I don’t like games or practice with pressure. That’s why I had to build a guitar fretboard practice tool, with a nod to the homework Joe Satriani gave to Steve Vai: “learn all the notes on the neck”

d-led.github.io/fret-flashca...

apps.apple.com/us/app/strin...

m.youtube.com/watch?t=1237...

BBC went from 100s of servers to 12

Insane cost savings and still able to deal with the tsunami of traffic with no risk of failure.

Think of the peace of mind it gives engineers and the business!

Talk is cheap. There's a difference between knowing something can be done and having done so. Sometimes you just have to externalize the thoughts by implementing them. Zero downtime upgrade of a server-side rendered animation of SVG.

#myelixirstatus #elixir #phoenix #erlang #otp

2017 AI will replace radiologists, 2022 well not really...
2019 AI will replace truck drivers, 2023 well not really...
2022 AI will replace software engineers, 2025 well not really...
2025 AI will cure cancer and provide endless prosperity, ...

Struggling to churn out legacy code fast enough? Deadlines, missed costs overrun? Generate legacy code at 1000x the productivity, now with a 99% test coverage! Our newest code generator generates exactly what you tell it to do at a fixed price. Skip the wait. Use your valuable time wisely.

From a rather stressful interaction elsewhere: architectural rules are nice but why not enforce them via a unit test? “Ports and Adapters” pattern written out as an #ArchUnit test github.com/d-led/ports-...

“One can always use the … library / C API”! Yes, one can, but have you done so? In several talks late Joe Armstrong said “Pony is cool”. Well, actors play well together; Why not try building a #Pony C-Node to interface it with #Erlang or #Elixir github.com/d-led/otp_po...

Some discussions create an urge to try things out before discussing or committing to architectural options. From one such occasion: communicating #Python and #Elixir processes in an OTP cluster (demo): github.com/d-led/pyrlan...
Via an unfortunately abandoned library Pyrlang: C-Nodes in Python

The current al (approximation of life) algorithms remind of other topics that cause humans to fail to have a civilized discussion. As with those, the test is easy: if al is intelligent, would you trust it to fill your taxes? If an authority finds errors, who's the perpetrator? 1/

Humanity, stop wasting electricity on useless computations, please! Fancy pics? No need because they devalue visuals. Fancy chatbots? Build a better search engine. Generating content? There’s enough good and bad content already.

Introducing... the Stupidity Manifesto! Which might not be what you think. Have a look at it, see if you want to sign up: insimpleterms.blog/the-stupidit...

Don’t confuse your 90+% accurate model with the infinitely complex world.

I never automate anything until I’ve consulted this XKCD, and at the moment I’m still firmly in that upper left corner :o).