I find myself thinking that these are the most important things for
wireless:
1. advertise up front that wireless is an experimental, not production,
service.
2. make the wired network production quality. make sure that there are
wires to all essential services (e.g. audiocast, chairs, presenter,
jabber scribe), and make sure those wires are reserved for their
intended uses. if the wireless network works well, so much the
better.
3. NOC team really needs a visible bidirectional communications path to
the users that doesn't rely on the network. I'm thinking in terms of
a live person and a large white board in a highly visible location -
say, near the message board. The white board should report what the
NOC team thinks is the status of the network, summarize what is
reliably known about the nature of the problems, and list known
workarounds. The person needs to write down reports of problems and
interact with users to try to precisely identify the conditions that
produced the failure.
The reason that this is needed is that rumor and exaggeration and
idle speculation on the part of IETF attendees seem at least as
damaging to the repuation of the network (and the host) as any
equipment problems. Such rumor and exaggeration and idle speculation
may even have an adverse operational impact, as users try to route
around their perceived (and sometimes incorrect) perceptions of damage.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf