Thanks Nick and Florian. Yes, it has been a very dynamic conversation. :)

Our Summit team here at the Foundation thinks traditions are an important part of our community culture, but we hate to do things (like voting) just because that’s what we’ve always done. In making the proposal to end voting, I think we’ve challenged our thinking around it and come up with some really good ideas. I am now personally leaning toward keeping voting this round with some tweaks and better communication around the purpose, with the option to reevaluate after Barcelona.

If possible from a development standpoint, we would support the following ideas in this thread:
- enabling comments to provide track chairs more feedback ahead of their decision and speakers more input as they prepare their talk
- removing static/linkable URLs for promotion to reduce the noise and gaming
- clarifying the purpose of voting as a piece of input for track chairs in the speaker selection process

We're on the fence about displaying each voter a limited, random subset of talks, because some folks only care about specific tracks or topics, and some folks have diligently rated every single talk in the past. We think it would be very difficult and counter to the purpose of community voting if we tried to limit the pool of voters to speakers or conference attendees (the latter simply because a very small percentage are actually registered at the point in time we start voting). 

Additionally, we would support publishing the names of track chairs earlier in the process with clear guidance not to contact or badger them about your talk.

On a related note, as we continue to produce more and more content across OpenStack events we've opened a position for a content manager role to help with this workload.


On May 19, 2016, at 5:26 PM, Nick Chase <nchase@mirantis.com> wrote:



On 5/19/2016 4:40 PM, Florian Haas wrote:
So to summarize, it looks like there are four options currently being discussed:

(1) Keep everything exactly as it has been (this implies rejecting
Claire's original proposal to drop voting).
(2) Drop voting, put everything in the track chairs' hands.
(3) Use some form of random talk selection, in combination with
submission pre-filtering and public comments.
(4) Use an approach where speakers vote on a random subset of talks
and rank them, also with public comments.
You forgot:

(5)  Keep everything as-is but change the "voting" to "commenting"
(6)  Keep voting but prevent social spam by removing static URLs
(7)  Change the voting structure ...
(8)  etc...

In other words, there are a LOT of options being discussed here, to be honest.  It's not as simple as just "should we vote or not".  In fact, we've left out the simplest version:

(9)  Keep everything as-is but make it more clear that the votes are just a guideline

I'm not sure the best way to proceed here from a problem-solving standpoint, to be honest.   I'd suggest that we had people come up with concrete proposals for up/down votes, but that's kind of what we've been doing here and the options have expanded rather than contracting. :)

----  Nick

_______________________________________________
Community mailing list
Community@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/community