How I Learned to Embrace my Chaotic Brain

Boriss Mejías — Fri, 10 Oct 2025 21:05:51 +0000

More than a decade ago, I applied for a job in Brussels and part of the process was to pass a psychological assessment. It was a very interesting experience and I learned a lot from the results. The report said something like: "the brain of Mr. Mejías is unstructured and chaotic. Nevertheless, he manages to solve complex problems and has an eye for details". Based on that, the recommendation was to work on improving my methodology of working. The whole report was so accurate that I decided to follow that advice and I went in search for a methodology that would give some structure and reduce some entropy in my brain.

Around that time I was reading "The Long Dark Tea Time of the Soul", by Douglas Adams, which is part of the series of Dirk Gently, the holistic private detective. In that book, Dirk Gently was saying to someone in a café: "I am not like other private detectives. My methods are holistic and, in a very proper sense of the word, chaotic. I operate by investigating the fundamental interconnectedness of all things." Ha! Might this be a method for me? I thought. And despite the fact that Dirk Gently uses the term holistic to scam his clients to justify trips to the Bahamas because they might be somehow connected to the current investigation, we are talking about Douglas Adams here, and his jokes always have a hidden truth, despite how silly that truth might be.

I know the term holistic sounds very esoteric and it is usually presented in opposition to the more scientific approach of reductionism. I say that it can be used not in opposition but complementary to reductionism. In reductionism, one splits a complex system into smaller and simpler parts, so that by understanding the parts, one can understand the more complex system. However, many times we are faced with the situation that a system is larger than the sum of all its parts. This is because the parts are connected to build the system, and they influence one another. Therefore, the system is formed by its parts and their interconnection. A small but relevant detail.

Why am I telling you all this? Well, this week I had the honor to participate in the "Talking Postgres" podcast, hosted by Claire Giordano, where we talked about the fundamental interconnectedness of all things. I’m still wondering how I managed to convince Claire to bring this topic to the podcast. I think she just trusted me on this. And the reason I wanted to talk about this topic is because I know that there are many people whose brain is wired in a similar way to mine, and that the ideas presented in the podcast might help a few of them to find a method that works for them. It might also help folks whose mind is well structured and organized, because the fundamental interconnectedness of all things is there to be found regardless of how your brain works.

In the podcast we talk about solving technical problems around PostgreSQL, chess, music, building the PostgreSQL community, and the connection between perfectionism, Darwinian evolution, and PostgreSQL releases.

The podcast can be listened to on Pocket Casts, YouTube and different platforms as listed on the Podcast website https://talkingpostgres.com/. If you happen to listen to the episode, I’d love to ask you to send me your thoughts, or maybe we get to meet at a PostgreSQL conference and we can talk about these ideas.

Hail Slonik!

PgDay Paris - Postgres Community, cheese and wine

Boriss Mejías — Wed, 20 Mar 2024 13:03:54 +0000

If one thing can be said about PgDay Paris, it is that they really focus on the community. One can see this from the selection of talks to the interactions between the attendees in the hallways. And all this mixed with great technical content.

I really like the fact that the Keynote talked once again about community. Valeria Kaplan (Data Egret) presented a very nice overview of all the different groups and activities going on within Postgres, combined with what we can expect in the professional realm. Valeria also gave us a remarkable quote from Stefanie Janine Stölting:

"Postgres doesn’t have a community, it is a community"

We should make t-shirts with that. And to exemplify how the community interacts beyond the software, the PostgreSQL Chess Club brought two boards and played several blitz games between talks, inviting new members to join the club. The boards were brought by Anthony Nowocien and Derk Van Veen, and I remember Ea Phan Van and Dmitry Dolgov among the players. The players discussed the games after they were over, which is a good practice to learn from each other. And thanks to the event organizers for giving us the space to play the game.

From blitz games to Lightning Talks

Another thing to thank the organizers for is the Lightning Talks. Brilliant idea to schedule them right after lunch. If there is something to keep you awake after a French meal it is a good load of energy coming from 5-minute talks with excellent content and a great lineup. I not only enjoyed this session the most, I also learned a lot. Great usage of pgvector to find cheese, by Matt Cornillon (Aiven). Brilliant geeky delivery of electric elephants by Chris Ellis (Intrbiz) (someone should invite Chris to a podcast… he has so many interesting things to talk about).

And my two favorite speakers of the event were also in this lineup: Claire Giordano (Microsoft) and Floor Drees (Aiven).

Claire was motivating people to fight the butterflies in the stomach (or to embrace them) and give their first Postgres talk. I’d say she also reached out to a larger audience motivating experienced speakers to keep on submitting and to improve their talks. Her slides and storytelling are really well done (by the way, submission for PgConf Belgium closes on March 25).

Floor’s talk reminded me of a lightning talk I gave in October at Devopsdays Eindhoven. Whereas I was talking about what makes Postgres a sustainable project, Floor confronted us with the risks and the challenges we are facing. Especially those we don’t really see, like reliance on unpaid labor, unclearly defined roles of contributors, and a skewed gender distribution (i.e. too many dudes, we should learn from the Vikings). She also confronted us with some pictures of young Postgres developers I didn’t recognize, and a picture of Black Sabbath… if you want to know more about Floor’s talk, you definitely want to read this post

More on talks

This post wouldn’t be complete if I didn’t mention other talks that fed my todo list with more things to investigate. Dirk Krautschick (Aiven) got everyone interested in the PG_PROFILE extension to go beyond pg_stat_statement for query analysis. Funny way to motivate the talk: he got sick of the previous one that only referenced PG_PROFILE. I still don’t know how he managed to place the word “Sustainable” in the title. Dirk, you owe me a beer for that.

Dave Pitts (Adyen) gave a very interesting talk explaining B-Trees and LSM trees (Log-Structure Merge-trees), comparing the use cases, and pros and cons of each of them. The style was quite academic which I appreciate a lot. My understanding is that with high-volume data injections, we will see more and more use cases for LSM trees.

Working with Derk

On a personal note, I gave a presentation with Derk van Veen (Adyen) on the collaboration between developers and DBAs. It is a talk about the interaction between people with a common goal, but with different motivations and priorities, and with different prejudices. Completely aligned with Ilya Kosmodemiansky’s worst practices 5 and 6: "5. as a DBA, never talk to your developers team," "6. As a developer, never talk to your DBAs team".

During our talk, Derk and I managed to define a better table partitioning strategy, and to agree on a contract written in code to avoid ambiguity, and to improve the maintenance of table partitioning in the long run. This idea of Contract as Code also helped us to explore the power of JSON support in Postgres. Whereas I love JSON in Postgres, Derk uses it just for convenience, and to be able to communicate with developers.

The talk itself was a challenge for me because Derk was coming with a lot of new ideas about how to build the talk, some of them scared me I must say, but that’s why the whole process was fun. I really enjoy working with Derk and I hope we will do more collaborations in the future (hard to do more collaborations in the past).

It is another example of how the Postgres community collaborates because both of us work for different companies (EDB and Adyen). And with that sentence, I can come back to where we started. As Stefanie says "Postgres doesn’t have a community, it is a community", and PgDay Paris is one of the places where the community can flourish.