Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Tell Hacker News: Throwaway accounts
149 points by unalone on Nov 11, 2008 | hide | past | favorite | 58 comments
The user KrisZolar seems to have a vast amount of throwaway, spam accounts, which are upvoting one another and cluttering up this thread in particular: http://news.ycombinator.com/item?id=359551.

Those users are: http://news.ycombinator.com/user?id=ScottHanson, http://news.ycombinator.com/user?id=SkylerNovak, http://news.ycombinator.com/user?id=RobertHenderson, http://news.ycombinator.com/user?id=Huxley78, http://news.ycombinator.com/user?id=Nerdlinger, http://news.ycombinator.com/user?id=Dino, http://news.ycombinator.com/user?id=AVC, http://news.ycombinator.com/user?id=Haggen, http://news.ycombinator.com/user?id=KeshRivya, http://news.ycombinator.com/user?id=KimStarr, and http://news.ycombinator.com/user?id=HugotheMongoose.

I wanted to be absolutely certain of this before making the accusation. I'm dead certain of these names, for the following reasons:

1) They follow similar patterns. Either they are very old, inactive user names, or they are extremely new. They all make small comments in other threads that, while not contributing anything useful, are upvoted slightly - to the point where they can upvote other account names.

2) Each one has a similar, one-sided look at Ayn Rand. They all react with very immediate hostility and none of them stop to make decent arguments.

3) Every single one gets incredibly hostile towards people who claim that the article KrisZolar submitted is a poorly-written article. (That article, for the record, took a similar approach to debunking Rand.)

4) They respond to one another's comments, always in the affirmative. When discussing with other people, they each respond in turn, rarely starting a long back-and-forth discussion, and their comments are all upvoted similarly to their original comments: up one or two points in thin trickles, after they've been downvoted. This despite a lack of content.

5) Several of them have deleted their posts after being responded to.

6) http://news.ycombinator.com/item?id=360063 is a comment thread in which Nerdlinger, who has not been a part of the ongoing thread, responds in the defensive as if he has been here for a while.

7) Similarities in the way that they speak and the way in which they use grammar.

8) If you open each of these accounts at once, you see a pattern in their posts: each posts one or two at a time, in a cycle, never simultaneously. There are gaps between their posts in which none of them speak whatsoever.

I flagged their posts, but I don't know exactly how to approach this case. I've never seen something like it on Hacker News. Hopefully this is the right protocol.



You're right, except there are even more accounts. I killed them all. I don't know what made the guy go on a trolling rampage after all this time, but it's stopped now.

You should just send me an email when you suspect things like this; I can verify it a lot faster.


I wasn't entirely sure what your email address was, to be honest. I guessed when I wrote that one.


I can assure you he went on a trolling rampage because arguing and karma is a game. And us geeks love to argue and play games.

I am certain there are more like him and there will be a lot more in the future.

The interesting question is can something like a Bayesian spam filter be applied to trolling?


Having an absolute karma score and leaderboard no doubt encourage the game aspect of the thing.


Yes, absolutely.

I feel a bit guilty, though, since my posting this seems to have put me on the leaderboard for the first time. Talk about karma. :-/


Wouldn't it be a good idea to silently hide the posts from everyone except the troll. Now he (or she) has been public outed he'll probably try again with a new set of accounts.


I like the idea, although if he/she has several accounts, it won't be long before he/she notices the trick.


The thing is, Hacker News has too small an audience to make gaming very worthwhile. I think that clamping down once is good enough to send the message across. I'd hope so, anyway, since it takes a lot of effort to get to the point where you can vote things up and down.


Do you have an automated system to verify an account is being used for this sort of astroturfing? If so I'm sure many of the folks here would be interested in it. I think a service that helped forums identify astroturfers could even be the basis for a YC company.

I would imagine simply grouping accounts by their ip address, creation time and set of articles they comment on should easily catch the unsophisticated guys.


I hate to repeat what's been said already, but I think this is a different case: I applaud the effort of users like you who are preventing HN from becoming a cesspool, but this is beyond the average call of the good samaritan. Bravo. I am going to start paying much more attention to the comments here to try to weed out this kind of nonsense; sadly, I did not notice this guy at all.

Thank you.

I have to ask: how much research went into this?


Aw, shucks. Thanks a ton.

