PGConf.EU 2023 Review

Posted by ads on Friday, 2023-12-22
Posted in [Events][Postgresql][Postgresql-News]

PGConf.EU 2023 is over, time for my personal review of the conference. Overall we got good and very good feedback, for most parts. Just the coffee …

But let’s start from the beginning.


The conference took place in the Clarion Congress Hotel Prague. Perfect venue, if you ask me. Not too big, still plenty of space both for walking around during the conference as well as for the sponsors. Conference rooms large enough, a nice bar downstairs, and a shopping mall attached to the hotel. Rooms are clean, just the floor plan with the rooms is a bit confusing.

4 screens in the events area, showing the conference logo
4 screens in the events area, showing the conference logo

Subway is connected to the hotel, bus station is right outside. Going to Old Town is around 20 minutes by subway. That’s ok.

Best part? It’s December and the ice cream shop in the mall is open (Proof, proof, proof, proof)!

The only complaint about the venue was the coffee. Myself not being a coffee drinker I can’t chime in on that, but it has been mentioned several times that there is “one good coffee maker downstairs”, and the other machines are not good. People even went to the mall to get a coffee.

Conference shown on a screen in the elevator
Conference shown on a screen in the elevator


This year’s keynote was presented by Simon Riggs, who retired last year during PGConf.EU 2022. He laid out his vision and ideas for the next 20 years of PostgreSQL.

One particular quote sticks out:

More features (in #PostgreSQL) will be written in the next 20 years than have been written in the past 20 years.

And Simon also claims that your job is secure if you set on PostgreSQL. Glad to hear that!

Simon Riggs - PostgreSQL: The Next 20 Years
Simon Riggs - PostgreSQL: The Next 20 Years


A conference does not happen without speakers, and the Call for Papers Committee did a very good job selecting a broad range of interesting talks.

There is one talk on the schedule which we discussed beforehand, because we didn’t know if attendees will accept the topic. Among the team we made sure that an experienced room host is hosting this room, as well as having additional team members in the room, just in case. In the end, everyone accepted that it is what it is: a talk explaining how technology is used in controversal situations.

Before the conference we invited the speakers to a dinner. Tomas found the Restaurace Marina Garden, which is a nice little place which served plenty of good food for us. Very good choice.

By the way, the nice blue background for the screens was made by Chris Ellis. Some speakers adapted this for their slides as well.


Of course Slonik was in Prague!

Even! More! Pictures!


I attended a couple of talks during the conference:

Trying to be Barbie in Ken’s Mojo Dojo Casa House

Was really looking forward to this talk by Karen Jex and Lætitia AVROT. They present how women in tech are at a disadvantage in many aspects. The peg for this talk is the Barbie movie, which also takes an unexpected turn. The presentation was very well done, and they added plenty of facts (as you can see in the slides).

You can find the transcript here, and the slides here.

Karen Jex and Lætitia AVROT - Trying to be Barbie in Ken’s Mojo Dojo Casa House
Karen Jex and Lætitia AVROT - Trying to be Barbie in Ken's Mojo Dojo Casa House

Karen Jex and Lætitia AVROT, with room host Floor Drees
Karen Jex and Lætitia AVROT, with room host Floor Drees

At the end of the talk, they invited all women on the stage for a group picture.

Group Picture of all Women
Group Picture of all Women

From Map to Reality: Using PostGIS in Warfare

This talk by Taras Kloba explains how modern technology influences life even in war zones. It’s very interesting to see how fast technology can advance.

pg_failover_slots: The missing piece

This talk by Peter Eisentraut brings a missing piece for logical replication. We all know that dealing with replication slots is complicated and surely this can be handled better. This extension can do that.

Peter Eisentraut - pg_failover_slots: The missing piece
Peter Eisentraut - pg_failover_slots: The missing piece

So you want a PGDay in your city

Henrietta Dombrovskaya and Teresa Giacomini talked about how to start a PGDay in your city. The talk shows their experiences working with PgUS, but most of it can be translated to any new conference anywhere in the worlds.

Henrietta Dombrovskaya and Teresa Giacomini - So you want a PGDay in your city
Henrietta Dombrovskaya and Teresa Giacomini - So you want a PGDay in your city

Add geocoordinates to your addresses in PostgreSQL

I kind of had to go to this talk - I am the presenter. It’s a new talk, but a topic which I had talked about in the past. Resolving addresses to geocoordinates is always a hot topic, many services in our daily lives depend on coordinates. OpenStreetMap (OSM) allows to resolve an address to coordinates, and I show how to do that in the database. This has the big advantage that the data doesn’t have to leave the database, you can enrich your existing data with even more useful GIS data.

On top of the functionality, I validate how good the resolver (Nominatim) in OSM works, by loading sample data sets and verifying the results. Peek preview: for most data and use cases it turns out to be ok.

The slides for the presentation are available on my website. Claire captured me while talking, and so did Lenz and one of the volunteers on the official conference account.

Please note the t-shirt, which shows the Prime Meridian at the Royal Observatory in Greenwich, London.

me walking into my own talk, with ice cream in my hand
me walking into my own talk, with ice cream in my hand

Lightning Talks

Friday afternoon, right before the conference closes, is the rigjt time for Lightning Talks.

