On 18 Mar 2005, at 12:33, Keith Moore wrote:
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.
For people with operational responsibilities, this (above) presents a problem. But, see below.
Dislcaimer: I didn't attend the whole week; I arrived on Wednesday, and left on Friday, and I obviously can't comment on the quality of the network when I was not there.
Using my powerbook with the b/g apple adapter, the wireless network in the bar and in most of the corridors was good enough to be able to read mail, use jabber and ssh to routers to fix things. This was good. There were times when it dropped out, but they weren't too frequent.
The quality of the wireless network in the meeting rooms that I sat in was much worse, and was frequently not good enough to fix routers. That's largely ok, though; if I'm paying sufficiently little attention in the meeting room that I am tempted to fix routers, then I should quite probably go and sit somewhere else instead.
I think I've seen someone suggest that the terminal room (with tethered ethernet connections) is a reasonable fall-back to the wireless, which is true a lot of the time. It is notably not the case, however, if:
(a) you need to be able to talk to someone on a cellphone about a problem you are trying to fix; I didn't try, but I got the distinct impression that talking into a cellphone for half an hour in the terminal room would make be unpopular with everybody else in there.
(b) you need to be able to make or receive calls over IP. Making hour-long calls to people in different countries gets really expensive when you're forced to fall back to roaming GSM phones (and the quality is frequently bad, too; 1900 and 850MHz GSM coverage in the hotel was patchy in many places where the wifi was tolerable).
I get the impression that most IETF meeting participants don't have operational responsibilities, and if I've got that right, it does not seem reasonable that the requirements of the network should necessarily be elevated to suit the small proportion of people who do.
However, if there was some place that I could have simultaneously made phone calls and got reliable v4 transit (whether wireless or tethered), that would have been a fine and useful thing for me.
Joe
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf