[discuss] /1net Steering/Coordination Commitee

Phillip Hallam-Baker hallam at gmail.com
Sat Dec 21 16:00:55 UTC 2013


On Sat, Dec 21, 2013 at 10:26 AM, Stephen Farrell <stephen.farrell at cs.tcd.ie
> wrote:

>
>
> On 12/21/2013 10:37 AM, Roland Perry wrote:
> >
> > Quite a few people have been doing that for some time (myself, on and
> > off, since 1999 for example).
>
> I didn't mean to imply it hadn't been tried. But I think results
> so far are not yet a wild success (in RFC 5218 terms:-). From my
> POV I've not see that much success so far - the modus operandi of
> the IETF for example still seems to be unknown or mysterious to
> most government folks I've met. But then I don't come across
> government folks that often so it could seem better from other
> perspectives. (I'm not saying that the IETF's approach is the
> ideal btw, but it is an important part of the story here.)
>
> I do think that some good-willed work in this area could well
> be fruitful. I also think that folks who'd be good for doing that
> might be at that Brazil meeting.
>
> I've no idea if 1net is supposed to actually help stuff happen,
> or just organise for fun, but in the former case, this would be
> a good example of stuff to try help with (or try again).
>

You chopped the definition of what needed explanation.


There are two separate criteria that are relevant

1) How to get a standard adopted by the IETF
2) How to get the specification adopted by the Internet

Where government thinking almost always fails is that they assume that (1)
leads to (2). IETF approval is neither a necessary nor a sufficient
condition.


What is necessary is that a core constituency of parties needed to to
deploy the specification coordinate to agree a common, interoperable
specification. Proposals that have such a constituency have succeeded
despite having no IETF sanction (e.g. SAML). Proposals that have IETF
sanction but lacked buy in from critical constituencies have failed
(DNSSEC, IPv6)

If I want to write the best spec I don't invite 100 random people whose
qualifications are that they have an opinion and a keyboard to 'help'.

The reason for participating in IETF is that it is one forum where a core
constituency necessary for deployment can be formed. I don't need 100
people to write a spec but I do need buy in from a large number of parties
to be able to deploy. And I don't just need to convince the major players
that the scheme is a good idea, they have to be convinced that there is a
wider constituency.

So if I propose a change to TLS, the Microsoft representative is not just
going to decide whether to support it based on their own assessment, they
are going to look at what other people are saying.


It is like Lloyd's coffee shop, the place where the insurance market grew
up. The market was created by the presence of the brokers. The IETF is
merely the room with the tables and the chairs and the coffee.

So the governments proposing to move IETF decision making into the ITU or
some other forum completely miss the point that it is the engineers who
make the decisions, not the tables, chairs or the coffee.

It is hard to see quite how a takeover could occur. The IETF certainly does
not have a process that would make that possible. It has taken twenty years
to change the document format. But if there was such a move the technical
community would leave the IETF very quickly if they felt that it was no
longer serving their needs.


We have the greatest tool for human collaboration ever invented at our
disposal and we know how to use it.

The governments do not.

-- 
Website: http://hallambaker.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://1net-mail.1net.org/pipermail/discuss/attachments/20131221/dbb2978b/attachment.html>


More information about the discuss mailing list