20190513

2019#12 Readings of the week

Photo by Zsolt Palatinus on Unsplash
NOTE: The themes are varied, and some links below are affiliate links. Big data and data engineering, metallurgy, Rust (not the metallurgy-related rust). Expect a similar wide range in the future as well. You can check all my weekly readings by checking the tag here . You can also get these as a weekly newsletter by subscribing here.

Testing Incrementally with ZIO Environment

I’ve been reading a lot about ZIO lately, even to the point of writing a few simple thinks with the help of a more advanced Scala friend. As I’ve mentioned before, it looks solid

Medieval Africans Had a Unique Process for Purifying Gold With Glass

No plans on smelting anything soon, but this was interesting.

Rust Runtime

The proposed approaches for async-await in Rust are shaping up.

Moving from Ruby to Rust (at Deliveroo)

It should surprise no-one, except maybe hardcore rubyists that moving from Ruby to Rust yielded terrific speed-ups. Something similar should happen in Python, though.

Zero-cost futures in Rust

One of the core tenets of Rust is having zero-cost abstractions (as in, being as close to manually written). Here you can read how futures fit that goal.

A quick look at trait objects in Rust

Traits will be familiar to Scala or even Go or Java developers, but there are a few gotchas/differences you need to be aware.

Druid Design

A friend of mine has used Apache Druid before, and has been really happy with its performance, so I’ve been looking into possible use cases to road-test it.

Make your own GeoIP API (in Python)

I’ve linked before to GeoIP approaches, this one is interesting as long as you only want country-level resolution, I usually need city or better.

An in-depth look at the HBase architecture

Similarly to Druid, I’ve been checking several things to speed up some areas. Apache HBase might be one of them, and I wanted to see how it works under the hood, I love these details

🎥 ZIO Schedule

As mentioned above, ZIO is shaping up nicely, and the scheduling helpers presented here are one of the selling points, depending on what you do.

🎥 Pure functional programming in Excel

Duh... but... to be fair... there is a point in here.

🎥 Gradual typing of production applications

Incremental approach suggestions for typing in Python, at Facebook (Instagram to be precise)

🎥 Effects as Data

Everything here should sound familiar if you are anything into functional programming

Newsletter?

These weekly posts are also available as a newsletter. These days (since RSS went into limbo) most of my regular information comes from several newsletters I’m subscribed to, instead of me going directly to a blog. If this is also your case, subscribe by clicking here.
Written by Ruben Berenguel