At the end of April, Mark Wong & I headed up to Bellingham for LinuxFest. This was my first time at the conference in several years, and though the expo hall was a bit smaller than I remembered, they had 1849 registered attendees! That is a truly excellent number for a community-run conference in a far corner of the US.
Somehow we had three Postgres booths: PgUS, 2nd Quadrant, and SEAPUG. This caused some confusion among the attendees. As in past years, booth visitors ran the gamut between experienced folks, random people who dropped in to see what was going on, and That Guy Who Just Wants to Argue^W Discuss Why MySQL Is Better [tm]. I hadn’t actually seen him in a while, it was good to catch up. :koff:
Talks I attended:
Emily Dunham’s Thinking in git, where I learned about git commit –dry-run and git pull –rebase (which makes it look like you were smart & pulled when you were supposed to.)
Robert Bernier’s Welcome To Total Security gave a fun historical overview of Postgres, and very good coverage of both beginner and more advanced information.
Frances Hocutt’s Why are these people following me. The big lesson for me from this talk was “Just because the abuse isn’t aimed at you, doesn’t mean you don’t suffer from it”, as I am currently dealing with a similar situation. Also: “Assume good faith” doesn’t mean “ignore evidence of bad faith”.
Eric Worden’s start_date, end_date: Calculate! included a pretty appalling example of a temporal table design, and how to fix it. This just convinced me even more that range types are The Right Way to handle these requirements.
On a personal note: I’ve been doing Pg-only confs for the past few years and made a rookie presenter mistake with my talk: I assumed a level of familiarity with Postgres that my audience didn’t have. I’ll need to adjust for that next time.