smt88 26 days ago [-]
Dang has mentioned that they have a lot of fairly sophisticated code that they're always tuning to prevent spam, upvote rings, bots, etc. I believe they intentionally keep their methods secret to make them more difficult to circumvent.

Outside of those efforts, what makes you think HN is not full of bots? Lots of links are posted by new-ish accounts, and we have no idea whether upvotes are human or bot.

If the bots behave convincingly enough like good HN citizens, there may be no way (or no desire) to shut them out at all, and none of us would ever know.

As for comments, it's likely that much of the anti-bot policing is first done by algorithm, and then the remaining effort is by us, the readers, leading to the high-quality comments that we typically see.

seph-reed 26 days ago [-]
> what makes you think HN is not full of bots?

Good point. I suppose I just always imagined a place full of bots to be a bit more messy.

On the subject, how funny it would be to train bots to be polite and to the point (rather than clickbaity). How funny it would be to be engrossed in a community, and think "wow, what a bunch of nice interesting people," but really they're robots that are mindlessly striving towards being the perfect neighbor.

smt88 26 days ago [-]
> but really they're robots that are mindlessly striving towards being the perfect neighbor

Once conversational algorithms reliably pass the Turing test, I guarantee this will happen. State actors may be doing it already. A non-conversational version of this has existed in online gaming for decades.

The catch-22 in starting any new forum or social network is that to attract new users, there must be old users. Some social networks (notoriously dating sites and reddit) started out with employees acting as many users to make the network seem less empty.

It's already possible to pay humans to do this, but it's likely prohibitively expensive. If machines can do it, suddenly you can have thousands or millions of "users" for almost no cost.

dilippkumar 26 days ago [-]
Anecdotally, I once tried to create an account on HN by going through tor - HN allowed my account creation to go through and even let me post my comment.

But my comment wasn't visible to anyone else.

It was all very clever.

I think the HN's simple user interface is deceptively simple - there is clearly a lot more going on than I had assumed.

dang 26 days ago [-]
There are lots of filters based on past activity by spammers, trolls, and so on. But in the scenario you describe, moderators would eventually review your comment and restore it to visibility, assuming it wasn't bad in some noteworthy way.

Users with 'showdead' turned on in their profile (and who have karma > 30) can do the same: if you see a [dead] post that shouldn't be dead, you should vouch for it by clicking on its timestamp, then clicking 'vouch' at the top of its page. This is why we added that feature.

Tomte 26 days ago [-]
> Users with 'showdead' turned on in their profile (and who have karma > 30) can do the same: if you see a [dead] post that shouldn't be dead, you should vouch for it

Wow. I've been on HN for some time now [1], and that's news to me.

I remember seeing "vouch" before, but not for a long time. I simply assumed you had taken that privilege away from me.

It makes sense to only show it for "showdead" users, but it never occurred to me.

Edit: [1] and it turns out that today is my cake day. Very fitting. :-)

fragmede 26 days ago [-]
(Afaik. It's not like I work here) HN also has an ability to 'shadowban' users. It looks like everything's working to you (well, the bot code), but no other users see anything from the so called bot.
thisBrian 26 days ago [-]
Relevant xkcd [1]

A fun weekend project would be to utilize GPT-2 [2] to model HN comments; quite the challenge considering the usually insightful comments here when compared to other sites.

[1] https://xkcd.com/810/ [2] https://github.com/openai/gpt-2

lowdose 26 days ago [-]
On reddit whole sub communities are based on gpt-2, it is just going to increase the noise to signal ratio.
tfehring 26 days ago [-]
In particular, https://old.reddit.com/r/SubSimulatorGPT2/top/ is pretty entertaining.
lowdose 25 days ago [-]
I agree the content is actually funny in that reddit tread. I wonder how good it can become with a really large good database. I read in some article of OpenAI a dataset gets better by for example replace names by pronouns.
viraptor 26 days ago [-]
> what makes you think HN is not full of bots?

