Speaker Interview with Jeff Davis

Could you briefly introduce yourself?
I’ve been involved in the Postgres community since around 2000. I started as a web developer, then DBA, then database internals for a number of products that involved Postgres in some way.
Have you enjoyed previous pgconf.eu or FOSDEM conferences, either as an attendee or as speaker?
Yes, the event in Prague was amazing! The conference and the city.
What will your talk be about, exactly? Why this topic?
It’s about writing non-trivial Postgres extensions in Rust; the same kinds of extensions that can only be written in C today. Things like background workers, etc.
I believe it’s really important to connect like-minded communities -- in this case, the Postgres and Rust communities. Bringing together people and ideas can have a very big impact and help both communities grow together.
What is the audience for your talk?
People who want to build a Postgres extension that needs lots of libraries that are less accessible in C, but easy to find and use with a modern ecosystem like Rust. One of my examples uses tokio, a futures runtime, to make a concurrent network server as a Postgres background worker. That could be a lot of work in C, but in rust, it’s a much more reasonable thing to do.
What existing knowledge should the attendee have?
The attendee should have some knowledge of Postgres extensions in C, the Rust language, or both. I believe a talk is a starting place, so it’s not necessary to understand every detail as long as they understand why the topic is important by the end.
My hope is that a Postgres developer will walk away thinking: “Wow, it looks like the error handling can work across languages!”; that a Rust developer will walk away thinking “Wow, a Postgres extension would be a great way to expose this new library I’m working on!”; and that someone who knows both won’t stump me too many times during the talk.
What is the one feature in PostgreSQL 12 which you like most?
I like the automatic inlining of common table expressions (WITH …). I like the clarity of using WITH, but I always avoided it because it could be bad for performance. I look forward to nicer queries now that they are no longer penalized.
Which other talk at this year's conference would you like to see?
Just one? The Data Compression talk by Iwata Aya is definitely on my list.
Which measure, action, feature or activity would—in your eyes—help to accelerate the adoption of PostgreSQL?
Postgres is already doing an awesome job, so I don’t think any real course correction is needed here.
Of course, there is plenty of room for improvement, but a lot of that is already being done and the future looks very bright. Community development, enterprise features, security features, more extensibility are all looking good.
Some might say we are a bit slow on some features that have been wanted for a long time, but I think our pace has served us well over the long term.