I'm not convinced. What if the upvoting really is representative of a large number of the users?
I don't upvote that often on reddit, but probably half of the time (at most once per day) it is for a Ron Paul link (and it's always something that I read and liked).
RP may be marginal at the national level, but that doesn't imply that he's marginal at all other levels as well. And of course, "...great new things often come from the margins, and yet the people who discover them are looked down on by everyone, including themselves.
...
But the best thing of all is when people call what you're doing inappropriate. I've been hearing this word all my life and I only recently realized that it is, in fact, the sound of the homing beacon. "Inappropriate" is the null criticism. It's merely the adjective form of "I don't like it." So that, I think, should be the highest goal for the marginal. Be inappropriate. When you hear people saying that, you're golden. And they, incidentally, are busted."
That said, news sites probably need some kind of clustering in order to ensure that there is a diversity of topics on the front page. Google news and Techmeme both do something like this, of course. There's no reason reddit couldn't.
I notice that Paul Graham hates seeing Ron Paul articles, but Paul Buchheit does not.
Maybe PG is used to seeing his name in Reddit since its beginning? Paul Buchheit does not expect any articles to be about him, so when he sees " Paul " in an article, he does not feel an instinctive reaction of curiosity followed by quick repulsion.
I'm kidding, of course, but it could definitely be a learned reaction of seeing many Paul articles to be about oneself.
> I'm not convinced. What if the upvoting really is representative of a large number of the users?
My bet is that up voting is definitely representative of a large number of the users, but that many are voting out of support for Ron Paul instead of because they read and liked the linked article. You could argue this not necessarily a problem. I guess it becomes one if it hijacks the focus of the site for the remainder (or majority?) of users.
I dunno. I'm beginning to hear Ron Paul evangelism from RL friends, mostly over FaceBook. I'm pretty sure they didn't get it from Reddit.
I suspect it's mostly because people are so fed up with everything connected to the current Washington establishment that they'll vote for anyone who sets himself up as an outsider. Ron Paul's been one of the few candidates that's consistently against the Iraq war; many people will vote for him on that basis alone, regardless of the rest of his positions.
I think it's telling that the friend who's been trying hardest to convince me to vote for RP is a pro-choice Green Party atheist. When I showed him Ron Paul's recent voting record (which included a bill to eliminate all federal laws on abortion, a bill to privatize a large federal wildlife refuge, and a bill to prevent laws on school prayer), he said "It doesn't matter. He's against the Iraq war, and that outweighs everything else." He's willing to set aside every other political belief he has for that one issue. I suspect a decent-sized portion of the American electorate is the same.
No. It's supposed to point out that nothing you've said refutes what Paul said. Ron Paul's supporters may be zealots and yet still be representative of a large number of voters.
I've been a Ron Paul zealot since years before Reddit existed, but I'm not one of the people voting up the Ron Paul stories. (It's a rarity that I use my up arrow on Reddit any more at all.) My one data point does not make a trend, but I think it's pretty plausible that the sort of people who support Ron Paul are just the same sort of people who find their way over to Reddit. Ron Paul is a libertarian, and it's nothing unique to Reddit that tech-oriented internet forums tend to be populated disproportionately by libertarians.
The disagree&downvote pattern on comments is definitely a problem, but I think it's a different problem and it seems to happen on a wide variety of stories. (which isn't to say that there's not zealotry involved, just that I don't think that's the only thing driving the RP stories)
It could also be argued that your comment was completely off topic for the link, kind of like the guy who always complains about the xkcd links on reddit. Would you argue that xkcd is driven by zealotry, or are there just a lot of reddit users who who actually like seeing xkcd? (and again, I'm one of the people who occasionally upvotes the xkcd comics)
I think zealous voting is indeed representative of the current userbase:
For a long time the quality and diversity of links in Reddit was quite stable, reflecting the strong core userbase. Major systemic issues [1] led to more junk-food like submissions edging in, and slowly the forces involved drove out the original core audience, long-september style. This process continued gaining momentum, until fairly quickly lurching to the new stable state we find it in today. A similar process can be observed in wikipedia: those who care the most, have the most time, and the most tolerance for internet drama naturally dominate: the young and the passionate.
