On 9/20/23 17:04, Brian E Carpenter wrote:
This isn't a new topic, in fact it goes back to about 1993 when Vint
Cerf arranged formal liaison between the Internet Society (on behalf
of the IETF) and several ISO committees. For the formalities, see
https://www.ietf.org/about/liaisons/
Of course, real liaison isn't achieved by formal relationships but by
overlapping participants, and that isn't something the IETF can
control - there's a pretty large amount of luck involved.
One thing I realised when thinking about it is that the choice of
topics and WGs in the IETF is very much a bottom-up process: it's rare
that the initiative for a BOF or a new WG comes from the IESG or the
IAB. So there is no systematic management of overlaps with other SDOs.
It's usually more of an afterthought during the chartering process.
In terms of scope, it's always been pretty clear that the IETF is
"above the hardware" (the main challenge being that MPLS is right on
the boundary) and "up to generic applications" (the main challenge
being the HTTP/HTML and URI boundary).
The Linux Foundation is a special case. The IETF certainly aims to be
O/S-independent, despite the current prevalence of Linux. But getting
code into Linux is important.
There are times that I think the IETF shouldn't be entirely a bottom-up
process. It pains me, for example, to see some of our core protocols
(like email) neglected over time because some combination of (a) nobody
on the "bottom" pushed hard enough to get work started, and/or (b) the
work that the IETF does on existing protocols is too piecemeal because
it's easier to get consensus on small incremental changes than broad
changes (sometimes but maybe not always this is good), and/or (c) people
argue that "industry" should replace standard solutions with proprietary
ones and maybe things will work out anyway (history disproves that, IMO).
But as compared to other standards-making organizations, it's pretty
clear that there's a need for organizations that DON'T do things the way
they do. So imperfect alignment of IETF's organization of areas with
other SDOs' alignment of areas is a feature. If IETF did things their
way, we'd still be trying to communicate with X.400 over X.25.
Keith
