New plan, based on comments from people and the assumption that the maemo.org gang-of-four will expense Nokia directly rather than coming out of the sponsorship budget. Figures in brackets are expected number from that slot, and total so far:

1. Anyone speaking in a "proper" session (~30/30)
2. The current (at the time) community council (5/35)
3. 5 Mer slots for Stskeeps and anyone else he wants (5/40)
4. 10-20 slots for arbitrary decisions decided by the council: community bloggers, GSoC students, cool Fremantle developments, upstream community members (~15/55)
5. Top 5 current OS app authors (i.e. currently diablo; personally I hope it'll be fremantle by the time of the summit!) on downloads.maemo.org based on downloads & rankings (5/60)
6. Anyone in the top 25 of karma. (~15/75)
7. Top 5 people on tmo on thanks/posts ratio, where posts > 300 (~4/79)
8. Anyone in the top 50 of karma coming from outside of the EU (~15/94).
9. Any previous council members (~7/101).
10. Anyone speaking in two or more lightning sessions (~4/105).
11. Any remaining, based on karma.

So that allocation is looking at about twice as many potential sponsors than last year.

There's a lot of overlap in the groups, hence the lower figures for the expected number of people in each group. For example, someone like yerga might be giving a talk; has a lot of karma; has some highly downloaded apps; may stand in the next council elections and win. That wouldn't mean that we go to the next highest app author automatically.

Two questions:

1. Is the prioritisation roughly right? (ignoring whether it needs to be judgment based or just different numbers)
2. Should it pretty much all be merit based after the first item on that list? Or should things just be jiggled around/slots & expectations changed?