1. a) Reddit conflated two distinct actions into a single vote: training the recommendation engine (vote on what you like/dislike) and judging quality (what is good/bad). This lead to a culture of voting on taste, tending the system away from voting on quality, partly because of:
b) The nature of these systems, like the mainstream media, is to tend to the lowest common denominator as audience size increases. This should not have been surprising.
How to avoid these issues? Paul's suggestion (somewhere else on this page) about automated topic clustering is a good one. A similar approach I favor is encouraging clusters of users to form naturally, using a collaborative classification/filtering approach; even something as simple as allowing three tags per link could do this. Once clusters are established, it should be much easier to allow users to opt-in or out of the clusters they like/dislike. The editor/oracle approach and/or weighted voting should also work, but seems like a lot more effort.
[ahem. I feel like I should mention I'm somewhat biased, as I'm now working on something experimental in this domain]
The proliferation of Paul Graham stories on reddit were evidence of a design flaw in version 1 of social news.
Such a small proportion of people who visit news sites vote that a group of zealots can easily overtake a site like this. Spammers are comparatively easy to guard against; they don't care enough to create 100 sockpuppet accounts, and in any case the guys at reddit wouldn't have any qualms about killing spams that seemed to be merely for profit. This is a tougher problem.
You could say all the anti-C++ stories were already evidence of this problem. But a large proportion of programmers, even C++ users, disapprove of C++, and an even larger proportion of the kind of people who would use a social news site.
The Lisp zealots are another matter. This is more like spam; it's not just people who hate C++ so much that they reflexively upvote stories that sound critical of him; it's a deliberate attempt to market something online.
Incidentally, I'm not saying this because I have some kind of ulterior motives. I think C++ is the worst language of my lifetime. And I am not anti-Paul Graham. I know nothing about him except that every article he writes makes it onto reddit--no one writes that much gold.
I think he sees the world through his own views and everything follows.
So when he writes about anything i feel he's just writing how he views the world. He always gets upvoted because there are lots of people who like how he examines the world.
He doesn't write that much gold but how he views the world has been judged golden by a many reddit readers.
Lisp stuff practically only ever shows up on reddit; it is way overrepresented there compared to Slashdot et. al.. And it is a similar sort of thing--people want others to run into Lisp, while not actually finding the articles interesting.
I'll give it to you on your essays though, they do pop up all over (however, I will say that although I avoid Digg, Ron Paul can be found all over the place over there).
The problem with social news isn't the zealots. It's that people vote for stories they think would be good for other people to read instead of for stories they find interesting themselves.
That's why Reddit ends up having the personality of John Kerry.
(And yes, I realize that this view is not mutually exclusive with the original pg comment.)
I agree, and this also shows that you don't need any kind of hierarchy, status or conspiracy for a user-driven media outlet to gain a feeling that some kind of cabal is running things.
I mean, Reddit makes contribution (including filtering) about as easy as possible, there is no user class with special status, and not a whole lot of editorial discussion, and yet still people can complain that the output is disproportionately controlled by a minority with particular interests.
I think this is more of an enduring pattern showing up again than a particular weakness in Reddit.
It seems to be a different kind of upvoting. When I write something new it generally shows up on both reddit and del.icio.us/popular. Stuff about Ron Paul only shows up on reddit. What that implies is that people are upvoting it because they want other people to read it, rather than because they found it interesting themselves.
The reddit guys actually worried about this problem from the very beginning. That was why initially there was such an emphasis on voting as a way of training filters: if you voted for dumb stuff, you'd be punished by having to read dumb stuff. But the filter-training message never got through, perhaps because the filtering worked so badly for so long.
I don't think that is the case. Your essays are fundamentally different than the Ron Paul news stories. They have a limited shelf life. I don't really feel a need to bookmark the latest Fox News injustice towards Ron Paul or clips of his last debate or news that the s&p dropped 7% or news that Sarkozy won. I want to read about it; I think it is important; but I don't need to read about it 3 years from now.
