Re: Is IETF is being shunned for new protocol development?

[Eric Rescorla <[e-mail]>]

You write:


> But, well, if you count the number of actual STANDARDS that the IETF
> has published, as opposed to draft or proposed standards then I
> would argue that it is much harder to argue that the IETF has been
> successful as an SDO. Given that is has only published 102 standards
> over a 40 year period, that means that the IETF only actually
> manages to publish 2-3 standards per year ... Of course, obviously,
> this just really means that the naming of the documents is wrong,
> but why do you think that nobody has succeeded in fixing this over
> the 40 years that the IETF has been publishing RFCs.

I have a somewhat different view of this particular point:
The reason that this hasn't been fixed is that it basically just
doesn't matter. The set of people who understand the nominal
difference between a Proposed Standard and a Standard and don't
understand that there's no practical difference is tiny, and so nobody
has any real incentive to do anything about it. For my part, I'm an
author on at least one document that clearly meets the bar for
Standard (TLS 1.3), and I told the WG chairs that I wouldn't expend
any effort to make draft-ietf-tls-rfc8446bis a Standard, because I
just don't care.

To tie this into your point about modpod, I think there is a
relatively small set of people who are really invested in things the
way they are and resist any change. This is why modpod was so
expensive and why it's also too expensive to change to a one level
standards track [0], so instead the system just works around
it. Modpod is the rare example where it's actually worth real effort
to make a change, and so the organization had to absorb it.

-Ekr

[0] Also why it's so hard for people to admit that IDs don't really
expire, etc.