Reddit's Favorite Scientist Just Got Banned for Cheating the Site
motherboard.vice.com
motherboard.vice.com
Visibility on Reddit is essentially predicated on early votes. Get a few upvotes and even mediocre comments will often remain at the top due to momentum. Less than -4 and it's doomed to oblivion (ie. downvote the competition).
One of the most popular linked sites on reddit (quickmeme) got to where it was with only a few strategically placed votes for each link, and it took ages for them to be found out and similarly banned: http://www.dailydot.com/business/reddit-quickmeme-banned-mil...
More broadly, the algorithm reddit uses is not only known wrong/buggy:
http://technotes.iangreenleaf.com/posts/2013-12-09-reddits-e...
but also generally defective:
http://www.reddit.com/r/circlebroke/comments/vqy9y/dear_circ...
It's possible HN suffers from the same flaw (initial sort and feedback loop from such) since it's the natural (naive) way to produce such algorithms. The machine can't tell if it's quality content or just easy to upvote, and the latter is more common.
TL;DR: Visibility == early votes from people who have no interest in depth and don't read the articles anyway.
edit: fixed links
In general, the problem of path dependency based on highly-weighted initial changes can be modeled by the Polya Urn: https://en.wikipedia.org/wiki/Polya_urn_model
Over at kuro5hin (which in general should not be used as a model for successfully running an online community) rusty addressed the problem of initial votes being too influential by hiding the score of a comment until it had received 6 votes.
reddit also allows subreddit mods to hide comment scores for a predetermined period of time, if they choose to: http://www.reddit.com/r/modnews/comments/1dd0xw/moderators_n...
These solutions are good in general, but are still problematic when faced with sybil attacks / voting circles.
No, the utility of up/down votes is in retention. It's nothing more than a Skinner box that people check in on to see if they are winning or losing. Which is how Digg and Reddit both became huge in a short amount of time.
What ever system that will replace up/down voting has to do better than the ones it will replace.
It's not silly when you have a community this size. It might be flawed but it's one of the less flawed system.
> http://technotes.iangreenleaf.com/posts/2013-12-09-reddits-e....
Reading that article, I get sent to https://github.com/reddit/reddit/pull/583 which contains https://github.com/reddit/reddit/pull/583#issuecomment-35440... which notes that the issue discussed has actually been fixed.
I always assumed that's why blatant blogspam sometimes makes it to the front page of HN.
Deleted comment
Deleted comment
this doen't matters. Reddit is the bottom of the internet. Anyone cares bout 4chan or reddit.
Vote cheating was only the final straw: part of it was a silly argument between Undian and another user.
It was the blatant vote manipulation that did him in. The admins only ever ban users who break one of the five rules or make the site look bad. They should probably do a lot more...
The positive-feedback cycle stemming from early votes serves only to create an illusion of consensus, and many impressionable users begin to form thoughts like "that many upvotes can't be wrong" etc.
That, IMO, is the real problem with up/downvote systems.
The identity of the idol, and the lack thereof, can change the perception of this funny super-commenter completely.
Source: I gained internet notoriety through my comments on TechCrunch under my real name. As far as I know, I don't have any sycophantic zealots.
Not all the time - I'm sure there are plenty of counterexamples - but enough I thought it was worth pointing out.
It's worth noting that Hacker News disables downvotes on submissions older than 2 weeks.
Here's the thing. You said "vote manipulation is cheating the site."
Is it in the same family? Yes. No one's arguing that.
As someone who is a scientist who studies web forums, I am telling you, specifically, in science, no one calls multi-voting cheating. If you want to be "specific" like you said, then you shouldn't either. They're not the same thing.
If you're saying "vote manipulation family" you're referring to the taxonomic grouping of votidae, which includes things from posting quality original content to asking strangers for upvotes to running vote-bots through proxies. So your reasoning for calling multi-voting cheating is because random people "call the unusual votes cheating?" Let's get copypasta and image macros in there, then, too. Also, calling someone a human or an ape? It's not one or the other, that's not how taxonomy works. They're both. A bot-vote is a bot-vote and a type of vote manipulation. But that's not what you said. You said vote manipulation is cheating, which is not true unless you're okay with calling all members of the vote manipulation family cheating, which means you'd call good posting, bad posting, and copypasta cheating, too. Which you said you don't. It's okay to just admit you're wrong, you know?
Is there some context I'm missing here?
It's an article taking about unidan having a bunch of alts to vote himself up. Referring to that as vote manipulation is a completely acceptable practice.
