> Because value of a project is not in the code produced. It's in the amount of battle-testing that code has seen.
No, this is a classic mistake. The value of a project is in the instantaneous utility it provides integrated over time. Some things are valuable because they are throwaway single shots that do a thing that's hard. In the past, making those was too costly for the value, but as software drops to near-zero in marginal cost they're worth doing.
e.g. the costco receipts site involves clicking through to view receipts and only 10 'view receipt' buttons per page. I might have stuck it out in the past, but instead I just described the problem to my claw-like and gave it my user/pass and it found it for me while I was dressing the baby. It wrote some Playwright code and this and that and drove the browser interactively.
That code is throwaway code but it was quite useful.
No, this is a classic mistake. The value of a project is in the instantaneous utility it provides integrated over time. Some things are valuable because they are throwaway single shots that do a thing that's hard. In the past, making those was too costly for the value, but as software drops to near-zero in marginal cost they're worth doing.
e.g. the costco receipts site involves clicking through to view receipts and only 10 'view receipt' buttons per page. I might have stuck it out in the past, but instead I just described the problem to my claw-like and gave it my user/pass and it found it for me while I was dressing the baby. It wrote some Playwright code and this and that and drove the browser interactively.
That code is throwaway code but it was quite useful.