2020#14 Readings5 minutes read | 906 words by Ruben Berenguel
This past week has been Data+AI Summit, so there are several new product announcements from Databricks.
Facebook senior software engineer interview: the only post you’ll need to read
I’ve had Programming Pearls as a “some day I should have a look” for many years, maybe it’s time.
Domain Coloring with Adaptive Contouring
If GLSL was available when I worked in fractals I would have wasted so much time during my PhD work…
Zig, Parser Combinators - and Why They’re Awesome
The introduction to parser combinators is top. If you are interested in Zig, that’s good too but I wasn’t keen on it myself.
Reflections on Five Years of Making Art Through Programming
I started “seriously” (I have been creating “graphical stuff” since I’ve known how to code, but doing regular/large amounts) a year ago to this day. I don’t think I’ll ever be as serious, and have absolutely no plans to sell any prints. Although I’ve been meaning to get a print (or a few) of my Iris for some time.
Running Visual Studio Code on Raspberry Pi OS
Just a FYI: you can run VS Code on your shiny Raspberry Pi 400s.
Next gen Spark Shuffle Architecture
I mentioned the previous post introducing this shuffle architecture. I’m looking forward to the open source implementation.
Assembling a query optimizer with Apache Calcite
This was a tougher read than expected.
🔊 Rocket Men
When I was a kid, one of my favourite books was The Right Stuff, and this is a book that brought back the sense of wonder and discovery that you could find in that classic. Kurson is no Wolfe (also, I had the audio version), but it’s thrilling, entertaining and well-structured. Very recommended, so far the best non-fiction (non-technical) book I have read/listened this year.
A theory of how developers seek information
Excellent related links here. Also worth for seeing coderribon again, I remember seeing a pseudoimplementation for emacs some time ago.
For-Else: A Weird but Useful Feature in Python
Not sure about the “breaking of nested loops” example, I think the
else construct makes it harder to understand. But the other 2, 👌.
Some Ways that PyPy uses Graphviz
I also love Graphviz. I definitely recommend you learn to use it. It will take you an afternoon and will pay off for ages.
Implicit In-order Forests: Zooming a billion trace events at 60fps
I have the strange recollection of seeing a very similar structure 3 or 4 years ago, or a blog post with aggregation diagrams that looked strangely similar. In any case, this is great.
Registering Native Spark Functions
The examples here will come handy to enable SQL analysts with custom, fancy functions.
How Mathematicians Use Homology to Make Sense of Topology
The explanation of chain complexes here made me think of Ikea instruction sets.
🔊 Face the music: A life exposed
Paul Stanley (KISS' Starchild) biography. It is pretty good, although the final chapters are pretty meh. Also he lashes pretty hard (not without reason I’d say) against Gene Simmons, Peter Criss and Ace Frehley.
Introducing Databricks Unity Catalog: Fine-grained Governance for Data and AI on the Lakehouse
This is a welcome addition to Databricks' SQL handling, table control used to be messy.
Introducing Delta Sharing: an Open Protocol for Secure Data Sharing
This week is Data+AI Summit week, which usually means new features. Sharing datasets is always hard, and this will help a lot, at least for tech-savvy data partners. Some people still require access via CSVs stored in an S3 bucket (which is already an improvement over CSVs dumped in an SFTP server…).
Simulating brush strokes with Hooke’s Law in P5JS
The end result is impressive given how easy implementing it looks. I don’t have any project right now requiring brush-ing, but I’ll keep this in mind.
Endangered Iberian lynx population jumps 10-fold
jtpio/p5-notebook: A minimal Jupyter Notebook UI for p5.js kernels running in the browser
This is neat! I have built my own harnesses to work with p5js sketches, but if I had had this I’d probably have used it instead.
🐦 “I am reliably informed that today is #NationalBiscuitDay. Here is a 🧵 of all the biscuits I’ve made so far in 2021.
The first image is impressive… and the follow-up are even better. I could not eat these cookies!
Amazon Redshift ML Is Now Generally Available – Use SQL to Create Machine Learning Models and Make Predictions from Your Data
This could come handy to quickly explore feasibility of solving a problem with ML. You basically create models from SQL directly within a Redshift connection. They can then be served via other AWS offerings.