Why are you being so pedandic? You must have a reason. I refuse to believe you see this as a "Terrible trend that must be stopped". Because that would be hopelessly insane.
edit:
>Is there some context I'm missing here?
answer: YES
give me a link or something, jesus -.-
http://www.reddit.com/r/AdviceAnimals/comments/2byyca/reddit...
I don't see a reason to ban someone over this behavior as long as you can make sure that the fraudulent votes don't actually add any value to the content, which they shouldn't.
Visibility of the content is essentially predicated on early votes.
The problem is that an initial sort is built on limited data (quality content or just easy to upvote produces same results and latter is more common), and the subsequent feedback loop is built on initial visibility decisions.
Why do you say that? I've worked hard on this problem and would not call it easy.
In the less general case, you can still determine that a particular group of accounts never take action in the same small time windows.
The way I've approached this problem in the past relies heavily on meta data about the user. It's easy to determine if an IP is a proxy/VPN, which should immediately make a client more suspicious (when you're already suspicious about fraudulent voting).
Initially, account ages are important but if someone is able to get away with it for a long time, they can get complacent and keep using the same accounts, and at that point you can look at things like upvote/downvote ratios (or generally the lack of diversity in voting activity) compared to some expected average activity.
There's a lot of other tricks, but I don't know what your approach has been when working on this, and I know HN is a much different platform from what I worked with :P
> Why do you say that? I've worked hard on this problem and would not call it easy.
Are there reference data sets for this problem? If not, you or reddit should publish data sets.
I suppose the fact that there's already a detection algorithm biases the data, but it'd still be better than nothing.
That takes 30 people and it requires discipline to oney the die, but it'd be pretty hard to detect. I think, maybe I'm wrong?
What makes you so sure it doesn't?
Also, isn't the reddit source public? I presume that if it had this, it'd be easy to find it there, but I haven't looked, maybe I should :)
The problem with banning fraudulent users is that they are tenacious, thats's a big factor in their committing fraud in the first place, they are usually on a "mission," see: https://en.wikipedia.org/wiki/Wikipedia:Single-purpose_accou... (not really the same as a fraudster, but this is a common archetype)
They will just make another account or increase their efforts, better to keep them in their container within the system.
At which point, they create a new account.
> while shouldn't be that negative on the system as a whole, especially if there is suspected fraud.
Assuming it isn't computationally expensive to do this on a per-vote basis. If you have to do any sophisticated analysis, that may not be true.
Reddit doing it, although some would argue unethical, was meant to make reddit even usable. It was one of those sites where it takes a critical mass to become usable, and without that critical mass it would be very challenging to actually get it. So they faked it until they had it - and it worked very well.
Someone doing it on their own only serves to benefit them - and their karma count. The goal isn't to benefit the larger community or make the site more usable - it only is to boost their ego.
> Does it really matter if the content is good?
Unidan had a lot of very good content, I cannot argue with that. However he also had a lot of really shitty content that under anyone else would have been downvoted to hell. Unfortunately his user page was taken down so I can't provide examples. But the problem is, that this bad content pushed other people's content out of view.
What's the point of trying again, he's already lost the trust of the community and his name is a red flag. Is that subreddit super-forgiving? Or maybe the turnover is so high people will just forget?
He is just trying to play the nice guy game to save his image. His doing this is dishonest -- plain and simple. Don't get me wrong. He is really good in his field and knows a lot. I learned a lot from him and if he posts stuff, I will read it again. But this behaviour is dishonesty.
If he was being honest, would it look any different? Or is the pitchfork mob just demanding blood?
On the one hand a reputation takes years to create and only a second to destroy. On the other hand, holding onto anger is like drinking poison and expecting the other person to die.
Personally I think if you want to be angry at something there's much more insidious manipulation going on. Unidan makes great content and gets people excited about his biology research... remember him for that, not for his amateur inability to execute a proper astroturfing botnet campaign.
At the very least, let him be reborn with a new username and see what he can build. Isn't the ability to bounce back from failure and mistakes what we celebrate here on hn?
I can totally understand that, I used to spend a huge times on forums, communities, etc... And I also used to have a bad temper. Of course I created fake accounts to upvote myself and downvote others. I'm pretty sure anyone who is a bit clever and also a sour loser/has a bad temper, has done that before.
And only 5 accounts? Maybe there are others, but that's really minimal. That's "unprofessional" which is a good thing and I really think things are getting blown out of proportion. Reddit wouldn't be the same without this guy, that he acted like a kid or not.
PS: What is actually sad here, is that he wasn't doing it to give to the community, but just for internet points.
Better?
That's being too generous. I would vote for RobotRollCall as Reddit's Favorite Scientist.