Devin Mullins

(The other half is #112 for SXGs.)

On second thought, clients should probably not transparently handle conditional headers for SXGs. As implemented now, the packager will happily tell them to keep using an expired SXG as long...

Note the cache would need only store the SXG prologues (everything that goes into the [signed message](https://wicg.github.io/webpackage/draft-yasskin-http-origin-signed-responses.html#signature-validity)); needn't be a full content cache.

Maybe better yet, base it on the OCSP response (checksum? timestamp?).

I'm not sure what problem this was originally trying to solve, to know if it's worth doing the hacky thing above. The less hacky thing could be useful in the...

Oops, I missed this notification. Agreed, seems like a good idea. Any suggestions for metrics to expose? Off the top of my head: - \# of requests - failures (by...

Sorry for the delay on this. It's still top-of-mind, but life. Wanted to document a few other ideas: - time until cert expires - time until ocsp expires - ACME...

Yes, this is what I had in mind! I'm guessing for the initial implementation of this feature, it wouldn't have ACME support. @banaag is landing #349 which adds ACME support,...

Note that a fix is in #374, pending my review. Just got back from a 2-week trip to Singapore/Bali; pardon the delay.

Also note that a workaround, in the interim, is to run different amppkg instances with different configs.