Not that much, surprisingly. The guy really wasn't very subtle with it. I've messed with spoof accounts in my past - I used to have an unpopular forum and I thought it would help raise user activity - and there are a lot of things that can clue you off. In this case, his accounts weren't saying rational things. I dissed the article, and one person would tell me I was being irrational, I'd respond: another person would call me an asshole and a third person would agree and call me a jerkwad. And it all seemed focused around the one thread (I'm a big fan of Rand and Gladwell, and really liked the discussion). All I did was scroll through and look for the anti-Rand people who only had a few points of karma.

Once you have the names, it was just a matter of looking at their comments. In this thread, I said to look at the time between posts: it was a lot easier than that, because I was in class and could just refresh once every fifteen minutes. One name did the talking each cycle, and each time I'd get downvoted one and his names would all get upvoted one.

This was bizarre, though. He was incredibly aggressive about the post, and extremely blatant. On this site, that stood out a lot: usually, people who insult other people get instantly downvoted. That happened to me a lot when I first joined. Here, a bunch of people were doing that and getting upvoted. I just hope this site never becomes diluted enough that something like that fails to stand out.


I was on the verge of leaving HN for a while because of the level of comments, especially the particular story in question, with ScottHanson etc on it. You're done a fundamentally good thing, in my opinion.

To catch a thief: I think it's very revealing that the person who was able to recognize this was somebody who has spoofed accounts themselves, and who has insulted other people "a lot" on HN and therefore knew the normal response.

I kind of hope that the perpetrator also has an opportunity to redeem themselves - and who perhaps, one day, might be enabled to do something similarly good as a direct result.


The meta-question that concerns me is: are those kinds of articles attracting the sort of people that get involved in that kind of crap? That's one of the reasons I am very wary of them.


I don't know. The guy who made the spoof accounts submitted two stories about Malcolm Gladwell, and his spoofs commented on other Gladwell stories, but I don't think he was involved in the other ones. It's just that right now Outliers is the new big thing.

I've never seen something like "Rand versus Gladwell" get such a rise out of anybody, ever. Hacker News has religion-v-atheism topics that manage to stay relatively calm. The fact that this is what made somebody flip out I think is a sign more of desperate blog-spam than it is of the actual Hacker News people going berserk. Luckily this was a single person.

I think, though, that in the last few weeks Hacker News has tipped a little bit: a lot of new people are joining, and as a result, the site might be at risk of becoming diluted. Hopefully that won't carry too far (I remember reading that in Usenet, every September would be difficult as new people joined, but until AOL added the feature within a month everybody would be properly inducted).


"September 1993 will go down in net.history as the September that never ended."

http://en.wikipedia.org/wiki/Eternal_september


Yeah. But before then, Usenet could regulate itself even with the occasional surge.

This is a surge right now, in part due to the election. If the people who've been here for a while help regulate things and maintain the culture, hopefully things will all settle down.


Algorithms + "Smart People Police" make me feel safer. Nice one, unalone!


Wow, thanks for all the effort. Also, kudos for being cautious, and a good guy, about it.


Thanks for mentioning it in the original thread! I was suspicious, but had it down to paranoia before you mentioned that as well. The really old accounts made me doubt myself.


Same here. I didn't flag those, only the shiny new ones. I wish I'd been bolder now.


> I've never seen something like it on Hacker News

You may want to try and deduce the list of puppets used by Michael Arington (of TechCrunch blog). Something tells me it's going to be more impressive than this one.


TechCrunch doesn't need puppets. It has its fans. I'm not a big TC reader, but I like all the people I've met who work at TechCrunch. They're all genuinely nice folks (though Arrington has the troll gene, so sometimes he goes overboard and picks on someone smaller than him, which is rude and in poor form...but I don't mind when he picks on Google or Yahoo or MS or anyone who's accepted more than a few million in funding).


Agreed. Arrington has a huge media empire. Perhaps he trolled and spammed back when that all started (though I'm nowhere near certain of that, either), but now he's got nothing to benefit.


You don't need sockpuppets when you have sycophants!


you must have spent _hours_ investigating this guy. wow.


I know the feeling though:

"I... can't... go... to... sleep... someone... is... wrong... on... the ... internet"


Obligatory xkcd link: http://xkcd.com/386/


They all posted on the thread. I was in a 3-hour lab class, and happened to notice when they all commented after me.

After that, it only took about 20 minutes to check their posting times and write this up. The guy was really not subtle about it.


Good for unalone though and I agree with his assessment. It's quite impressive to see this level of commitment to HN.

Well done.


Impressive to see this level of commitment? Perhaps.

But this "level of commitment" can be extraordinarily unhealthy. I enabled the procrastination feature and I find myself hitting "override" every day.

sigh


Wait! Procrastination feature? What's this?

