PASS Summit Session Selection

Let’s make it democratic. Let’s ensure we get solid sessions from key people. And let’s save a ton of work in the process.

Spotlight Sessions:

There are certain folks that are extremely knowledgeable in their areas of expertise. They also happen to be excellent presenters. Have the spotlight sessions and invite them to present a talk. Limit the number of these, obviously. However, this ensures top speakers are presenting.

Let the Community vote:

Since folks had to update their community profiles in order to participate in the voting, let’s go down that road, except for session selection. It doesn’t matter if it’s a pre-con, a regular session, or a lightning talk. Put the abstracts up, complete with who is giving them, and give the community a chance to vote on a particular number. Perhaps for each track you get to vote for your top 10.

Does this skew things in favor of those who are more popularly known? Yes. But it also means the community is seeing who they want. So what about those who don’t have as solid a reputation? Let them build a reputation via the following:

  • local user groups
  • SQL Saturdays
  • virtual chapters

That’s what’s effectively being done by having a speaker rating score, anyway.

Set the deadline, tabulate the votes, and then take the top presentations per track and schedule them.

What if there’s a tie? Use a random number generator to make the selection.

But what if there are too many tracks?

Limit the number of submissions. Perhaps:

  • 1 pre-con
  • 2 regular session talks
  • 2 lightning session talks

Speakers try to game the system today because they don’t know what the selection committee for the Summit or for a particular SQL Saturday will want. So they submit more sessions than they actually want to give. Limit the # of submissions. What about panels? If the panel discussion is that important to you, then it takes one of your slots. No apologies, because you know that ahead of time. This causes a speaker to focus on what topics he or she really want to speak on and think will go over with the audience.

Will there be issues?

Of course there will. But this is more transparent than having selection committees behind the scenes. PASS, after all, is a community organization. It also eliminates any board influence (and there has been board influence in the past). So let’s keep this simple.

But that’s not how XYZ Conference does it!

No, it’s not. But XYZ Conference is probably not run by a community organization. If it is, perhaps they should follow the same model.

15 Comments (+add yours?)

  1. timcost
    Jun 27, 2014 @ 11:42:35

    Interesting. I’ve done a lot of presentations at SQL Saturday events, do they contribute to my speaker score rating? If they don’t, what does? How would I find what my rating is or when it changes? I wasn’t even aware I had a rating.

    Reply

    • K. Brian Kelley
      Jun 27, 2014 @ 12:29:52

      I’m not sure how you find out your rating. That puzzled me when I submitted a few years back. Or even what the real criteria were.

      As far as SQL Saturday, as an organizer for two of them, I never had a mechanism to get the ratings back to PASS. Unless that has changed, there isn’t one, meaning SQL Saturday ratings don’t go back.

      Reply

  2. timcost
    Jun 27, 2014 @ 12:41:17

    Thanks. I enjoyed your post. Very thought provoking. I hope it prompts change in PASS.

    Reply

  3. Steve
    Jun 27, 2014 @ 14:31:44

    I like this concept. I think it would be valuable to have an “up and comers” section to support SQL Saturday and UG speakers who don’t have national presence. It will add to the variety.

    Reply

  4. Jason Horner (@jasonhorner)
    Jun 27, 2014 @ 14:40:41

    @Tim – A component of your speaking rating was the box that asked about your previous speaking experience in the abstract submission form. You could list your previous there and the speaker review team would take that into consideration. Other than that, the speaker team was encouraged to find information about the speaker from blog posts and other sites.

    Reply

  5. Allen White (@SQLRunr)
    Jun 27, 2014 @ 15:56:53

    Good post, Brian, except we tried voting once and it led to excessive campaigning for votes. Many speakers, myself included, abhor the idea of campaigning to be selected, yet that’s what others used to get there. It’s wrong for the process, and it doesn’t make for a good experience for the attendees.

    Reply

    • K. Brian Kelley
      Jun 27, 2014 @ 16:05:58

      Keep in mind, though, that you will eliminate some of the campaigning by having the spotlight sessions, which is why I suggested keeping those around. Those folks won’t enter the general pool, meaning those folks won’t have a “need” to campaign. That also partially addresses the US bias issue.

      Reply

    • brento
      Jun 28, 2014 @ 15:24:17

      Allen – SQLbits does session voting, and it’s my favorite conference that I’ve ever attended. It seems to work out pretty well. The session voting isn’t the only factor taken into account, of course, and it’s just one part.

      I would also argue that public campaigning and competition to build the best sessions is not just a good thing – it’s a GREAT thing. The free market builds the best options. Having attended a few unconferences and open space sessions, I’m sold on the participant-driven agenda. What’s wrong with letting buyers pick the product?

      Does it mean that popular people will always get sessions? Sure, but remember why they’re popular – they have a following of folks who want to see their work. Set a cap on the number of sessions an individual can get, and set a cap on the number of sessions per topic.

      When I see people say “voting is bad,” what I really hear is “popular people don’t produce good sessions” – and I have a hard time stomaching that message.

      Reply

  6. Allen White
    Jun 28, 2014 @ 15:41:13

    Brent, we will never agree on this issue. I want the Conference team to select the best possible content designed to give attendees the content that meets their needs. In the PASS scenario, campaigning causes “popular” – read “loudest” – people to get more attention, thus more votes. I’ve seen “popular” people produce really bad content, and that is and should be unacceptable. I get it, you’re popular. I’ve also found your content worth paying for, as you know. Not everyone is you, so I’d rather leave it to the Conference to select the best material for the attendees.

    Reply

    • K. Brian Kelley
      Jun 29, 2014 @ 00:31:48

      This is why I indicated still having spotlight sessions. You’re going to get top content that way.

      As far as seeing popular people produce really bad content, we can easily make the same argument that we’ve seen some bad presentations that had excellent abstracts. Unless someone is assessing every session submission with a run through, bad sessions are going to get in.

      Reply

  7. brento
    Jun 29, 2014 @ 09:37:48

    Allen – I’d agree with you if the selection committee never picked a bad session, and attendee feedback was consistently 5’s across the board. It’s … not. And I just can’t believe that a handful of volunteers can do a better job of curating content in their spare time than the entire community could do.

    Can we really complain that Microsoft doesn’t listen to votes in Connect, and at the same time, say that votes aren’t the right thing to do for PASS Summit selections?

    Reply

  8. Trackback: codegumbo » A tiny step in the right direction #SQLPASS
  9. Trackback: A tiny step in the right direction #SQLPASS - SQL Server - SQL Server - Toad World
  10. Trackback: Is the SQL Server Community Too Successful? | Aunt Kathi's SQL Server Home

Leave a reply to K. Brian Kelley Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.