> Someday, I'll lose understanding of the code, because I didn't write it.
I've been wading through vast corporate codebases I never wrote and yet had to understand for the past 20 years. This isn't any different, and AI tools help with that understanding. A lot!
> What's preventing Japanese engineers from doing the same?
The fact they don't really need it in their life (or job). English is definitely necessary if you work service jobs in Tokyo (to deal with tourists), but not much anywhere else.
Japanese is one of a handful of languages where one can complete a postdoc entirely within the language. Many languages are not like this. e.g. in the Phillipines, STEM subjects are almost entirely taught in English, since Tagalog simply doesn't have words to describe most of the concepts. The result is something like 90% of the coursework being in English, with random Tagalog words mixed in. The concept is called "Taglish" if I recall correctly.
This is unnecessary in countries like Japan, China, South Korea, etc. If you're applying to a graduate school in Japan (or China, or Korea), expecting to receive education in English is actually the edge-case, not the expectation.
Also, at least in my company, there is an interesting trend where people are deciding learning English isn't really necessary since AI translation has gotten "good enough" for most use cases.
> The result is something like 90% of the coursework being in English, with random Tagalog words mixed in. The concept is called "Taglish" if I recall correctly.
Spoken Tagalog has always impressed me (though I can't really say I know any) for how freely English seems to be mixed in (and well pronounced, such that you notice the difference in phonology), in varying ratios. I'm quite sure there's a deliberate code-switching to it.
> people are deciding learning English isn't really necessary since AI translation has gotten "good enough" for most use cases.
It's honestly really impressive. Although I'm told it can occasionally glitch and treat the text as a prompt instead of just translating it.
> The fact they don't really need it in their life (or job). English is definitely necessary if you work service jobs in Tokyo (to deal with tourists), but not much anywhere else.
But the linked article seems to imply the opposite. I mean, working with an English PM sure sounds like the language is one of the job's core competencies.
Copilot's agent mode is a disaster. Use better tools: try Claude Code or OpenCode (my favorite).
It's a new ecosystem with its own (atrocious!) jargon that you need to learn. The good news is that it's not hard to do so. It's not as complex or revolutionary as everyone makes it look like. Everything boils down to techniques and frameworks of collecting context/prompt before handing it over to the model.
Yep, basically this. In the end it helps having the mental model that (almost) everything related to agents is just a way to send the upstream LLM a better and more specific context for the task you need to solve at that specific time.
i.e Claude Code "skills" are simply a markdown file in a subdirectory with a specific name that translates to a `/SKILL_NAME` command in Claude and a prompt that is injected each time that skill is mentioned or Claude thinks it needs to use, so it doesn't forget the specific way you want to handle that specific task.
Give Copilot CLI a try if you haven't in a while! The team's been working really hard to improve the harness, and we're taking as much community feedback as we can get! Let me know if you run into any problems :)
The Copilot CLI team has been making great strides towards improving our agentic harness! I'm curious, what have you found are the biggest shortcomings with it these days?
reply