It’s likely that there will be many implementations … servers pushing out XAML, and web browsers or other client-side systems interpreting it. Some from Microsoft, some from universities, some from other corporations, some from hobbyists.

And if they don’t agree on what something means, then they will do different things in response to the commands.

That variation in behaviour will be exploitable. We have the ‘storm worm’ racing around public Internet at the moment, creating networks which can be used for sending junk mail and worse. Broken or part-functional XAML interpreters will do much more of that.

What’s the resolution ? I’m not sure, but I think ‘open standards for interoperability’ ought to be part of it.

As for XAML, there are several issues here and FOSS users are bound to be affected most badly for a variety of reasons. As for reach, Microsoft may have plans for Yahoo. It’s not as though it has been gentle with established standards… ever in the past.

Regarding XAML I’m a bit more optimistic than you are. I don’t think Microsoft has any chance of succeeding to replace HTML with a proprietary format at a scale that really matters. Some people will use it internally just like they’ve been using Microsoft’s proprietary HTML extensions on intranets where IE is required. But I don’t see this spilling out onto the internet in a significant manner.