I think this thread has been fabulous a lot has come out of it :) As an example I'm still a naysayer, but I drove me to do the performance testing with different settings. Result was that i was able to increase performance by a lot. Also I went about going through different websites and found some interesting facts.
Canvas rending SUCKS. Even Cocos-x on android isn't all that good. While we are annoyed with 20fps issues they are having troubles with 30fps issues with the same sprite circumstances. There are gaming blogs that report game design considerations once you hit more than 30 sprites. With that part of the problems lay in the Tegra 1 and 2 where any transparent rendering has flaws which also drops performance.
What however changes is oddly enough OpenGL 3d based rendering. Strangly enough a full on 3d engine doing 2d can render 250 sprites animated at 30fps on a Tegra 2. I'm not sure why the difference or what's going on. but it is insightful. None of this would come to light if discussions aren't there.
Now to something important which I don't think the HTML5 choice will ever get over. Which is the other problem. Let's remove the entire performance speed issue. let's assume we are in happy land :) heres the problem
anything that is native code access
even the day HTML5 becomes this great at gaming. We can't access Steam Achievements. So there we go. Completly reliant on CoocoonJS which is closed platform at this time. Though apparently CoocoonJS has an extension manager in the works. Till that day :(
I have no idea how to monetize my Ouya game for a sale. I can't access the Ouya store. If I use Clay.io and am allowed to sell the game. this will require users to enter in the pay system by another method and well people don't like that :(.
I'm still a naysayer, but I'm willing to put performance to rest for now. I feel I can pull enough performance out what there is right now for my games. Though that doesn't change other factors :(
Anyways I thank everyone for speaking there minds and having an opportunity to speak my mind. Also Ashley participating and giving us a lot of company making decisions :)
as a final note on that however. I still think Scirra should do a kickstarter to gain funding for a 1 year staff to make the Exporter Development Kit. Once that's done. just let the community do the rest and Scirra can stay hands off towards maintaining. As Ashley pointed out Chrome was a communal developed project from many different code snippets and improvements. The EDK's exporters and plugins could operate under the same standards. Scirra can stay hands off past that entry point. But hey. That's it for me.
I hope many of you have learned as much s I have :)