Rubén Fernández

In my view, advances in medical science feel useless as long as people can't afford medical care in the first place. The easiest way to improve public health is not cutting edge medical research, but simple measures for affordable care.

Time travelling monad? Pfft, we have a time travelling OS!

I've used it to painlessly try out packages from the unstable nix version (or the next stable version, before I'm ready to do a full upgrade), or even actually instantiate them (without the full upgrade). I've been calling it "borrowing packages from the future".

Parse, don’t validate https://lexi-lambda.github.io/blog/2019/11/05/parse-don-t-validate … by @lexi_lambda

A tutorial also forces you to understand your library from the perspective of a new user. Hard to explain? Then it's probably hard to use! For me this is a great forcing function to reconsider the APIhttps://twitter.com/GabrielG439/status/1185947474331983873 …

Writing a tutorial tends to improve the API because developers usually prefer to spend days coding (to simplify the API) rather than spend hours writing (to improve the tutorial)

RT if you support unions.

Had a similar feeling when I used a garbage collected language for the first time - the liberating feeling of being able to just forget about a piece of data once I am done with it, not having to worry about disposing of it.

When I use a good type-safe language, the compiler often feels like I have a personal assistant - taking care of much the drudgery I'd otherwise would have to pay attention to myself, taking away the focus from tinkering creatively.

If I give another talk, I wanna make it about dispelling the myth that type systems and compilers constrain your creativity and prevent tinkering and force you down some narrow path My compiler is *what lets me* play fast and loose and try different ideas with fast feedback!

From "Gender is not a boolean but a float" to "Gender is not a partially ordered set but a commutative ring"

Hey #ActivityPub & #DistSys twitter, I'm looking for resources, papers, articles, on the subject of building (little to middle-scale) federated systems. Concerns like moderation, propagation, stuff like that. What would you recommend? (also if you could RT that would be)

I'm starting to suspect that, the more natural and inevitable something seems to experienced programmers, the more it is anything but.

Screw the concept of the 10x engineer. Instead, cherish people who genuinely enjoy explaining concepts/reasoning and sharing knowledge with their team mates. This inevitably helps everyone around them (and themselves) grow while building a strong collaborative team culture