But Ron Paul does end up on Digg and Technorati. Ron Paul posters are all over my campus. There are packed rallies and meet ups. I think that Ron Paul is important news right now. A lot of young, technically proficient people seem to agree. (These are a lot of the same people who like you by the way.)
It would make good business sense to implement some sort of clustering method. So, a great number of minority groups could happily use the site and view the ads.
But, if you were the editor of reddit, I would hope you would not censor Ron Paul's stories. The community would if it wanted to; if you search for Ron Paul on reddit, you will find that the vast majority of his stories have 0 points. This invisible hand of the reddit market would surely please Ron Paul.
edit: My god does that last sentience sound dirty...
You're 100% correct. While PG is likely genuinely dissatisfied with the current algorithms of reddit and digg, in this post, he was simply expressing helplessness about the many Ron Paul stories he has had to see on one of his favorite sites, and feels like he is responsible enough for the creation of the site to try to do something about it, but not really being able to do so.
For instance, Paul's argument did not take into account that voting up Ron Paul related articles is more about certain principles than about the man himself.
Some people like that Paul is pro-drugs. Some people like that he has the most consistent voting record. Some people like that he is libertarian. Some people like that he's fiscally responsible. That is not a single issue, unlike what PG has said in another post.
But beyond voting up unexpected political principles, readers also like to vote for articles that mention that media outlets reset or deleted polls and misrepresented their results, as well as those which criticize internet users for heavily favoring Ron Paul the most--which redditors and diggers feel is an attack on them or their intelligence.
Another reason users vote up Ron Paul stories is because users like rooting for the little guy.
Another reason is that users feel there is an injustice being done, such as that the main-stream media is not giving enough face time to all candidates.
Another reason has political roots--some users feel the other candidates aren't good enough, and regardless of what they think about Ron Paul, he has a consistent voting record.
Finally, many users vote up Ron Paul to vote for policy change, and realize there is little chance Ron Paul will actually be elected.
It has also become a self-serving prophecy in the main stream media that internet users heavily prefer Ron Paul and are responsible for online and text messaging votes, so those who use the internet think that's who they should vote for.
There are many, many reasons why Ron Paul is being selected. When articles with his name come up, there are a variety of reasons to vote them up. Those may not be good reasons to vote up those articles, but they're genuine, and I doubt there's a conspiracy.
I don't think anybody (reasonable) believes there's a conspiracy. PG's original point is that a large number of users seem to be voting for Ron Paul stories not because they find the stories interesting and want to read more like them, but because they want to help Ron Paul's campaign or express a political opinion.
The intent of reddit is that users vote for posts they find interesting and against posts they find boring. Any situation that leads to posts gaining or losing points based on other criteria is a sign of a flaw in the system.
I'm not sure what the solution is. I don't find all Ron Paul articles boring, but I only need to see one article about how Fox News tried to make him look bad, not 30. Some sort of grouping might help. Once they have tags, I think it would help to create a flagging system somewhat similar to Craigslist that would allow users to suggest tags for posts. Once a certain number of users suggest a tag, the tag is added to the post. That way, users could filter out tags they didn't want to see, and override anybody who tries to game the system.
"The intent of reddit is that users vote for posts they find interesting and against posts they find boring. Any situation that leads to posts gaining or losing points based on other criteria is a sign of a flaw in the system."
No, the intent of reddit was to get as many users as possible, whoever they are--even if they like Ron Paul stories and pictures of cats--and then sell it for a profit or go public. It's both the most logical conclusion and also the advice of Paul Graham, their investor and advisor. Is Reddit (and digg) not more popular than ever, and did Reddit not get bought? There is no stipulation in Paul's articles or speeches about making something the perfect x, y, or z, but about making something users want.
Paul says that News.YC is better, but it's because it has a much tighter focus with the addition of moderation and the aspect of sharing news and ideas with real people about real issues. It's also because those who have a lot of free time on their hands and want to go where many users are, have the choice of Reddit or Digg. If this site had been the first social news site, it, too, would have deteriorated quickly.
