[TLS] Re: 2nd Working Group Last Call for The SSLKEYLOGFILE Formatfor TLS

["Bellebaum, Thomas" <[e-mail]>]

> Some things should not be standardized.

The problem is that there is a longing for such standardization.
And not just at the IETF. The idea of the draft is, after all, to republish a "de facto standard" already implemented by many applications and debugging helpers alike. Someone had the need for such a format, implemented it, and others followed from network effects.

This creates an ecosystem, and to be honest: If I was writing a new application for TLS right now, I would probably at least think about how I could leverage the existing debugging tools, which work with this file format.

I am still in favor of *some* standardization. Not despite, but because of the enormous risks involved in such a standardization.
Doing so at the IETF grants us the leverage to bring in the voices of many and to shape a de facto emerging standard and how it will be applied, IF we apply enough foresight.
Not doing so at the IETF leaves the important details to those implementing applications, and I am not convinced that their interests and focus match mine...

I am of the opinion that the current draft however misses this opportunity to shape the way it is applied to applications, and the unintended effects it may have on users.
A mere republishing of a file format is little more than a big +1 on just the considerations of the existing implementations.

- TBB