Software is more like science than mathematics. Mathematical expressions are provable, scientific theories are merely demonstrable. Math is deductive. Science is empirical. Software is empirical. We demonstrate, but do not prove, correctness.

you call a library, but the framework calls you

libraries favour composition, frameworks favour inheritance

(1/2) It is much harder to teach OO programming today than it used to be because of 30 years of brainwashing: plainly wrong views of OO with grave misunderstandings that found their way into the dominant ones . You have to unteach first. Example from yesterday:

💡 Recordando #esLibre2022: 📌 Filis Futsarov: "Cómo hacer que tus tests te mantengan enfocado" 🐧📽️: w.wiki/5Yer ‼️ Abierto envío de propuestas para #esLibre2023 hasta el 15 de enero. 🔗: eslib.re/2023/propuestas

the activity of modelling can be, sometimes, more important than the model itself

“If you understand your painting beforehand, you might as well not paint it.” 🙃 ― Salvador Dali

"The value of a class-oriented system is maximally the sum of its parts. The addition of explicit information about the runtime structure and behaviour of the parts can make the value of a DCI Context greater than the sum of its parts." - Working with objects in computer and mind

youtu.be/3ka4KY7TMTU?t=…

"Decoupling" is way overused. Coupling is unavailable. If A uses B, A is coupled to B. If B changes its interface used by A, A will have to change: A->B You might alleviate the direct dependency: A->B1->B Yet, A is still coupled to B, and now B1 too. A->B is just less obvious.

Seeing the Nette Framework official documentation translated into so many different languages is simply beautiful!

|￣￣￣￣￣￣￣￣￣￣￣￣￣￣| | Self-hosting is the future. | |＿＿＿＿＿＿＿＿＿＿＿＿＿＿| \ (•◡•) / \ / —— | | |_ |_

for each desired change, make the change easy (warning: this may be hard), then make the easy change