zlacker

>>laiysb+(OP)
Interesting that every comment has "Help improve Copilot by leaving feedback using the or buttons" suffix, yet none of the comments received any feedback, either positive or negative.

> 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.

>>diggan+L1
At least we can tell the junior developers to not submit a pull-request before they have the tests running locally.

At what point does the human developers just give up and close the PRs as "AI garbage". Keep the ones that works, then just junk the rest. I feel that at some point entertaining the machine becomes unbearable and people just stops doing it or rage close the PRs.

>>mrweas+ga
When their performance reviews stop depending upon them not doing that.

Microsoft's stock price is dependent on them proving that this is a success.

>>pydry+rb
What happens when they can't prove that and development efficiency starts falling, because developers spend 50% of their time battling copilot?

>>mrweas+ue
they'll just add more and more half-baked features

it's not as if Microsoft's share price has ever reflected the quality of their products

>>blibbl+Zi
I think it did, then they built up a most that made it very hard to turn momentum the other direction. It's turned now but it happened very slowly. Who knows if it ever falls of a cliff, all I know is that moats are only broken when momentum is going in the wrong direction and so they are certainly more vulnerable now than they would otherwise have been if they hadn't pissed so many people off about their products.

At one point, their desktop user experience was actually pretty good. And that was all their products back then. They definitely didn't get to where they are now by selling products that were bad. You could make the argument that some of them were bad but they were cheap, but if price is a big aspect of what makes a product good in the eyes of the consumer at the time and nobody else is competing on price, then that isn't "bad" in the sense I'm using the word.

I don't think I'd have called them out for always making terrible products all the way through till about Windows 7. I had no major complaints about that release, cloud was in its infancy, no pushing 365 etc. After that, quality started to go downhill. To the point that I'd argue with a straight face that most major community supported Linux DEs provide an objectively better and more stable user experience for both technical and non technical users.