I agree with your hatred of the term "coding". It rings like "typing" or having typist skills. It demeans the many years of time it takes to learn real software engineering skills, which includes proper software requirements analysis, design, testing, packaging, and deployment.
I work with a bunch of data analysts teaching them programming for data analytics and machine learning. If I use the term "software development" the managers think it is out of scope for their job ("that's something IT does, not our group"). But when I use the term "coding" the managers don't seem to mind. They need to grow up and appreciate what it takes to develop analytics software.
I like what you said "apprenticeships at body shops isn't going to create software developers, it's just going to create terrible software with obvious rot and decay as the years go own." These junior apprentices produce stuff that barely works, and it is rotting crap whose "technical debt" never disappears, and eventually requires a senior software engineer to rewrite it after it fails.
Maybe so, but can you please not post like this to HN? We're trying for thoughtful conversation here, and that means being kind and responding substantively (or not at all), even when another comment is completely wrong.
If you'd read the site guidelines and take the spirit of HN to heart, we'd be grateful. We're trying to be less hostile and a bit more interesting than Internet Default, if you know what I mean.
Yes, staying in software development, even as a trainer requires a daily and constant investment of time to advance. This is not something most people can keep up for decades on end. At 61 years old, I still maintain "a daily student schedule" to learn SOMETHING new every day with programming. I must study daily to keep advancing, or at least maintain "even position" with my peers, and not be passed by because things advance so fast in software and biotechnology.
This discussion is nonsensical. If a theoretical physicist is so over-educated that he can figure out how to parlay his skills and knowledge into a $200K+ job in Quant Phynance, insurance, or as a full-fledged Data Scientist, then he/she needs to reskill on street smarts.
It's not all about getting paid the most for some people. I think a challenge for the case of physicists that want to work in that field is that there are more of them than are jobs.
Compare to programmers (of all levels), where there is basically endless demand in most big cities, and in the top say 15 tech centers in the us probably a few 100k jobs are available in total. I'm sure Seattle has 50k jobs.
https://www.reddit.com/r/programming/comments/44bmmq/why_agi...