But the lack of an .exe exporter means it's a no for me.
I don't think you should put a vote against the licensing model because you don't like the features - they're different things. If we did an EXE exporter, that would not change the proposed licensing model! So the question is, is it a good licensing model?
We've been over the EXE exporter in other threads - we knew we'd disappoint many people with HTML5, but we took a risk and did it anyway, and we still want to eventually produce an EXE runtime. So hopefully everyone is happy in the end!
The fact that you said that c2 is going to take up to 2 years to catch construct 0.x.
0.x was developed very irregularly, part time, in odd evenings here and there. If we go full-time, I think we could catch up within 6 months - but that's not a guarantee (I probably said 2 years to avoid disappointment, or because we weren't planning to go fulltime yet). Feel free to wait until you think it's mature enough to buy a license, but then you might end up paying more (because the early adopter licenses are significantly marked down).
I think being a games publisher or "the youtube of games" is way too crowded a market. There are a ridiculous number of other game arcades out there, some even running on HTML5 now. What's so special about a Scirra arcade? I really can't see why people would come to our site over an established arcade, so I don't think that's the way to go for us. Besides, we want to attract game creators, not just game players. We're about the program.
lucid: as the original post says, the HTML5 exporter and all its plugins will be open source. Third party exporters and plugins will be allowed, either free or commercial, regardless of whether or not they're on an official "exchange".