I try to take things in shifts. Some days I don't check Hacker News at all: I work on writing and design. But I usually don't write unless I'm absolutely in the mood, and so this is a perfect outlet: it's a great place for me to refine my thoughts, and to talk to people that disagree with me about things. It helps me keep an open mind. And once in a while, I read something here that provokes me to write something new. (I wrote one of my current favorite poems after clicking a link on Reddit, for instance.)


In your profile, set noprocrast to yes. maxvisit is the number of minutes you can browse HN per session. minaway is the amount of time you'll have to find something else to do.


Oh, wow. That's excellent. I wish every site had something like that.


Why is there even an override options, given that the point of using the feature would be that you have low selfcontrol?

Btw no anti-procrastionation software will ever be useful until it prevents you from changing the OS time...

HELP!


I'm guessing it's in case you set minaway to 4294967296.


Haha, I just had a look the original post you were referring to and this conversation struck me as particularly funny:

unalone: Them, and AVC, and HugotheMongoose, and KimStarr, and Huxley78. It's baffling. I've never seen something like that happen before.

Nerdlinger: So what? I signed up to post.

unalone: I didn't mention your name, so entering right now is only slightly suspicious. :-)


Awesome work, thank you for helping keep YC great.


I'm pretty sure pg said there's some anti-sockpuppet provisions in news.yc. I believe upvoting between sockpuppets is ignored.

But I agree, this seems suspicious.


Yeah there is - I've come across this (innocently, naturally) before. Not sure of the exact algorithm.


I love Hacker News. I started one of the big fights when I wrote that the guy made a flawed analogy. I leave HN for the day, come back, and all hell has broken loose with trolling and fake accounts... and then I see the guy's been wiped off the board. Awesome - thanks PG, and Unalone.


I recently had a comment mildly critical of Ayn Rand mysteriously downmodded. Maybe this is why.


No. That was me, unfortunately. I downvoted before I checked your user name, which was a big mistake. I was going to go back and reply, but I completely forgot. Sorry about that: I'll respond to you right now.


Kudos for the work!

Like DavidW, I'm interested in the meta-issues involved here: how many more fakers are out there? Are we attracting this kind of activity in some way? If we can block spam with code, can we block this activity with code?



Have you considered a career as a private detective?


Looks like solid evidence to me, good work.


email pg...


I wrote to pg@ycombinator.com. Hopefully that's the correct address.


It is. (btw, impressive effort)


I hope he has a good spam filter... I know pg pioneered bayesian spam-filtering, but I wonder how well it copes with publicly known email address?

I tend to assume Gmail has the best spam filtering (with their massive database of email), but I still get about 7 per day that it's not sure of, and my account name isn't very publicized.


It would cope brilliantly.

Because PG gets that spam is a personal definition (http://www.paulgraham.com/better.html - last two par's before notes). Running a spam filter at the level of a company, or several companies, will never work properly. Ours doesn't, despite running several popular anti-spam engines together.

It's good compared to not running any filter, but it can't be tuned with enough granularity, and it doesn't learn.

It's a combination of "crummy but available is good enough", failure to believe/see that it could be better, and Yudkowsky's "not holding ourselves to a high enough standard".

GMail could do it, since they have access to not only all your mail, but also the possibility to identify "mail arriving at x many accounts system wide".


I get a relatively high number of false positives, but I admit that my situation is probably an outlier. I get error reports, mailing list moderation messages, and other stuff that must be hard to filter on.


Do you click the report spam button?


The 7 per day that it's not sure of are already in the spam folder. On the rare occasion one gets to my inbox, yes I click the report spam button.


So it works... your spam goes to the spam folder.


Yes, it does. I'm a bit confused though - I wasn't in question about that.

Let me clarify: I was wondering if pg's bayesian filter works so well that publishing his email address won't cause a bump in spam for him - since he seems to hide it (as I do mine). I was using Gmail as one example of bayesian filtering.

Mainly, I was gently suggesting it's better to not publish anyone's address here where it will get harvested.

BTW: When I think about it though, Gmail didn't work as well for me in the past.. but I don't know the cause of the improvement: it could be because I've been a lot more careful since then; or because Gmail has improved in general; or it's learnt for me specifically. I think a big factor is my old uni accounts stopped being diverted to it, and they used to catch a heap of spam. That's probably it.


hey unalone, i take it THIS is the site you were referring to on reddit earlier today ;)


Yessir. I didn't exactly know this would be happening when I mentioned it, though.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: