Fast languages

Continuous Deployment means fast OODA Loops.

CD is for code, but what about for ideas?

For ideas in general, probably tweets.

For computational ideas, languages that let you move fast. This is different from the usual notion of ‘fast’ languages. C is fast but doesn’t allow you to move fast. For our purposes here, value programmer time more than the computer’s time.

These would be languages with a REPL, and mostly would be dynamically typed. These considerations will increase as the complexity of what you are dealing with (not complicated-ness). Static types in complicated contexts, and dynamic in complex? There is a robustness-efficiency tradeoff.

And that’s how you end up with R and Python for Data Science.

This is probably also the reason Paul Graham et al claimed the use of Lisp was a significant competitive advantage in his startup. Startups need to be able to have high tempo; this is a defining difference from large companies.

Notes mentioning this note

Trainability wins at scale
Early guns were very bad. Cavalry was better. And yet the guns won because a horse-rider is hard to make,...
Trainability wins at scale
Early guns were very bad. Cavalry was better. And yet the guns won because a horse-rider is hard to make,...

Here are all the notes in this garden, along with their links, visualized as a graph. If you don't see any nodes try zooming and panning in the grey area.