Seriously. American kids can start to learn to program at age 6. Non-English-speaking kids can do so once they're old enough to know English.
Who do you think will have the edge?
Plus, there's a ton of coding in all fields. Learning english to be a programmer is perhaps fine, but to be able to analyze data in seemingly EVERY field of endeavor? Or automate basic routine tasks?
The language issue is an issue, and one worth solving.
> Seriously. American kids can start to learn to program at age 6. Non-English-speaking kids can do so once they're old enough to know English.
I don't think that follows at all. Non-English-speaking kids can also start to learn programming at any arbitrary age. What they need to learn for that is a programming language, not English.
Sure, most programming languages are subsets of English -- but extremely limited subsets. And the meanings of the words in them are (often radically) different from the original meaning of the same word in English; I suspect English-speaking learners have only a tiny initial advantage of the words being similar, but in effect have to learn them as if they were from a foreign language anyway.
I agree with that, but what I've explained in the article is that if you don't know English is harder to access English-only documentation, resources, etc...
This can be changed by the community to be more inclusive, especially for newcomers, kids, ...
Yes, I do think we should have translations for everything.
We should have infrastructure to view code in multiple languages. Blockly games, Hour-of-Code, and parts of Microsoft MakeCode do this already. You flip a button and "on start" becomes "nar programmet starter" or "all'avvio."
Programming is easy, since code is structured, and it's easy to parse what's a variable and what's a function call. The translation problem, outside of comments and documentation, is a lot easier than natural language.
Furthermore, a lot of this stuff generalizes since the same words are used so often. Once there's a translation for "page_loaded," you can use that across many platforms.
I should be able to select a drop-down, and have the same for any program I can find on github or gitlab, and edit code in my native language.
It's a (huge) one-time investment. But once that investment is made, sustaining is cheap. A lot of translation works well enough automated (through Google or Amazon translate), and what doesn't translate well can be cleaned up by crowdsourcing. Google Translated comments are usually actually good enough.