Talk selection procedure

[David Wolever]

Hey guys,
Unless anyone can point to a better way, here's how we're going to be selecting talks:

- The talk review party will be on Tuesday, Oct. 2nd. The party will take place at 7:00 UTC-4, physically at Morgan's[0], virtually at irc://irc.freenode.net/#pyconca-talk-review. Reviewers must bring an internet-enabled device (laptop, phone, etc).

- Talk reviewers will be everyone present — either physically or virtually — for the party.

- Selection is done entirely during the selection party on Oct 2nd (ie, reviewers don't need to plan ahead).
- One person will be assigned the role of "transcriber", who will be responsible for communicating with the people on IRC.
- One person will be assigned the role of "assigner", who will be responsible for assigning talks to reviewers.
- There will be a short discussion of "what constitutes a talk that's “good enough”".
- Each talk will be reviewed by two reviewers to determine if it's "good enough".

- After being assigned a talk, reviewers will read it, privately make the "good enough"/"not good enough" decision, then get another talk from the assigner.

- Efforts will be made to ensure that talk submitters don't know who reviewed their talk.
- Once all talks have been reviewed by two people, the assigner will read out loud the list of talk names and authors. The two reviewers will give either thumbs up ("good enough") or thumbs down ("not good enough"). Talks that are good enough will move on. Talks that are not will not. Talks which may or may not be good enough (one thumbs up, one thumbs down) will be set aside to discuss later.
- At any point any reviewer can flag a talk for further discussion (ex, if they would really like to see it, even though it was rejected). These flags will be handled as appropriate.
- The number of accepted talks will be counted. If there are enough, we're done — those are the talks. If there aren't, we'll break off into smaller groups and discuss, in an ad-hoc manner, the talks which may or may not be good enough, ordering them roughly by preference.
- The top N of these talks will be accepted, where N is the number of talks we need.
- The Board reserves the right to make any and all final decisions (although we hope not to).

Additionally, I would encourage all reviewers to read http://www.doughellmann.com/articles/how-tos/review-conference-proposal/index.html before the review party.