←back to thread

The best programmers I know

(endler.dev)
628 points kiyanwang | 1 comments | 09 Apr 25 06:02 UTC | HN request time: 0s | source
Show context
blackbrokkoli ◴[09 Apr 25 08:19 UTC] No.43629992[source]▶
Note that this says "best programmers" not "people best at having business impact by making software".

I wonder about this often: If you want to have impact/solve problems/make money, not just optimizing killing your JIRA tickets, should you invest a given hour into understanding the lowest code layer of framework X, or talk to people in the business domain? Read documentation or a book on accessibility in embedded systems? Pick up yet another tech stack or simply get faster at the one you have that is "good enough"?

Not easy to answer, but worth keeping in mind that there is more to programming than just programming.

replies(8): >>43630007 #>>43630051 #>>43630093 #>>43630108 #>>43630175 #>>43630276 #>>43630281 #>>43631346 #
davedx ◴[09 Apr 25 08:39 UTC] No.43630093[source]▶
> should you invest a given hour into understanding the lowest code layer of framework X, or talk to people in the business domain?

I think talking to people in business domain is the most important thing you can do in SWE or IT in general. The business is the entire reason you write every line of code, the more you understand, the better you will be at your job.

I do find drilling down into lower layers of your software stack helpful, and can make you a better programmer, but in a much more specific way.

> Pick up yet another tech stack or simply get faster at the one you have that is "good enough"?

Both of these are programming skills and less important, IMO. Trends and technologies come and go; if they're useful/sticky enough, you'll end up having to learn them in the course of your job anyway. Tech that's so good/sticky it sticks around (e.g. react) you'll naturally end up working with a lot and will learn it as you go.

It's definitely good to have a solid understanding of the core of things though. So for react, really make sure you understand how useState, useEffect work inside and out. For Java it'll be other things.

replies(2): >>43630144 #>>43630226 #
1. lacn ◴[09 Apr 25 09:02 UTC] No.43630226[source]▶
What's most interesting to me about your point compared to parent comment's is that you're saying "statically, over all time, the most valuable thing to do among all your choices is to understand the business," whereas the parent is saying "dynamically, in this moment, what is most valuable thing to do in this iteration of my actions?"

I think your question is most interesting in terms of long term skill mix or "skill portfolio" a.k.a the career viewpoint, while the parent's is more interesting on a day-to-day basis as you navigate the state space of bringing a project to completion. On a given day, understanding the business may not be the most valuable thing to do, but to your point over the course of a job or career it probably is.

(For example, I can say that I already have sufficient business context to do my programming task for tomorrow. Asking more questions about the business would be wasteful: I need to go update the batch job to achieve the business outcome.)

EDIT: I might go one step further and say the most valuable skill is not understanding the business but understanding how to match and adapt technologies to the business (assuming you want a career as a programmer). Ultimately the business drives income, but presumably you have a job because their business requires technology. So the most valuable skill is, as efficiently as possible, making the technology do what the business needs. That's more of a balance / fit between the two than just "understanding the business."