In theory it seems like del.icio.us should have much better content for this reason. But in practice making the front page of Reddit is good for 5-10k pageviews, whereas making the front page of del.icio.us is good for 1,000 at most. This raises the question as to whether the zealotry actually helps make Reddit more popular, or whether the low traffic of del.icio.us is due to bad design or some other factor.
I've thought a lot about this. The answer is that there are several senses of better. The stuff on del.icio.us/popular is "better" in the sense that people only bookmark stuff that means a lot to them. But lightweight stuff [PIC] is easier to click on and thus "better" in the sense of generating more page views.
It seems like the ideal social site would use a del.icio.us-like mechanism for promoting resources and reference type stuff, a techmeme-like system for promoting insightful and intellectually interesting stuff (based partly on the amount of discussion), and a Reddit-like system for promoting Lightweight Digital Refreshment. Right now each social site uses one algorithm for all types of content, which is why they're all broken.
My take on switching to News.YC is that as there is no down vote, its not an "agreement" vote for me, its more that its just want I want to read. Because there is no down (unless its hidden?) my up has more value, I am more careful what I give it to then on reddit. Probably upvote half as much.
I'm not sure If I'm a counter example to this or not. I never voted on any stories on reddit. So that would make me part of the non-voting majority. But recently I felt the urge to upvote all Ron Paul stories to help his campaign.
Don't make my life difficult. I have other stuff I need to work on. Don't make me spend hours babysitting news.yc and implementing countermeasures against abuse.
We already seem to have a nice tradition of giving lots of votes to "hey, check out what I did ..." articles. I really like that - it adds to the sense of community that other people are supportive of your efforts.
So the obvious question becomes: How does version 2 of social news prevent this? How does/will news.yc deal with it?
For the sake of argument (though it may be true anyways), what if we assume the Ron Paul voting is totally organic? What if over time a critical mass of existing reddit community members decided it was important to knee-jerk upvote any Ron Paul story without reading each story? So you have a block of legitimate users using the site for something other than the expressed goal of the site (social news), presumably to the detriment of the rest of the user base.
I mentioned one way of dealing with lame submissions when we switched news.yc to Hacker News: have human editors who mark good and bad submissions, and weight people's future votes depending on which they predict. That would probably be enough.
There are other possible solutions. For example, the reddit guys tried to solve the problem of all the political crap on the frontpage by creating a subreddit for it. But of course this is what happened:
There are also mechanical solutions along the lines of the type used to defeat SEO. I.e. if a group of the same people always upvote the same stories, count their votes less. If people vote on something without reading it first, count their votes less. Etc.
The editorial solution is interesting. It's just a way to scale out the tastes of a group of editors beyond the articles they can look at themselves. If it works well then eventually the only users who would stick around would be people with similar taste to the editors.
The subreddit did not work because they did it backwards. You don't ask users you don't want to move somewhere else. You leave the bad users where they are and move yourself to a new site. Sounds familiar.
> There are other possible solutions. For example, the reddit guys tried to solve the problem of all the political crap on the frontpage by creating a subreddit for it.
That would probably have worked better if people were forced to choose a subreddit for their submissions, and stories from all subreddits got aggregated onto the main reddit page - and logged in users could choose not to see links from certain subreddits. (That's an aspect of slashdot that does work quite well.)
The idea of a popular and high quality "social news" site is fundamentally broken. Either you want a mass market or you want quality content. The mass market has an IQ of 100 and will not choose content that generally appeals to people here puzzling over "social news". Any fix that gets rid of the stuff morons like will cost you the mass market.
The metafilter people already went through all this. They started to achieve mass market popularity and quality started to go to shit. They cut off new membership and let attrition get rid of some people. Then they were happy again with their small community.
If you introduce an editorial role, well guess what, it's not "social news" anymore. Slashdot and Fark are far from new and web2.0.
This problem is nothing new. Look at reddit's sister company, Wired. Wired went from great in the 90s to mostly moronic and consumerist now. It's also a lot more popular now.
