Re: [rfc-i] draft-kuehlewind-update-tag/

On Mar 26, 2020, at 1:25 PM, Eric Rescorla <ekr@rtfm.com> wrote:
> As I said in my initial mail, the first rule of holes is "stop digging”.

Again not really an actionable comment, though, unless the first rule of code optimization is to never optimize.

What all of these “rules” mean is that you don’t have solid data to back up your opinion—it’s just your opinion. There’s nothing wrong with that, and you might be right, but you can’t end a debate by expressing an opinion.

What might be a more actionable rule would be “don’t optimize until you need to,” which I think we have all taken to heart by this point in our careers.  I think you think we don’t need to optimize here; clearly the authors think we do.

A way forward is probably not going to involve someone convincing everyone else to have the same opinion.   Can we maybe do an exercise of analyzing some set of past RFCs to see how we think this document would apply to them?  That might help us to collect more data, so that we don’t just have to rely on opinion.