Before the last IETF meeting I asked some IETFers in my company about the rationale for having the final IETF schedule so close to the meeting date that one can't necessarily get a reduced-fare ticket for the days one would want to be there. I heard back that this is indeed a dead horse, so I guess I'm beating it publicly this time....
Perhaps a while back it was a given that people could afford the time and money to attend IETF for the entire week, and therefore the specific agenda is not so important. And various people say how important the cross-fertilization is, and how terrible it would be if people went to the IETF meeting just for a WG meeting or two and then went home. I've heard this viewpoint before, and I sympathize, but I think the whole scale of the IETF has shifted dramatically, and the IETF should be pragmatic. When there were 500 people attending, and they could all come for a week and know everything that's going on, that was fine. But the economic realities can impinge on a company's ability to send dozens of people across an ocean (or even halfway across a continent) for a week at a time (though I do grant that if you do it for a week, at least you can get the cheap airfare :), and the people who participate have ever-increasing other demands on their time. Has anyone done a study to get an idea of what fraction of attendees currently stay for what fraction of time? This might shed some light on the subject. It's one thing to decree that it's a good idea, and it's another thing to recognize that in practice maybe that's not the way it works anymore... BTW, I also heard that WG/BOF chairs would kvetch if they had to ask for a slot earlier. I can't buy this whatsoever -- if the whole schedule were known well in advance to start and end two weeks earlier, where's the pain? Fred
