Remix.run Logo
0xbadcafebee 5 hours ago

New technology does not eliminate old technology or craftsmanship. It just shifts who uses it and what for.

- Power tools didn't annihilate the craftsmanship of hand-tool woodworking. Fine woodworkers are still around and making money using hand tools, as well as hobbyists. But contractors universally switched to power tools because they help them make more money with less labor/cost/time.

- A friend of mine still knits on a loom because she likes using a loom. Some people knit by hand because they like that better. Neither of them stopped just because of large automated looms.

- Blacksmiths still exist and make amazing metal crafts. That doesn't mean there isn't a huge market for machine cast or forged metal parts.

In the future there'll just be the "IDE people" and the "Agent Prompt people", both plugging away at whatever they do.

hackyhacky 5 hours ago | parent | next [-]

You give examples where crafts based on pre-industrial technology still exist. You're right, but you're proving the GP's point.

200 years ago, being a blacksmith was a viable career path. Now it's not. The use of hand tools, hand knitting, and hand forging is limited to niche, exotic, or hobbyist areas. The same could be said of making clothes by hand or developing film photographs. Coding will be relegated to the same purgatory: not completely forgotten, but considered an obsolete eccentricity. Effectively all software will be made by AI. Students will not study coding, the knowledge of our generation will be lost.

Ronsenshi 4 hours ago | parent [-]

> Effectively all software will be made by AI. Students will not study coding, the knowledge of our generation will be lost.

Given the echo chamber of HN when it comes to AI that certainly seems inevitable. The question is - who would work on novel things or further AI model improvements if it so happens that knowledge of writing software by hand disappears?

hackyhacky 3 hours ago | parent [-]

I can answer your question in two ways:

1. AI will work on AI. 2. People will work on AI, but as a rare niche, not a mainstream area of software development.

trollbridge 4 hours ago | parent | prev [-]

The examples given are using tools to do well-defined, repeatable processes. So far, despite many attempts by upper management to make software the same way, it hasn't happened, and AI doesn't appear to be any different.

I don't see a huge difference between people writing in a high-level language and people writing complex prompts.