There is nothing wrong with the article (from memory, I read it some time ago), the problem is going from "there is a small minority of users who abuse user-generated content platforms" to "zomg, this platform is a horrible and encouraging all sorts of illegal things!" One does not follow from the other. The problem with the article isn't that it exposed a problem, that part was excellent journalism, the problem is also that it advocated for all sorts of companies to boycot PornHub, as well as being laced with a kind of accusatory attitude towards PornHub I don't much care for, almost as if PornHub themselves uploaded the videos.

The same problem exists on YouTube, Twitter, reddit, and even Hacker News to some degree. Of course, being a text-only platform makes the scope a lot less dramatic, but I've seen some people advocate for violence against certain groups and the like. I'm sure dang can tell you about some of the horrible things he's had to ban people for.

This is a difficult problem to solve, I don't really have a good solution for this, I'm not sure if anyone does. I do know that private companies acting solely in their own interest acting as custodians of what is and isn't allowed based on what does and doesn't generate "negative publicity" is most definitely not a good solution.

How often do we see "{YouTube,Apple,Google,...} blocked my account and is threatening by business and I don't have a good recourse"-stories on HN? Too often, and those are just the ones that manage to get publicity. There is often no good recourse to address mistakes or really get a good impartial judgement on things that might be on the border.

You might say "these are private companies and can reject the customers they want, you don't have a right to do business with them", well, fundamentally I don't really disagree with that, but a lot of these companies are monopolies or duopolies they can really have a big impact and even make basic participation incredibly hard or even impossible. Sometimes that's not a bad thing, but again, it doesn't seem to me that we really want to leave it up to private companies acting in their own interest to make these kind of decisions.

You don't seem to be talking about the article that was linked to in the comment you replied to.

The entire first paragraph was, and the rest is indeed about the larger issues at play.

I don't disagree with the article overall, but it is very sensationalized. Statements like this

> It is monetizing video compilations with titles like “Screaming Teen,” “Degraded Teen” and “Extreme Choking.” Look at a choking video and it may suggest also searching for “She Can’t Breathe.”

imply Pornhub is going out of it's way to specifically monetize rape or child videos, when the truth is they monetize every single video on their platform. If it was as easy as creating some sort of child or rape filter and de-monetizing those videos as the sentence seems to imply they should be doing, they would just remove the videos entirely, not de-monetize them.

And the bit about related suggestions sort of implies to a non-techie that Pornhub employees are generating these suggestions, and it's not the only place in the article they make this point.

The reality isn't an evil mustache-twirling Pornhub employee sitting at a desk and brainstorming "Ah yes, if they searched for ‘Young Asian’ they probably want to search for 'young tiny teen. The reality is an algorithm detecting patterns in user searches, and some of those users are sick people, and there's basically no level of human moderation of videos or search terms happening at the scale of usage Pornhub gets.

Sentences like

> Mindgeek’s moderators are charged with filtering out videos of children, but its business model profits from sex videos starring young people.

Implies that Pornhub is getting a substantial amount of it's profit from illegal porn, which I highly doubt is the case and the article provides no evidence of.

Again, I agree overall with the content of the article, and approve of Pornhub's reaction to only allow verified videos (which was suggested by the article). I'm not angry at Nicholas Kristof for sensationalizing some things, as it appears that is what it took to actually make something happen. But the article is definitely skewing towards misleading in some sections to get more reach.

The problem is real amateur porn were kicked out because many feared to get verified which means people seek other less reputable sites to upload their videos on. Sites which some don't care about if it's legal or not like pornhub did and of course people will flock to sites that provide what pornhub destroyed. So many shady porn sites will get an income boost from what pornhub did in panic. Would you consider that a win since you agree with the verified only move? I definitely don't. It was a very bad move without any consideration of the consequences. You're supposed to support good sites. What pornhub was. Now it's just filled with professional garbage. I've been on these sites and stumbled upon illegal stuff. I never did that on pornhub. Nicholas Kristof only made the problem much worse. Journalists never consider the consequences. They only care about writing hit pieces.

This all seems extremely hand-wavey to me.

> the bit about related suggestions sort of implies to a non-techie that Pornhub employees are generating these suggestions

They are, in a sense. They're not actively doing it in person but the code they wrote is generating those suggestions. Changing the code is within their power, yet they don't. If the code processing payments accidentally started charging everyone $0 you can bet it would get fixed pretty quickly. Why not fix the code that suggests rape videos?

> The reality isn't an evil mustache-twirling Pornhub employee sitting at a desk and brainstorming "Ah yes, if they searched for ‘Young Asian’ they probably want to search for 'young tiny teen. The reality is an algorithm detecting patterns in user searches, and some of those users are sick people, and there's basically no level of human moderation of videos or search terms happening at the scale of usage Pornhub gets.

"We created a feature that can be actively harmful, but there's no way for us to moderate it. Oh well!". Wouldn't it be better to just... not have that feature? Why isn't it an option to just turn it off?

> Why not fix the code that suggests rape videos?

None of those titles say rape or child in them, and the article already covered that they have a blacklist of search terms. Ever played a videogame with a censored wordlist for profanity? People will still find a way to call you a racial slur, even if it involves some really creative uses of text to get around the filter.

> Wouldn't it be better to just... not have that feature? Why isn't it an option to just turn it off?

Related searches in a search engine is a pretty core feature, and it's not like it prevents people from just coming up with related searches with their own mind and typing them in manually. And your statement sort of applies to every facet of Pornhub. Uploading user submitted videos can be actively harmful. Comments can be actively harmful.

Edit to respond to this bit:

> Changing the code is within their power, yet they don't.

I'm pretty certain it's not just taking search term and running it through a thesaurus or something like that. It likely correlates searches from users that make similar search terms. The code isn't really generating any suggestions, just aggregating correlated search terms. I don't have hard proof of this of course, but it's a very common implementation for related searches.

> Why not fix the code that suggests rape videos?

Deciding what is and isn't rape is an AI-hard problem.

Pornhub literally hasn't even let you search "rape" in years. Go ahead and search it. 0 results. This precedes the now infamous editorial. Same with "forced" and "brutal".

Consensual non-consensual (CNC) and even "rapekink" is a thing. There is nothing wrong with videos having "rape", "forced", or "brutal" in it as long as everyone performing is doing so out of their own free will.

This is a bit akin to saying that a film like reservoir dogs encourages theft, torture, and murder.

You may not like it personally, and that's perfectly reasonable, but a lot of people – men and women alike – do.

This doesn't help, just ask Winnie the Pooh. (Xi) People are very creative with language to get around filters.

So it being difficult is enough to absolve the company of responsibility?

The article uses innuendo to suggest that nonconsensual videos were very popular on the platform and that its proposed policies will improve performers' conditions. These questions are not seriously examined, but the article is focused on sensational stories and big numbers while skipping context and consequences.

https://newrepublic.com/article/160488/nick-kristof-holy-war...