Hacker News Reader

Comment by Waterluvian | Hacker News Reader

I expect to see the entire gamut of possible reactions with a sufficient number of bot PRs. But in looking at 10 of them at random, I didn't find a single "negative response."

(I don't think ignoring it is invalid or wrong by any means, given there's so many reasons one might not engage in a timely manner, or at all, in the issues section or PRs. I don't monitor my repos issues because I just don't feel interested in supporting my code. Feel free to fork or ignore!)

View on HN

Some negative reactions:

https://github.com/mitmproxy/mitmproxy/issues/5285

https://github.com/Qiskit/qiskit-terra/issues/7981

https://github.com/beetbox/beets/issues/4340

I do think those concerns are legitimate. (I also think more tooling is a good thing!)

Thank you for sharing these links!

I looked through all three. The first isn't really a complaint because the bot acted in good faith and found an error. In the second one they complained abiout a missing unsubscribe link (reasonable) and in the third one, the author should update their code so it doesn't create a variable named path, then a non-f-string that includes "{path}". I had to stare at the author's comment that it was a false positive for quite a bit to convince myself they were right.

I will point out that in the first two issues, the repo owners also edited the initial report with something along the lines of "removed ad".

I disagree that the first isn't a complaint -- the owner stated that this behavior isn't appreciated but decided to let it slide because the issue was valid.

In the third issue, the owner also explicitly stated: "I don't think bots posting unsolicited static analysis results are a good idea." I have no opinions on whether the code should be clearer, but that doesn't change the validity of the reaction.

The bot-account's (apparently human-written) reply of "you're very welcome" to the complaint in the third issue is downright dismissive of the problem and kinda passive aggressive. While it seems that the bot did good work overall, the human(s) handling edge cases need work.

We've blocked the bot after their script malfunctioned and they opened a second issue with exactly the same text (https://github.com/mitmproxy/mitmproxy/issues/5286).

klyrs was right about the reply from me (a dev behdind Code Review Doctor) being dismissive in the issue. I apologise for that.

FWIW my reaction was classic "expectations not meeting reality": weeks of work to do (what I thought) was a mutually beneficial helpful thing. I was naively not expecting non-positive responses and was ill prepared when you raised valid concerns I had not considered.

Again, I am working on that and sorry I was passive aggressive to you.

Thank you - that explanation makes sense, apology accepted. :)

No good deed goes unpunished.

Nice response, thanks.

I do think the question of "how should bots that do static analysis work" is an important one, but in the meantime, people are gonna bot and repo managers are gonna complain.

Reply on news.ycombinator.com