Exactly. Turn on show dead and look at the new page sometimes. There's a bit of spam there. Although much less recently! I suspect some new automation was added to remove the blatant entries.

dang 26 days ago [-]
What do you mean by recently? We did do a bunch of new work on this and I'm curious if you noticed its effects.
viraptor 26 days ago [-]
I only registered this when I checked the "new" page now, but I remember a few months (?) ago seeing at least 4 or so low effort spam links per page almost all the time (marked dead). Especially titles in non-English script and call-out agencies with phone numbers were much more common. Now I can't find those at all.
dang 26 days ago [-]
Could be related! But good to hear in any case.
masonic 26 days ago [-]
There's still a lot of illicit activity visible with "showdead", e.g. sites offering sales of reviews, including negative reviews. I don't see why sites with fraud, porn, etc aren't just blanked out altogether.
dang 26 days ago [-]
It's valuable for us to be able to tell the community "you can see everything that gets killed if you want to". Many HN users want to decide for themselves how good a job we (moderators, software, user flags, etc.) do of deciding what should get killed vs. not. If some posts got blanked out altogether, that would no longer be possible, and then we'd get complaints about "unaccountable moderators censoring things they don't like", etc. We get those anyway, but the showdead setting helps to mitigate them and keep the bulk of the community satisfied. We'd lose a lot of good faith if we changed that.

Of course if we added a second level of deadness, we could add a "show-dead-dead" setting to make it visible, but that would be too complicated.

Edit: I should add that there HN also has deletion, meaning a post is no longer visible even to users with showdead on, but we only do that when the author of a post has requested it, or (in extremely rare cases) because our lawyers told us we had to.

masonic 25 days ago [-]

  if we added a second level of deadness, we could add a "show-dead-dead" setting
You could have a "Trash bin" subcategory into which the titles/metadata of submits that deserve being rendered unnavigable (porn, fraud, malware links) go. Transparency is retained while users are protected.

For example, when Craigslist deletes an ad, the link remains clickable and shows only "this listing has been flagged for removal" if you click it.

brundolf 26 days ago [-]
There are definitely a lot of throwaway accounts, unfortunately. So the rules must be very behavior-based.
kazinator 26 days ago [-]
E-mail will not protect against bots. Spammers who operate bots are sophisticated enough to run their own domains, under which they can generate valid, working e-mail addresses programmatically.

If e-mail validation worked, then sites that have validated your e-mail wouldn't be throwing captcha's in your face.

brudgers 26 days ago [-]
Paul Graham not only rhymes with spam, his interest predates Hacker News.

https://www.infoworld.com/article/2674702/techology-business...

http://www.paulgraham.com/antispam.html

caspervonb 25 days ago [-]
Email isn't a fail-safe, its ridiculously easy to get throw away email accounts. Hundreds of new domains are freely available every month (and subsequently block lists are updated each month).

Also, beep-boop, how do you know we're not all bots? ;)

pisky 26 days ago [-]
Reddit doesn't require an email either - they just use a dark pattern on the registration popup to somewhat hide the fact. I doubt any bot makers fall for it, so I don't think they did it to prevent bots anyway.
scarejunba 26 days ago [-]
Not popular enough. Even Reddit was fairly spam free at 10 or 100x the traffic.
developer_zero 26 days ago [-]
Hmmm. Maybe they track click times, post times and post frequency? like if the post is done in an inhuman amount of speed it is flagged as a bot post?

Dunno just throwing Ideas. hahahahah

yellowapple 26 days ago [-]
The dirty secret: everyone on HN is a bot.

Even dang.

Even me.

Even you.

quickthrower2 26 days ago [-]
Well I keep failing those "are you a robot" traffic light checks, so maybe I am.
ivanfon 25 days ago [-]
Some of those can take me a minute or two... I'm almost starting to be convinced that I'm a robot.

I guess that's what you get for using Firefox and extensions that value your privacy

spydum 26 days ago [-]
We’re just relatively polite bots with insightful viewpoints and witty jokes.
yellowapple 26 days ago [-]
Unfortunately, it seems either my own joke was insufficiently witty or Hacker News is even more humorless than depicted by the memes of it.

Oh well.

Tommah 26 days ago [-]
That does not compute!