How We Selected Talks for PGConf US 2015

In the spirit of open-source, we would like to share how we handled the talk selection process for PGConf US 2015. This post will discuss the entire process of how a talk ends up in one of our United States PostgreSQL Association conferences; our goal is to help you understand what our conference is looking for in talk proposals and help you decide what you submit for PGConf US 2016!

In the past conferences we organized in New York, we had to do a lot of direct outreach in order to attract speakers, to the point where we would have just enough talks to put on a program. This changed this year; while we did do some direct outreach, we were able to rely more on prior attendee / speaker experience, referrals from the PostgreSQL community, and our marketing efforts, which lead to 107 talk submissions!

At first, we were planning to do 3 conference tracks across 2 days, for a total of 32 speaking slots. After Jim & myself did a first pass on the talks, we decided that the overall quality of the submissions warranted us adding either another day to the speaking sessions, or another track. Because of our contract with the hotel and the fact that we know people already booked their travel, we opted to add a 4th track, and expand to 44 speaking slots.

We assembled a committee with the understanding that it works similar to an operating committee of a business: committee members were given the option to say "Yes" or "No" to talks and provide feedback as to their reasoning. However, Jim & I ultimately make the final decision on all of the talks. Overall, most of the talk selections were in-line with the committee consensus, but there were some exceptions which I will discuss later.

We chose the committee members based on their history with the PostgreSQL, experience in technology, and industry they are representing.

  • Mehboob Alam: Has helped with marketing and organizational efforts in previous conferences; experience in healthcare.
  • Jonah Harris: Longtime PostgreSQL community member, has worked with large-scale PostgreSQL deployments as well as Oracle. Jonah also had a lot of feedback on the talks from last year's conference, so we challenged him to help us with the selections this year ;)
  • Bryan Doyle: Helping to organize the Regulated Industry Summit; experience in enterprise and data deployments in regulated industries

For our part, Jim comes from the enterprise consulting world and has seen PostgreSQL deployed in many different configurations. I consider myself on the app development / startup side of the equation, thus I try to consider talks from that perspective.

Two quick interjections:

  1. The talk committee did an outstanding job; we gave them less than a week to review 107 proposals and they all followed through. Great job team!
  2. If you are interested in being a part of the talk committee for PGConf US 2016, please talk to Jim & myself at the conference :) Our intention is to grow the conference even further in 2016, which should mean more talks to review and the need for fresh perspectives. And yes, attending one of our prior conferences is a requirement to be on the talk committee.

When it came to reviewing the talks, we gave the committee very general guidelines, and this was on purpose: We wanted everyone to bring their unique perspectives and analysis to the reviews. However, there were some central preferences that came out of the review process, which were:

  • Does this proposal have real-world applications? Case-studies scored major points with everyone.
  • If the proposal revolved around a tool or utility, would the talk cover real-world deployments of the tool and what problems it helped to solve?
  • Did the talk offer a fresh perspective on a PostgreSQL topic?
  • If the topic was forward-looking, would it be appropriate for a user audience and provide enough information so users could take action now?
  • Did the proposal provide enough information about the talk? Generally proposals that were more detailed met with more favorable reviews. Also, the actually wording of the proposal counts; if you do not write your proposal properly, the odds are you will have trouble conveying your message while speaking.
  • There were some cases where Jim & myself reached out to talk submitters to clarify points in the proposals, request changes, or make suggestions. We were pleasantly surprised and glad that everyone we reached out to got back to us almost immediately!

    After the group reviews process, there were about 63 talks that met with favorable opinion, and about 55 that were universally favored. This of course makes ita challenge when you only have 44 slots to select, thus we had to make some tough decisions. When it came to make the final cutdown, we considered a few things:

    • Does the talk cover a topic that is not addressed by any other talk submissions?
    • Do we have the correct balance between beginner / intermediate / advanced talks?
    • Do we have the correct balance between different topics, with heavier weight placed on case-studies / PostgreSQL deployment strategy talks?
    • How cool is the content of the proposal? We don't want attendees to be bored ;) As an example, there was one proposal that talked about using a foreign-data wrapper to control a Lego robot - no pressure, but we are expecting a live demonstration of that!
    • Does the speaker have previous speaking experience, and at a PostgreSQL conference? This by no means was a deal-breaker; in many ways, we were looking for new speakers to bring into the community. We do want to make sure we are bringing in speakers who are comfortable with public speaking and can deliver clear messages to their audience
    • Does the talk fit the overall theme we are trying to convey at the conference? In general, the PGConf US series is looking for people who can talk about a particular feature or concept within the context of real-world usage, and that can include talking about technologies other than PostgreSQL.
    • Will the talk be able to better educate people how to use PostgreSQL within the greater world of technology?

    Even with this criteria, on the day we notified speakers, we were still left with 12 talks for 4 slots to fill. In a perfect world, we would have chosen all 12 of those talks, and we had to make some tough choices. There were some talks that the entire group universally voted "Yes" on that we ended up turning down, usually for one of these reasons:

    • The topic was covered in some way by someone else who also received universal approval
    • We ultimately decided it did not provide enough use-cases based on the proposal
    • We just ran out of space :(

    Looking towards PGConf US 2016, we are planning on adding a 3rd day of talks, which will mean we will have anywhere from 48 to 68 slots available based on the number of talk submissions and/or the number of concurrent tracks we want to run. We may consider adding a "PostgreSQL Hackers" track to have some more abstract, forward-looking concepts, though the criteria of having a real-world application will still apply to those talks.

    If you have any questions, comments, thoughts, or whatever, you can reach out to us at pgconf@postgresql.us