HYTRADBOI

Hard refresh? They've been up for a long time. How long could I possibly have set the cache headers for...

I'm kind of doubtful that it's the ui that affects behaviour. I suspect the difference is actually that a) a big conference like splash has a significant influence on your career and b) people who speak regularly are often used to writing their talk on the flight to the conference.

I wrote a post-mortem. Significantly less mortem compared to 2022 though.

www.scattered-thoughts.net/writing/hytr...

All the videos are up at www.hytradboi.com/2025#program now.

"The website comes back ... 👀"

Here's an awesome talk by Iain Ireland to get you started!

www.hytradboi.com/2025/0a4d08f...

Eating chocolate, drinking barley wine and watching #hytradboi: Perfect Friday night 🥰

This is running now! I have a talk on how debuggers work in 30 mins

We're starting in 30 minutes. I hope you all remembered to bring your databases.

Get a ticket to watch @hytradboi.bsky.social tomorrow and discuss some wacky stuff in Convex! Or wait, the conference videos go up later—but this is a remote conference that's fun to participate in live, there are some real neat folks in the discussions.

"Never write a database. Even if you want to, even if you think you should. Resist. Never write a database. Unless you have to write a database. But you don't."

Talk 28: "Good thing we're not writing a database" by Peter van Hardenberg (@pvh.ca)

Talk 27: "Let's run a million benchmarks" by Yao Yue (@thinkingfish.bsky.social)

0900-1000: Compilers Database query compilation: our journey Thomas Neumann & Viktor Leis A YJIT interview Maxime Chevalier-Boisvert A quick ramp-up on ramping up quickly Iain Ireland Can we democratize JIT compilers? Haoran Xu Safe and productive performance with user-schedulable languages Jonathan Ragan-Kelley 1000-1100: Understanding programs How debuggers work Sy Brand Debugging compiler-optimized code: how it works and doesn't Stephen Kell Side-Eye: ask your programs anything Andrei Matei Let’s run a million benchmarks Yao Yue Rocket science of simulation testing! Alex / matklad 1100-1200: Wild ideas Back to modularity Daniel Jackson DB usability: as if Jonathan Edwards Twizzler and far out memory sharing: precise abstractions Daniel Bittman Programming without pointers Andrew Kelley Throwing it all away - how extreme rewriting changed the way I build databases Tyler Neely 1200-1230: Programmers are people A case for feminism in programming language design Felienne Hermans Malloy, mic drop, peace! Michael Toy 1230-1300: Lightning talk buffet Learning about the odd bits of SQL by reading the PostgreSQL docs Chris Zetter Hacking Observable notebooks from within Tom Larkworthy Zero copy data structures Evan Chan Reliable serverless needs distributed transactions Stu Hood Pangeo is a database Alexander Merose Rubbing a database on a language server Philip Zeyliger Language agnostic simulation testing on a budget Stevan A Shapeshifter: using LLMs inside a database for schema flexibility David Nachman Why S3's conditional writes made people excited Miikka Koskinen pghttp: backend-free, lowest latency web apps Damir Simunic 1300-1400: Query languages Pipe syntax in SQL; it's time Jeff Shute PRQL: a modern, pipelined SQL replacement …

The final program:

HYTRADBOI is tomorrow! Feb 28 0900 PDT. The chat will run for days, so it's worth joining even if you can't make the early hours.

Bring all your friends too! Re-skeet, share on your favorite forum, put flyers under your neighbours doors etc.

See you tomorrow!

www.hytradboi.com/2025/#tickets

A table examining different consensus protocols across many categories, showing that Raft is never the best option in any one category.

Talk 26: "Enough with all the Raft" by Alex Miller (@alexmillerdb.bsky.social)

I have a talk about S2 at HYTRADBOI this friday! www.hytradboi.com/2025/

A diagram of the baseline interpreter. The core is a loop over a switch on opcode. The branch for each opcode is generated by the compiler, and the dispatch logic is duplicated into the end of each branch.

Talk 25: "A quick ramp-up on ramping up quickly" by Iain Ireland

Building a KV-store with S2 ● Multi-primary, horizontally scalable, distributed KV-store ● All writes regionally durable ● Linearizable reads from all replicas https://github.com/s2-streamstore/s2-kv-demo

Talk 24: "Serverless primitives for the shared log architecture" by Stephen Balogh (@sbalogh.bsky.social)

A large PRQL query.

Talk 23: "PRQL: a modern, pipelined SQL replacement" by Tobias Brandt (@tobiasbrandt.bsky.social)

User-schedulable languages decouple what to compute from how to compute it (exposed for programmer control).

Talk #22: "Safe and productive performance with user-schedulable languages" by Jonathan Ragan-Kelley

Other kinds of talks.

www.scattered-thoughts.net/writing/othe...

Someone start CheckIfItActuallyWorksCon. I'll help.

It's starting to look a lot like ~Christmas~ HYTRADBOI!

Just one week to go.

Thomas scribbles furiously on a whiteboard.

Talk #21: "Database ideas in Convex" by Thomas Ballinger

HYTRADBOI ("Have you tried rubbing a database on it?", @hytradboi.bsky.social) is happening again! It's a really cool, forward-looking online event on everything databases. Highly recommend to check it out and attend. Happening on Feb 28, 9:00-15:00 PDT.

www.hytradboi.com/2025

Lightning talk #9: "Why S3's conditional writes made people excited" by Miikka Koskinen

Just a couple of days left to submit your lightning talks! www.hytradboi.com/2025#lightni...

Lightning talk #8: "Shapeshifter: using LLMs inside a database for schema flexibility" by David Nachman

Lightning talk #7: "Language agnostic simulation testing on a budget" by Stevan A

A demo showing the query "select symbol_name, count(*) from symbols group by 1 order by 2 desc" running over the data exported by a typescript language server.

Lightning talk #6: "Rubbing a database on a language server" by Philip Zeyliger

Alexander talking. The captions say "Zarr is OLAP with all the abstractions peeled away".

Lightning talk #5: "Pangeo is a database" by Alexander Merose (@al.merose.com)

The difficulty of vector-at-a-time processing. > [...] needs to handle the N-ary nature of the operators. As a result, expressing complex relational operators in a vectorized modele is a challenge in itself. Marcin Zukowski, doctoral dissertation

Talk #20: "Database query compilation: our journey" by Thomas Neumann & Viktor Leis