Thanks for your comments, Jordi!
I'm replying to part of your note, and changing the subject line to get different topics on different threads..... I do think we need some kind of IETF consensus on the criteria for venue selection - and once we have that documented consensus, we need to evaluate how well we follow them...
This is not trivial, nor is it easy to get everything in line; for instance, one of our requirements is fairly large (and expensive) offsite Internet bandwidth; one potential US site was able to deliver that bandwidth as part of the conference package - BUT insisted that it be able to perform its usual packet filtering on the traffic, and refused to allow alternate bandwidth provisioning on site.
Not something we expected five years ago. Foretec declined the offer - wisely, in my opinion.
The subject of continent selection is, BTW, one of the real touchy ones; my statistics show that in the North American contingent, the attendance is cut in half when we meet outside North America; in the Asian contingent, the attendance from the host country quadruples or more when we meet in their country, while Europeans show more steady attendance statistics.
Fred Baker formulated a principle of IETF meeting placement as "if you contribute to the IETF, the IETF sometimes holds a meeting near you". Measuring contribution is a difficult thing - it's certainly not the same as attendance! - but I think the principle is not unreasonable.
And after having picked a continent, we need to pick a venue - which can depend on the wishes of the sponsor, cost of doing business, availability of reasonable venues (I have been told by Foretec that the sensible Vancouver venues are booked solid on the dates we want for the next 2 years, for instance). And that sometimes interferes with the selection of continent - for a while, I was told that Foretec was looking at Europe in spring 2005; now I'm told it's looking at summer 2005 - because venues worked better for August.
I think the organizer needs to be able to make these tradeoffs in real time, and without going back to the IETF for a consensus process on individual meetings - but we do need to have our criteria right out in the open.
I would prefer to split the process into two rather independent parts: One (open) that sets the criteria, and one (subcontracted) that attempts to find sites that fulfil the criteria. Then we can evaluate the result - for economics, for venue performance, for sponsor satisfaction (that too matters!), and for "fairness".
As to how to achieve all that.... I'm not at all sure.
A long note about a subject that is a small part of the "reorg" issue...... hope this makes sense to you!
Harald
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf