Just my 2 cents:
I'm not necessarily convinced that more time equates to more getting done. My observation was that if you had a 7 day conference, folks would show up for a few days and that you'd end up having "quorum" for only a brief period. The reality is that having 5 days versus 3 days isn't going to mean the difference between a release or not.
while I agree that the focus of our meetings has tended not to cater to end users, i think that's more a result of the audience.
one could also make the argument that we might not be drawing a larger crowd for any of the following reasons (not necessarily inclusive):
- lack of advance notice on dates and locations - length of time: most folks have jobs that prevent them from taking a whole week just for something like this - perception of a too-highly technical crowd
My preference would be for more "meetings" of shorter duration. That way, there are more likely to be "hits" for people's schedules. I understand that getting a large proportion of the core development team is a goal, but that's more a matter of coordination than anything.
It's less the case for people like me, but folks in europe are blessed with inexpensive air travel that makes things like twice yearly, in-person meetings of a day or two possible for a large number of our audience.
Also, I'm not opposed to making the trip to Sweden. I've never been, and I hear it's lovely :)
Bill
the last two conferences were more like 5 days. and i'd like to spend more time actually working on code and a 7.8 release. like an extended international version of that last meeting over kladdkaka.
that's not for everyone of course, but it's not like we are drawing a crowd, so we might as well do things to move pike forward instead of trying to attract a lot of pike users. (not that i am against attracting pike users, only suggesting that the conference might not necesarily be the venue for it)