Nicholas Hurley

Hrm, so for these, I think I would rather see some logging happen, as well, similar to how we do other errors that we forward on through other objects: ```...

I'm not so sure about that. It may not be as simple as I was originally thinking, but I think it's doable. When I was looking at this, I wasn't...

Yeah, I think that seems like a reasonable way to go about it. Once draft14 lands (which should be as soon as Martin has the spec revision ready, unless he...

I can't speak for @molnarg , but from my perspective, unencrypted h2 shouldn't exist at all - HTTPS is the future of the web (even for HTTP/1.1). Given that, I...

Not to get too pushy on this (patches are good, and I welcome them!), but the dev/debug tools that don't support TLS are, frankly, going to have to for h2,...

I've locked this because I'm not quite ready to wontfix it. As noted (though as a non-sequitur to the actual purpose of this bug) this isn't the most high-performance server....

@ralph3991 not true, as of v3.2.0, node-http2 advertises h2, not draft 16 (any documents that say otherwise are just out of date - pull requests welcome to fix it!) Also,...

I'm not convinced the breaking API changes need to happen - it's just as easy to add arguments (which will default to undefined) without breaking backwards compat (and the argument...

Hrm... that feels even less desirable, to me - mixed metaphors in a way, with some arguments being explicit/positional, and some being in a (positional) dict. I'm tempted to go...

Sorry for the lag here (the downside of having little time to work on this). Are you talking about one physical host having SNI or Subject Alternate Names in the...