been made to the UI framework for years. During the Parc/Digitalk merger, the VW UI was abandoned, and when the Jigsaw effort failed, the remaining team had no UI members left on it. Thus the UI stagnated.
When the decision to create Widgetry was made, no thought went into making changes to the UI, as "Pollock was coming". All work in that direction was frozen, as we waited. Plenty could have been done, plenty should have been done. Now, plenty will be done.
Now, as to the theory that the existing UI framework is essentially frozen:
Evolution of the existing code base may require changes to customer code, but the key thing is this: it will require relatively small changes over time, instead of a huge change all at once. For an example, look at the NetClients code in VW. I can speak to this one personally, having used the code extensively over the last few years (Silt, BottomFeeder). With each new release since VW 7, I've had to make minor changes to my code. They were all documented though, and all minor. I didn't have to drop all of my HTTP handling code and start over - I had to apply a change over here, a modification over there.
The difference between what would have happened with Widgetry and what is happening now is the difference between revolutionary change and evolutionary change. Change there will be, but it will not be massive and all at once.