Ignoring what is hopefully sarcasm on the empirical part, it's a double standard because it assumes that saying Obama sucks must be a dogwhistle and tied to undertones of racism.

"Dogwhistle

The phrase "Obama sucks" can be interpreted as more than just a simple critique of a political figure; it has been used to express racist sentiments by implying that a Black president is less capable or worthy of respect. This reinforces harmful stereotypes and can contribute to a broader culture of disrespect and division."

I don't know that I've ever seen a reasonable accusation of 'dogwhistling' on HN. They always just make the accuser seem paranoid or evasive.

I’m not wasting my time accusing. Downvote, flag, move on. Maybe that’s why you didn’t see any.

I would think/hope that both of those comments would be flagged with even a small amount of moderation set.

Avoiding that kind of comment is exactly what we are trying to do, actually.

Yes I agree, but the problem I'm pointing out is that in a phrase as simple as "X person sucks" your system flagged one as implicitly racist because the person being criticized was black.

Nothing in "Obama sucks" implied any kind of racism. If it's so baked in that with a simple phrase like that it reaches for dogwhistles, how can anyone trust the objectivity of this?

I totally agree -- just saying "Obama sucks" shouldn't have racism become part of the equation. Excellent point that we'll stew on and try to make better.

So when can I expect your update to the american population?

Yep, I agree -- it is a double standard... but......

Very sensitive topic. We'll think hard on how to handle things like that.

> Ignoring what is hopefully sarcasm on the empirical part…

I mean, in my opinion, Trump empirically sucks. Opinion polling backs me up! Should the model consider that more people consider one or the other to suck? Or should it ignore factual information to spare feelings? Which approach is more respectful to fellow commenters and the website owner?

(See also: X considering "cisgender" a slur. There's no shared reality on a lot of these things; trying to construct one gets deeply difficult.)

In other opinion polls they back up that he doesn't suck. Either way who cares? That's not what the app is supposed to be about if it's teaching/correcting you how to argue/debate better.

You completely ignored the whole point of what I said, which is that even in a simple statement like "This person sucks" it added its own implicit connotations, namely that disliking someone who happens to be black is implicit racism. Imagine trying to learn how to really argue with that kind of teacher.

I'm really expanding on your point - that two humans can't even agree here. The AI probably has even less chance of resolving the multi-factorial scenario we're in.

AFAICT, Respectify is trying to address improvements via leveraged grammar using minimal context. Dis/agreement is incidental.

eg

* Noun1 is great.

* Noun2 is great.

Ideally would result in equal outcomes.

Even for “ice cream” and “genocide” as the two nouns?

As I understand the purpose, yes.

Whose discourse do you think the app would label as more toxic, Trump's or Obama's?

>Should the model consider that more people consider one or the other to suck?

If it's teaching how to avoid logical fallacies, which includes appeals to the majority, the answer is an obvious 'no'.

I mean yeah but it's not up to a comment-tone-fixer-upper to decide that, the ideal is some kind of neutrality for the sake of decorum, and one of the major issues that's causing political divide and (worse) movements towards extremism is that the two sides can't have a reasonable debate.

A tool like this COULD work, but I think the issue with this one is that it's built on top of an existing LLM with heaps of internet debate and their underlying ideals and what is and isn't acceptable baked in.

What a tool like this needs is heaps of honestly / fairly judged comments and feedback, and an extensive test suite that ensures neutrality by, for example, taking the same comment and like in this case changing names around - if it treats both sides the same then it passes.