I don’t remember all of them, but I do remember a fewm:

  • Kai Wagner (Percona) talked about “Encryption is not a crime
  • Boriss Mejias (EDB) announces the PostgreSQL Chess Club
  • Christoph Berg and Devrim Gündüz talked about the Debian/Ubuntu and the Red Hat/Fedora package infrastructure
  • Andreas Seltenreich talked about tracking RFID chips, specifically tire pressure sensors, and identify specific cars, models and tires over the year

The four Platinum sponsors have a keynote on Thursday afternoon. From the feedback I gathered on Thursday evening and Friday, three of the four keynotes have been good and did target the audience. One sponsor needs to work on the content, quote: “Not aligned to the target audience”. Let’s leave it at that.

GSoC Students

This year the Call for Papers committee accepted two talks from Google Code of Summer (GSoC) students, both from India. We as the conference organizers decided to sponsor travel for both, which is a good tradition in the PostgreSQL Project: the project offers students that they can travel to a conference if they get a talk about their GSoC project accepted.

The visa worked without problems - given past experience that’s already surprising. And then they told us on the day of their travel that their flight from New Delhi is cancelled. bummer They were supposed to arrive on Tuesday, in time for the speaker dinner. They then got re-routed to Mumbai and Istanbul. And while tracking their flight, one flight radar (see screenshot) looked for a while like they have to dump fuel and make an emergency landing in Azerbaijan or Iran. But it turns out that this website had a data glitch, and two other websites showed the plane on track. In the end they made it to the conference on Wednesday, right in time for the evening reception.

Flight 6E-17 from New Delhi to Istanbul
Flight 6E-17 from New Delhi to Istanbul

Akshat Jaimini wrote about their journey here and here. Rajiv Harlalka posted about the GSoC participation earlier, and also here.


As a volunteer-run conference, quite a few volunteers help us every year to make every conference a success. People contact us at some point before the conference, we then gather the names and a few weeks before the actual conference we bring everyone together.

For the volunteers we had a box of cookies (very delicious) and a set of crystal glasses with the conference logo on it. The glasses look very nice, and reminds me of the glasses we had back in 2012.


We had a record number of 720 registrations! A few people didn’t pick up their badge, the real number is somewhere around 700 people. Still the largest PostgreSQL Conference around!

Attendee badges, sorted by name
Attendee badges, sorted by name

Job Board

For the third time we had the job board up. Everyone was allowed to post there, no matter if you want to hire someone or if you are looking for a new job. The board was in good use, and I’ve seen people checking out the postings during the entire conference. Good luck with your search everyone!

Job Board
Job Board

Community Reception

On Wednesday evening, after the last talks, we had the Community Repection in the large sponsor area. Everyone was talking with everyone, not too loud music, drinks and some finger food. In general, everyone seemed to have a good time. We should keep the reception at the conference venue, it’s much more convenient.


We had - for the second year in the row - childcare available. The hotel moved this into a room a bit away, where no one needs to walk by. That was good, no one was disturbing the kids there, and the professionals could take care of the kids. A couple of attendees took up this offer, brought their kids along and attended the conference while their kids were entertained.

PostgreSQL Project Steering

A hot topic during the entire week in Prague has been the steering for the PostgreSQL Project. This came up in several discussions, a couple of times in connection with the now solved trademark issues.

From a developer perspecitive, it is still too complicated and too tedious to develop for the core project. My understanding of the discussions is that everyone agrees that the development model can’t be simply changed to something like Pull Requests on GitHub or Merge Requests on GitLab. However new developers need Sitzfleisch until they get even simple patches accepted and merged. And the discussion culture on the mailing lists is in need of improvement. Robert Haas blogged about this. The same Robert Haas blogged earlier this year about development numbers: “66% of the new lines of code were contributed by one of 14 people, and 90% of the new lines of code were contributed by one of 40 people”. For a project the size of PostgreSQL, that’s not a lot of contributors. Given the amount of emails reported in the same posting (another topic which drives away contributors), it seems that the majority of contributions come from a small(er) number of companies which pay people to work on the code.

Auxiliary projects are mostly fine, although there are some which are only maintained by one or two people and it is not clear if anything is even happening there.

What came up several times during the discussions is that people wish for more transparency: “what are certain projects within the community doing”, “who is even working on what projects”. A good example is the Code of Conduct commitee, which is both transparent in what they are doing and also has yearly reports. Many other oprojects simply exists, and no one knows details. More like what the CoC is doing, please.


On Monday evening, we had dinner in the Pohostinec Karlín restaurant, about 30 minutes by subway. Or about 5 km walking back to the hotel. Very nice place with good food.

Thursday evening a small group of friends went to the Old Town and we had dinner in the Restaurace White Horse. The restaurant is underground, outside is the Christmas market and the Prague Astronomical Clock.

On Friday evening, after the conference was over, we went to the 2Baldies Restaurant nearby for pizza and other good food and wine.

What’s next

The next conferences (for me) are FOSDEM PGDay and FOSDEM Devroom in February in Brussels, as well as PGConf.DE 2024 in Munich, Germany. See you there!