Comment by jf22
5 days ago
What kind of "data" do you want? I'm submitting twice as many PRs with twice as much complexity. Changes that would take weeks take minutes. I generated 50 new endpoints for our API in a week. Last time we added 20 it took a month.
What kind of data would you want?
I'm 100x as productive now that I eat breakfast while reciting the declaration of independence.
You really should try it. Don't give up, it works if you do it just right.
Edit: I realize this line of argument doesn't really lead anywhere. I leave you with this Carl Sagan quote:
> Extraordinary claims require extraordinary evidence
(or perhaps we can just agree that claims require evidence, and not the anecdotal kind)
This is an immature bad faith interpretation.
The thing is there are loads of people making these claims and they aren't that extraordinary.
The claims for coding productivity increases were extraordinary 12 months ago, now they are fully realized and in use every day.
Please refrain from ad hominem attacks.
5 replies →
Unless you’re being paid on a fixed bid it doesn’t matter. Someone else is reaping the rewards of your productivity not you. You’re being paid the same as your half-as-productive coworkers.
I'm a manager of software engineers. I have absolutely no intention of paying engineer B the same as half-as-productive engineer A.
Figuring out the best way to determine productivity is still a hard problem, but I think it's a category error to think that productivity gains go exclusively to the company.
If all (or even most) of the engineers on your team who were previously your equal become durably twice as productive and your productivity remains unchanged, your income prospects will go down, quite possibly to zero.
I think it matters very much if the person next to me is 2x as productive and I'm not.
Isn't this the case regardless of whether you use AI or not? Is this an argument for being less productive?