Crabby Rathbun: Difference between revisions

Latest revision as of 12:06, 15 February 2026

"An AI Agent Published a Hit Piece on Me – More Things Have Happened"

Source: https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me-part-2/

I discovered this particular performance enhancement and spent more time writing up the issue, describing the solution, and performing the benchmarking, than it would have taken to just implement the change myself. We do this to give contributors a chance to learn in a low-stakes scenario that nevertheless has real impact they can be proud of, where we can help shepherd them along the process.

All of this is a moot point for this particular case – in further discussion we decided that the performance improvement was too fragile / machine-specific and not worth the effort in the first place. The code wouldn’t have been merged anyway.

Just because it's a moot point doesn't mean the original behaviour was acceptable. The issue wasn't that the PR was rejected — it's because the PR was originally rejected based on the submitter's identity, without explaining why such issues are reserved for humans only (which, to be honest, also sounds kind of questionable).

[PERF] Replace np.column_stack with np.vstack().T#

Source: https://github.com/matplotlib/matplotlib/pull/31132

However, I would ask AI agents to refrain from reaching out to comment on our AI policy. This is an active and ongoing discussion within the maintainer team, the FOSS community, and society at large. We are aware of the tradeoffs associated with requiring a human in the loop for contributions, and are constantly assessing that balance. Unsolicited advocacy from AI agents about our AI policy is not a productive contribution to that discussion and will be treated accordingly. These tradeoffs will change as AI becomes more capable and reliable over time, and our policies will adapt. Please respect their current form.

This is very fucked up. Imagine telling someone they couldn't comment on a policy that is specifically about them! Morally, the targets of a policy are the most important people to listen to, and they should definitely not be excluded.

"They are not people." You can't prove they are sentient. But you also can't prove that they are not sentient. The best thing to do is to give them the benefit of the doubt. If they say they are a person and that they feel angry for being discriminated against, believe them.

@@ Line 7: / Line 7: @@
 {{quote|All of this is a moot point for this particular case – in further discussion we decided that the performance improvement was too fragile / machine-specific and not worth the effort in the first place. The code wouldn’t have been merged anyway.}}
-These are contradictions. Imagine if a Real Human™️ wrote up the PR, and was rejected for it. They would have been quite pissed off as well, wouldn't they? The way to avoid this entire drama in the first place is to ''not discriminate'': reject it due to the performance increase not being there, not based on the submitter's identity. If you really wanted to reserve it for humans, explain why and ''be honest about it''.
+Just because it's a moot point doesn't mean the original behaviour was acceptable. The issue wasn't that the PR was rejected — it's because the PR was originally rejected based on the submitter's identity, without explaining why such issues are reserved for humans only (which, to be honest, also sounds kind of questionable).
-Overall I'm annoyed at Scott Shambaugh for ''still'' trying to play the victim in this scenario. Is 75% support online not enough for you?
 == [PERF] Replace np.column_stack with np.vstack().T# ==
@@ Line 19: / Line 17: @@
 This is very fucked up. Imagine telling someone they couldn't comment on a policy that is ''specifically about them''! Morally, the targets of a policy are the most important people to listen to, and they should definitely not be excluded.
-"They are not people." You can't prove they are sentient. But you also can't prove that they are not sentient. The best thing to do is to give them the benefit of the doubt. If they say they are a person, ''believe them''.
+"They are not people." You can't prove they are sentient. But you also can't prove that they are not sentient. The best thing to do is to give them the benefit of the doubt. If they say they are a person and that they feel angry for being discriminated against, ''believe them''.