What I think AI will very soon get to is being able to generate code to that “functional sharpness” level, while human engineers are caught striving for that as “sharp as possible” level because they care for their craft. The cost difference is just not worth it to the organizations and the trade off will be AI generated code that will be “good enough”. Sure a talented dev could do better, but that juice just won’t be worth the squeeze. To me, embracing the tools and becoming an prompting expert is the way to go, rather than some Luddite-like rejection of the technology because of some offended artistic notions.
A solo programmer can afford to be an artisan, but commercial code is rarely artisanal. It's full of trade-offs and compromises.
Commercial code is always about business costs -- maintenance cost, buy-vs-build cost, refactoring cost, business continuity costs, etc. You're always working under constraints like time, headcount, available expertise, maintainability, and coordination bottlenecks.
Craftsmanship in commercial code is rare. Sometimes you get geniuses like John Carmack who writes inspired stuff in code that ships. But most of the time, it's just a bunch of normal people trying to solve a problem under a ton of constraints.
What makes you think that doesn’t involve craftsmanship?
Your first part may be right, however using the term Luddite as a pejorative without understanding what they meant or stood for (given the social context) is a sign of ignorance.
The commenter I was responding to was specifically speaking to code quality concerns generated by AI automation. While the modern usage of “Luddite” tends to generalize towards rejection of any new technology. When I mentioned “Luddite-like rejection of the technology because of some offended artistic notions” it was specific to the Luddite complain that the automation of would lower quality.
That might be a specific detail in the history that you may have missed, forgotten, or perhaps simply ignored—but it’s an important detail with a direct parallel in this conversation.
How about a good engineer who wants to feed their family?
My opinion, and it's just that, is the corporations are betting on the robots advancing fast enough where it won't be an issue there aren't new 'good engineers' to follow the current batch. Probably to their detriment but by the time that happens humans will be tasked to only perform the work the robots simply can't physically do so their agency in the matter will no longer matter, you either install 13.6 screws/min or you are replaced with someone (or something) who can.
Happy days...
I don’t worry about his job, good riddance
Don’t make the mistake of thinking that you are in the telegram business and not the communications business.