> This seems like it's fixing the symptom rather than the underlying issue?
This is also my experience when you haven't setup a proper system prompt to address this for everything an LLM does. Funniest PRs are the ones that "resolves" test failures by removing/commenting out the test cases, or change the assertions. Googles and Microsofts models seems more likely to do this than OpenAIs and Anthropics models, I wonder if there is some difference in their internal processes that are leaking through here?
The same PR as the quote above continues with 3 more messages before the human seemingly gives up:
> please take a look
> Your new tests aren't being run because the new file wasn't added to the csproj
> Your added tests are failing.
I can't imagine how the people who have to deal with this are feeling. It's like you have a junior developer except they don't even read what you're telling them, and have 0 agency to understand what they're actually doing.
Another PR: https://github.com/dotnet/runtime/pull/115732/files
How are people reviewing that? 90% of the page height is taken up by "Check failure", can hardly see the code/diff at all. And as a cherry on top, the unit test has a comment that say "Test expressions mentioned in the issue". This whole thing would be fucking hilarious if I didn't feel so bad for the humans who are on the other side of this.
I feel the same way today, but I got started around 2012 professionally. I wonder how much of this is just our fading optimism after seeing how shit really works behind the scenes, and how much the industry itself is responsible for it. I know we're not the only two people feeling this way either, but it seems all of us have different timescales from when it turned from "enjoyable" to "get me out of here".
Then one day I woke up and realized the ones paying me were also the ones using it to run over or do circles around everyone else not equipped with a bicycle yet; and were colluding to make crippled bicycles that'd never liberate the masses as much as they themselves had been previously liberated; bicycles designed to monitor, or to undermine their owner, or more disgustingly, their "licensee".
So I'm not doing it anymore. I'm not going to continue making deliberately crippled, overly complex, legally encumbered bicycles for the mind, purely intended as subjects for ARR extraction.
We should not forget that on the other side of this issue are equally smart and motivated people and they too are aware of the power dynamics involved. For example, the phenomena of younger programmers poo pooing experienced engineers was a completely new valuation paradigm pushed by interested parties at some point around the dotcom bubble.
Doctors with n years in the OR will not take shit from some intern that just came out of school. But we were placed in that situation at some point after '00. So the fundamental issue is that there is an (engineered imho) generational divide, and coupled with age discrimination in hiring (again due to interested parties' incentives) has a created a situation where one side is accumiliating generational wealth and power and the other side (us developers) are divided by age and the ones with the most skin in the game are naive youngsters who have no clue and have been taught to hate on "millenials" and "old timers" etc.
> Because other professional fields have not been subjected to a long running effort to commoditize software engineers.
> And further, most other (cognitive) professionals are not subject to 'age shaming' and discounting of experience.
When young people learn that I am a software developer, their eyes light up (thinking that I make huge money working for FAANG). Frequently, they ask if they should also become a software developer. I tell them no, because this industry requires constant self-learning that is very hard to sustain after 40. Then, you become a target for layoffs, and getting re-employed after 40 as a software dev can be very tough.