The most important things I pointed out in my previous review have been at least mentioned in the new draft - the need to look at complete ecosystems (new paragraph in the introduction), better description of re-signing (section 4) and the use of a server as a signature generator (7.3.7).

The discussion following the review also indicated a strong community interest in deploying this technology.

If my misgivings about the approach (which I still have) are wrong, this is a good standards track document.
If my misgivings are correct, this will show itself in deployment.

I don't think we should block this going on the standards track in its present form.