I don't think too much should be read into the recent activity on Reddit. Ron Paul has actively promoted his campaign on the internet and has been a large proponent of internet freedoms. So users' behaviour on reddit probably isn't representative of the average American but actually quite representative of the average Internet user. In other words reddit's ranking system hasn't been abused in this sense.
People have given millions of dollars to Ron Paul. That's an ungame-able vote. If Paul is getting 1-2 percent of real support among voters, then there should be no shortage of Guiliani and McCain and Romney fans to quickly downmod any stories they don't like.
But evidently, Paul is the only Republican candidate with fans on reddit. If there's a "flaw" that's the only one.
Maybe Ron Paul proliferation is due to the man's ideas and his honesty?! And his popularity on reddit and other sites is due to the fact that he resonates well with what people think? Maybe people are sick of candidates that lie to you all the time. You know... just stating the obvious.
You don't seem to grasp that we're talking about a completely orthogonal issue. Ron Paul could be the best presidential candidate in history. The problem I'm talking about is simply a structural one: because such a small proportion of visitors on social news sites vote, 100 of his vocal supporters can dictate the stories for a site with 100,000 users.
Man is politics a tar baby. I wish I'd never made that comment on reddit. Nothing personal (I mean this more about the reddit commenters than you), but it's a topic that brings out the dumbest in everyone.
"...100 of his vocal supporters can dictate the stories for a site with 100,000 users."
Only for as long as the 100,000 is reasonably happy with what the 100 is providing them. Otherwise you'd think they'd have the sense to either start downvoting, or simply stop using reddit, rendering it a site with 100 readers who all vote. More likely many of the 100,000 see reddit as their best source for all the Ron Paul news the 'mainstream' media is 'trying to hide' from them.
No, I think the real problem is that talking about politics brings out the worst in people who disagree with you (because you make them feel angry), and the only the mildly unsavory in people who don't (because you make them feel smug). How many people responded to What You Can't Say with some form of "Yes for thee, but not for me"?
Telling people an important part of their opinion-forming mechanism -- their facts, or their fact-finding, or their ability to tell "probative" from "normative" -- is just as special case that gives everyone an equal opportunity to show his ass.
I actually spent some time thinking about how you could do better than what's out there now -- provide a way to enforce good rhetorical behavior; set some rules about evidence; and especially, make prominence based on credibility or evidentiary support.
But in the end, I don't think many forum users would really prefer the hard work of keeping their opinions in line with all the available facts to the satisfaction of beating a disembodied un-person in a non-argument. And also, there are a lot of factors working against you:
Ron Paul was just a poor example for what you were trying to describe, particularly as a reddit specific phenomenon. Ron Paul appears to be "unusually" popular on reddit, digg, slashdot, youtube, myspace, and facebook (where no sockpuppeting should be possible) among other places. Could there be 100 people inflating those vote? There probably are. But Ron Paul's popularity is a more complex phenomenon and distracts from the general case.
If Ron Paul is the worst thing you found on reddit lately, you haven't been paying enough attention. The site has more or less been 'owned' by the batshit crazy fringes:
"I know there isn't widespread support for him because I rarely see references to him anywhere except sites like this."
The best measure of any election is without a doubt the gambling market. They've proven far more accurate than any poll. My favorite is http://www.intrade.com where Ron Paul is tracking at around 4% to win the Republican nod.
I don't upvote that often on reddit, but probably half of the time (at most once per day) it is for a Ron Paul link (and it's always something that I read and liked).
RP may be marginal at the national level, but that doesn't imply that he's marginal at all other levels as well. And of course, "...great new things often come from the margins, and yet the people who discover them are looked down on by everyone, including themselves. ... But the best thing of all is when people call what you're doing inappropriate. I've been hearing this word all my life and I only recently realized that it is, in fact, the sound of the homing beacon. "Inappropriate" is the null criticism. It's merely the adjective form of "I don't like it." So that, I think, should be the highest goal for the marginal. Be inappropriate. When you hear people saying that, you're golden. And they, incidentally, are busted."