(I think what Warren, Randy, and others have to say is more
relevant to most of this than my opinion - unless you count a
handful of end networks with VPN connections among a subset of
them, I haven't had either ops responsibility or even direct or
indirect management responsibility for those who do for a very
long time. I haven't been to a NOG meeting for even longer.
And I have absolutely no delusion that I'm current. So the
following comments are observations about general principles
only.)
--On Friday, May 31, 2013 02:19 +0100 Adrian Farrel
<adrian(_at_)olddog(_dot_)co(_dot_)uk> wrote:
Hi,
This thread is helpful to me.
...
ultimately call the IETF's legitimacy and long-term future
into question. As you suggest, we may have good vendor
participation but the operators are ultimately the folks who
pay the vendor's bills.
I agree so far.
But who pays the operators' bills, and do we need to encourage
participation at that level as well?
The customers of the ISPs. The technical sophistication of
those folks and their (non-ISP, non-operator) support structure
differs hugely, but most of them neither speak network nor have
any desire to learn. That is not necessarily bad.
Let me suggest an analogy. The electrical distribution industry
has its own standards and standards development processes. In
most countries, those standards body attract engineers and
designers, but very few of the actual operators who run the
power plants and worry about the mains. You presumably pay
utility bills (directly or indirectly) and thereby support the
whole process. But your ability or mine to plug something into
a mains socket and take advantage of the stuff that comes out
does not, in general, qualify us to participate in the process
of defining transmission standards. And that standards body
should probably not lose much sleep over the fact that we aren't
showing up. Indeed, if we did show up, we would probably be out
of place, wouldn't speak the technical language, and might not
understand the culture. At least in my case, if I made a
suggestion, it would probably be stupid enough that people would
either feel that my presence was wasting their time or would
feel obligated to try to explain basic principles to me in the
interest of outreach (or both).
There are certainly engineers with protocol design background
who have ended up with operations responsibilities and acquired
a good deal of additional perspective as a result. Several of
them have made significant contributions to the IETF and, at
least IMO, that dual perspective has helped. But I'm pretty
sure that the majority of ops people don't have that background.
For those who don't, who needs to know that the equipment
and configurations they use will work but who may not have
strong preferences about the details of how, and, when things
don't work, are a lot more likely to blame vendors (or other
ISPs) than us, maybe "how do we get more of them to come to the
IETF and participate in the ways that those of us on this list
participate" is just not the right question.
best,
john