Common sense, process, and the nature of change

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



At the plenary last night, Andrew Sullivan set off a series of
conversations at the mic lines by asking what seems at first to be a
fairly simple question:  why is that we seem now to have more process
and less reliance on common sense?  As those at the plenary will have
noticed, the conversation did not stick directly to that question, but
went through some interesting history, called for a bit of data,
entertained a plea for mentoring, and then questioned whether point
solutions or major change would actually work to improve the IETF.

Thinking a bit about the directions that conversation took, I think
there is both a relatively simple answer to Andrew's question and a
much larger piece of context that need to be teased out of the
discussion.  The relatively simple answer is that we don't just use
common sense any more because we don't want to trust individuals as
much as we used to.  That lack of trust isn't directed at the current
IESG, IAOC,  or IAB, but at future incumbents.  We have come to the
idea that allowing a current set of office-holders to make ad hoc
decisions implies that all later incumbents will share that ability.
Since we don't know those later incumbents (how could we?), we don't
trust them; since we don't trust them, we don't want to cede to them a
power that might later get abused.  So we attempt to use structure and
process to restrict those unknown future incumbents.  That's
interesting in part because we believe in precedent enough to worry
that ceding decision making will grant to later officer holders
equivalent power, but we don't believe in it enough to believe it will
guide what the later officer holders will do.  That, again, likely
stems from a lack of trust.

So why don't we trust the future IETF NomCom that picks those folks or
the incumbents when picked?

I think that's where the larger context comes in.  The IETF is not
simply an engineering organization, it is a mission-based
organization.  Our mission is to make the Internet work and grow.
Belief in that mission is something built into the context of the
IETF, and it is part of what helps each of us guide our decisions
here.  Where some SDOs get compromises entirely by horse-trading, many
of the compromises that let the IETF work by rough consensus actually
come about because of that shared mission.  We recognize that
compromise to get interoperability is a key part of what lets the
Internet continue to work and grow.  We both give our technical
insight to that mission and we subordinate our technical desires to
it.

That works in part because we know that others here share that
mission.  We don't recite it at meetings or make it the subject of
ceremonies, but we recognize that spending enough time here will cause
both that mission and the whole-Internet engineering perspective to
seep into the bones of the participants.  We get a direct sense of it
from those with whom we interact (Pete and Murray's exchange last
night was a great precis of that process), we get an indirect sense of
it from broader social networks, and we have institutionalized it in
things like the NomCom eligibility rules.

I suspect that some of the trust issues we have with imagined future
incumbents actually comes from a subconscious fear that we won't be as
successful at passing on a belief in that mission as we have so far
been.  That may be because the current mechanisms are largely ad hoc
(as Joe's comments on mentoring hinted); it may be more free form than
that.  To counter that concern, we may want to extend the methods we
already do have (Edu teams, newcomers socials, and so on) for longer
parts of the initial participation periods.  We may even want to
consider new ways of generating affiliation to our core goals.
However we do it, it seems likely that energy put into making sure
that the IETF's mission is part of each participant's understanding of
their work will return benefits both to the IETF now and when those
unknown future incumbents take office.

The other thing that context suggests is probably equally important.
As a mission-based organization, we have a natural touchstone for
evaluating change.  If a proposed change furthers the mission of the
organization, we can likely manage the transition it implies, whatever
the scale might be.  If it hinders the mission of the organization, it
shouldn't be taken on however cheaply and easily it might be done.

Maintaining the power of the incumbents? Not important.  Maintaining
the current structures? Not important.  Change for change's sake?  Not
valuable.  Making sure the mission gets done?  Pretty much the only
thing that matters.

Since this is all long since baked into the bones of most IETF
participants, this no doubt seems a bit trite or even silly.  But
there are times when it is important to say things out loud, and this
may be one of them. If we are considering why change in the IETF
increasingly looks like ossification and if we are considering how to
fix that, we should keep our mission in mind.

My two cents as an individual,

Ted Hardie


[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]