I have been on holidays this week, playing VR and preparing videos for an online Python event I co-organise.
PyDayBCN, an event we run each year in PyBCN is today, should be great. Will be great.
Christopher Dresser’s Studies in Design
A Twitter thread where I address how I classify projects in my task manager.
Another thread, where I have some circular arguments in favour of why Scrum makes sense and why tasks should be small. Don’t take it too seriously 🤣.
I love the nitty-gritty details of how database engines work. And this has another reference to the
MonetDB paper… I should read this one soon.
I’ll keep insisting: Ray and Dask are going to make a difference.
Any book by the Heath’s (or partial Heath as is this case) is a good book.
I should refactor some code in my personal Haskell projects to use State.
Rust has potential in the big data space, as Andy Grove proves here.
If you have ever wanted to understand what all these plans are, this is a good post to get you started.
This is a lot of deletion.
A good explanation of what partitioning is in Spark, how to tune it and why you should (or should not).
Sadly it may take a while for Spark to be available with Scala 3 support, and most of my Scala lately is only Spark related.
I had assumed that Photon was based on Flare, but I was wrong. This video covers enough material and shows enough code to give an idea of how Photon works, and it is fascinating.
It’s a pretty good mini-series, go watch it.
This is fascinating to read, and is giving me many ideas.
Any new charting option is a good thing to have at hand.
This is a clear example of how to use
Numba for something real. Always useful to have more tools for speed.
There are many kinds of CTOs depending on company and timelines, and in this post this is excellently addressed.
This is fascinating, particularly the part about using a memory-mapped
This was interesting and crazy to read. I’d love it if Rust had HKT though.
Pretty entertaining, even if I don’t like bicycles. Technologically, though, they are a special case as you can see here.
This is a good introduction to parser combinators in Haskell. After having used them for several personal projects, I would recommend using
Megaparsec instead of Parsec. The error messages are a lot better in Megaparsec, and it quickly pays off when working with them.
This is a horribly clickbaity title. It’s a pinhole camera, and the camera happens to be a beer can. It’s an 8 years exposure, that’s th impressive part.