[Chapter-delegates] ISOC and its involvement with the IETF

[Andrew Sullivan]

Hi,

On Mon, Sep 13, 2021 at 08:02:32PM +0300, Hank Nussbacher via Chapter-delegates wrote:

>I recently learned of an IETF standard (RFC8890) entitled "The 
>Internet is for End Users" which was published a year ago:
>https://datatracker.ietf.org/doc/rfc8890/

To add a little to what Bob Hinden said (and this will doubtless seem eye-glazing to many on this list), that RFC is neither and IETF product nor a standard.  It _is_, however, a valuable document, and I'd encourage people to read it.

So, what do I mean by this?  Well, two things.  First, RFCs come in different flavours, and while all standards that are produced by the IETF are found in the RFC series, not all RFCs are standards.  In this case, the document is an Informational document.  This is actually stated in the document, in a bit of boilerplate that essentially nobody ever reads, near the beginning, entitled "Status of this Memo":

    This document is not an Internet Standards Track specification; it is
    published for informational purposes.

But, the second thing is that the document is not even from the IETF.  When an RFC is produced by the IETF, even an Informational RFC, it goes through the IETF Last Call procedure, which is where anyone in the IETF community can make arguments about why the document is wrong, or ought to be changed, or is entirely correct, or whatever.  This document, however, came from the IAB stream of RFCs.  The IAB stream reflects the opinions of the Internet Architecture Board, but not necessarily the opinions of the IETF.

Now, it is important to remember that the IAB's charter gives it two roles.  One is as a committee of the IETF, but the other is as an advisory body to the Internet Society Board of Trustees.  In that sense, we can think of the chapters having the ChAC, the organization members having the OMAC, and the IETF having the IAB, in relation to the ISOC board.

>In section 4.1 "Engaging the Internet Community" I am surprised that 
>ISOC is not mentioned at all as being a representative of all Internet 
>end users.  Why is ISOC not seen as representative of end users 
>worldwide?

I think the document goes out of its way especially to note that there really _isn't_ anyone who is a representative of all Internet end users.  For my part, I don't believe it is possible to collect such representation into a single organization.  What I think is true is that the Internet Society _as a society_ could probably do a better job of capturing and channelling the views of various kinds of users into IETF discussions about technology.  I will note that many (I haven't checked whether all) of the people explicitly acknowledged in the document are Internet Society members, so that is certainly encouraging.

>Looking at the authors of RFC8712 I am curious which ISOC staff or 
>Chapter members were polled or played a part in the writing of 
>RFC8712?   I cannot identify whom from ISOC was involved in the 
>writing on this RFC.

Gonzalo Camarillo (one of the listed authors) was the board Chair at the time.  In effect, that RFC encodes the new relationship between the IETF and the Internet Society that was a direct result of the IASA2 changes that resulted in the IETF LLC coming into existence.  So, it's really descriptive rather than prescriptive (you will note that it is also Informational).  You could certainly see comments that were received on the document in the IASA2 working group; the list archive is at https://mailarchive.ietf.org/arch/browse/iasa20/.  I definitely reviewed this document before it was published and I believed it to be an accurate description of how things worked.

>I fear I see a pattern here of a disconnect between the IETF and ISOC 
>which needs to be rectified by a more active involvement of ISOC and 
>its chapter members in the IETF.

What do you mean by "more active involvement of ISOC…in the IETF"?  It is is worth remembering that 1/3 of the trustees are appointed directly by the IETF through the IAB.  Many other trustees have historically also been active at the IETF.  Moreover, in a formal sense the IETF is a part of the Internet Society: legally, it is a separate corporation (IETF Administration LLC) that is disregarded for tax purposes; so when someone donates money to the IETF, it lives with the LLC but is counted for tax purposes as a donation to the Internet Society.  We also have staff who pay attention to work that goes on in the IETF, but because we provide so much of the IETF funding, and becaise of the formal legal relationships between ISOC and the IETF, we try to avoid having staff act as part of the formal structures of the IETF (it is rare that our staff are WG chairs and staff are now formally prohibited from being IESG or IAB members, or part of the IETF Administration LLC board or the IETF Trust).  We also discourage staff being authors on Internet Drafts that aim to become RFCs except under fairly narrow, special circumstances.  All of that is intended to ensure we do not create any possible perception of "staff capture" of the IETF.  Enthusiastic participation, particularly through high-quality reviews of drafts, _is_ encouraged.

As for ISOC chapter members being actively involved in the IETF, I am the first to encourage it!  In one sense, the barrier to entry is low, of course, because all you need to do is join a mailing list and read some freely-available documents.  In practice, however, the IETF is an incredibly hard club to join.  This is a long-standing problem in the way the IETF functions, but it is certainly not something that I think ISOC-the-corporation should try to "fix", any more than it would be ok for us to try to interfere in the internal operations of chapters beyond the minimum agreed as part of chapter agreements.

Best regards,

A

-- 
Andrew Sullivan
President & CEO, Internet Society
sullivan at isoc.org
+1 416 731 1261