Hacker News new | ask | show | jobs
by TheOtherHobbes 2295 days ago
I suspect everyone would love K if they had an extra 30 IQ points to work with.

The reason we have Python, Javascript, and Java is because most people (including me) just aren't bright enough to work fluently in these terse APL-alike idioms.

They fail because it's hard for most people to keep that many terse symbols and symbolic relationships in memory at the same time, without English labelling and all the other usual memory aids.

(In fact basic functional programming is a bit of a stretch for the average commercial developer.)

An interesting hypothetical is what a language would look like if it needed an extra 60 IQ points...
2 comments
rtikulit 2295 days ago
In the 70's, in high school, I taught myself APL (mostly TOPS-10 APL SF and a bit of IBM APL SV) well enough to grasp the fundamental functional approach and write real programs (and the obligatory game of life in 1 line :-)) At the time I was also using PL/I and the difference was mind expanding.

I'm quite sure I don't have an 30 extra IQ points. Mostly I was just mildly obsessed (and these days I find persistence is a workable substitute).

It is incredibly valuable to have experience across the range of programming paradigms.

link
bartread 2295 days ago
IQ points aren't really anything to do with it: I suspect most programmers, and certainly those who are average or better, could learn to be proficient with K if they invested the time.

The real issue is knowing your userbase.

Example: we have this in-house DSL that's been around for years and uses a symbolic representation not a million miles from K that we use for automating reports. The problem is it's a pain to onboard people who, whilst they're bright enough, aren't programmers and in any case have many other responsibilities to attend to. Giving them something more readable rather than arcane makes life easier for everyone.

And fundamentally we want to make it easy for everyone, not just the top quartile or the top 10%, to quickly crank out high quality automated reports across different market sectors, geos, etc., as well as to modify and maintain these reports into the future.

link