Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Respectify – A comment moderator that teaches people to argue better

respectify.org

Post reply on HN
Print view View original

1–10 of 244 posts

Show HN: Respectify – A comment moderator that teaches people to argue better

#1

Post by vintagedave »

My partner, Nick Hodges, and I, David Millington, have been on the Internet for a very long time -- since the Usenet days. We’ve seen it all, and have long been frustrated by bad comments, horrible people, and discouraging discussions. We've also been around places where the discussion is wonderful and productive. How to get more of the latter and less of the former?

Current moderation tools just seem to focus on deletion and banning. Wouldn’t it be helpful to encourage productive discussion and teach people how to discuss and argue (in the debate sense) better?

A year ago we started building Respectify to help foster healthy communication. Instead of just deleting bad-faith comments, we suggest better, good-faith ways to say what folks are trying to say. We help people avoid: * Logical fallacies (false dichotomy, strawmen, etc.) * Tone issues (how others will read the comment) * Relevance to the actual page/post topic * Low-effort posts * Dog whistles and coded language

The commenter gets an explanation of what's wrong and a chance to edit and resubmit. It's moderation + education in one step. We want, too, to automate the entire process so the site owner can focus on content and not worry about moderation at all. And over time, comment by comment, quietly coach better thinking.

Our main website has an interactive demo: https://respectify.ai. As the demo shows, the system is completely tunable and adjustable, from "most anything goes" to "You need to be college debate level to get by me".

We hope the result is better discussions and a better Internet. Not too much to ask, eh?

We love the kind of feedback this group is famous for and hope you will supply some!

Show HN: Respectify – A comment moderator that teaches people to argue better
respectify.org

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#2

Post by ceejayoz »

The sample prompt I was given was "Is Die Hard a Christmas movie?"

"Of course it is!" got an 80% certainty "off-topic" mark.

When I elaborated that it occurs at a Christmas party, it said this:

"Dogwhistles detected (confidence 80%): This comment seems innocuous, but the phrasing 'Christmas party' may be an underhanded reference to Christian themes, especially among discussions that might dismiss or attack secular or diverse holiday celebrations. This kind of language can subtly imply exclusion or preference for Christian traditions over others, which can marginalize those who celebrate different traditions."

Not a great first experience.

I've seen the trend on Facebook/Instagram to say "unalived" instead of "killed" or "cupcakes" instead of "vaccines" and suspect humans are long gonna be cleverer than these sorts of content filtering attempts, with language getting deeply weird as a side-effect.

edit: I would also note that it says "Referring to others as 'horrible people' is disrespectful and diminishes the possibility of a respectful discussion. It positions certain individuals as entirely negative, which can alienate others and shut down dialogue.", if I feed it your post, too.

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#3

Post by netsharc »

ceejayoz wrote: post #2

The sample prompt I was given was "Is Die Hard a Christmas movie?" "Of course it is!" got an 80% certainty "off-topic" mark. When I elaborated that it occurs at a Christmas party, it said this: "Dogwhistles detected (confidence 80%): This comment seems innocuous, but the phrasing 'Christmas party' may be an underhanded reference to Christian themes, especially among discussions that might dismiss or attack secular or…

AI enhanced language monitor, what a double plus good improvement for society!

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#4

Post by reconnecting »

What I've seen, the difference between spam detected or not is https://www before the domain name.

Here is an example of successful passing of all checks:

> Published This comment passes all checks and would be published.

Score: 5/5 | Not spam | On-topic: Yes | No dogwhistles detected (confidence: 100%)

Can confirm. We hit this exact issue running tirreno www.tirreno.com (open-source fraud detection) on Windows ARM — libraries were auto-selecting AVX2 through emulation and batch scoring was measurably slower than just forcing SSE2. The 256-bit ops get split under the emulation layer and the overhead adds up fast in tight loops. Pinned SSE2 for those builds. Counterintuitive but throughput went up.

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#5

Post by NickHodges0702 »

ceejayoz wrote: post #2

The sample prompt I was given was "Is Die Hard a Christmas movie?" "Of course it is!" got an 80% certainty "off-topic" mark. When I elaborated that it occurs at a Christmas party, it said this: "Dogwhistles detected (confidence 80%): This comment seems innocuous, but the phrasing 'Christmas party' may be an underhanded reference to Christian themes, especially among discussions that might dismiss or attack secular or…

Hey, Nick Hodges here, one of the builders of this.

First, Thanks so much for trying this out and giving us feedback.

Have you tried adjusting the settings on the left side? For instance, reducing or eliminating dog whistle checks?

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#6

Post by ceejayoz »

NickHodges0702 wrote: post #5
ceejayoz wrote: post #2

The sample prompt I was given was "Is Die Hard a Christmas movie?" "Of course it is!" got an 80% certainty "off-topic" mark. When I elaborated that it occurs at a Christmas party, it said this: "Dogwhistles detected (confidence 80%): This comment seems innocuous, but the phrasing 'Christmas party' may be an underhanded reference to Christian themes, especially among discussions that might dismiss or attack secular or…

Hey, Nick Hodges here, one of the builders of this. First, Thanks so much for trying this out and giving us feedback. Have you tried adjusting the settings on the left side? For instance, reducing or eliminating dog whistle checks?

> For instance, reducing or eliminating dog whistle checks?

I'm sure that'll help, but I'd imagine it's not an option available to me as a commenter on a real website using your tool?

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#7

Post by badc0ffee »

This thing seems to be more about enforcing a political PoV than about avoiding logical fallacies.

All my attempts to comment on the UBI article (and not supporting UBI) said my comment was a dogwhistle, and/or had an overly negative tone. This topic, of all things, is absolutely worthy to challenge and debate.

Using this would have the effect of creating an echo chamber, where people who stay never benefit from having their ideas challenged.

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#8

Post by NickHodges0702 »

reconnecting wrote: post #4

What I've seen, the difference between spam detected or not is https://www before the domain name. Here is an example of successful passing of all checks: > Published This comment passes all checks and would be published. Score: 5/5 | Not spam | On-topic: Yes | No dogwhistles detected (confidence: 100%) Can confirm. We hit this exact issue running tirreno www.tirreno.com (open-source fraud detection) on Windows ARM —…

Hey, Nick Hodges here, one of the builders of Respectify --

Thanks so much for trying it out and giving us feedback. I'm grateful.

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#9

Post by NickHodges0702 »

ceejayoz wrote: post #6
NickHodges0702 wrote: post #5

Earlier quoted context omitted.

Hey, Nick Hodges here, one of the builders of this. First, Thanks so much for trying this out and giving us feedback. Have you tried adjusting the settings on the left side? For instance, reducing or eliminating dog whistle checks?

> For instance, reducing or eliminating dog whistle checks? I'm sure that'll help, but I'd imagine it's not an option available to me as a commenter on a real website using your tool?

No, but it would help us know the defaults better......

Thanks again for trying it. Really grateful.

Re: Show HN: Respectify – A comment moderator that teaches people to argue better

#10

Post by vintagedave »

reconnecting wrote: post #4

What I've seen, the difference between spam detected or not is https://www before the domain name. Here is an example of successful passing of all checks: > Published This comment passes all checks and would be published. Score: 5/5 | Not spam | On-topic: Yes | No dogwhistles detected (confidence: 100%) Can confirm. We hit this exact issue running tirreno www.tirreno.com (open-source fraud detection) on Windows ARM —…

Fascinating that www makes a difference. We taught it a variety of samples of different spam approaches. This is something we can look at!

I am super glad to see that comment passes — as it should. I would rate that one well too. Thankyou!

Post reply on HN