http://www.youtube.com/watch?v=3cukFl7gpjE (Crockford takes the stage ~5:09 )
Forking a project isn't a healthy thing to do, unless there's something seriously wrong with the existing community (e.g. XFree86). As far as I can tell:
* The nodejs community is healthy and vibrant.
* There are no systemic problems with contribution.
* Isaac Schultner has his head on straight, has reasonable priorities, and is a pleasant character.
* Much, if not most, of node's development is happening outside Joyent anyway.
* Nothing is stopping Yahoo from contributing to the existing project.
Even if all of the above were false, why does Crockford think the remnants of Yahoo's engineering can do better? Joyent is filled with Sun's top talent, and the engineering division is run by an enlightened alpha engineer.
I cannot help but think this is a case of sour grapes by Crockford, because there just doesn't seem to be a good reason for his stance.
Edit: as an aside, his idea of rewriting everything in node is a bit ridiculous too. I've been a part of several projects to rewrite code, for smaller things than all of Yahoo's online assets, and it sure took longer than one year to do a good job even with excellent engineers. This is a long-term project, not something you do to fix the immediate bleeding. This, and forking node, are very strange things to focus on right now given Yahoo's difficulties.
Or you have goals which are substantially different from the goals of the original project.
Lets not forget that parts of Android are 'forked' from Linux (and were later merged back).