Yesterday the preeminent socially generated news site, digg.com, nearly exploded. The way the site is supposed to work is that users submit links to stories (and web sites, photos, and videos) they think will be interesting, and others give those stories a thumbs-up (a "digg") or thumbs-down. If enough users digg a story, it will be promoted to the top of the front page, and this constantly-updated page will reflect a diversity of interesting stories from around the world.
Ideally, this system is completely controlled by Digg's users, and the only interference from "management" is to remove spam and improve the algorithm controlling the system. But things went horribly wrong yesterday when users began to notice that certain stories were inexplicably being excluded -- those which revealed a DMCA-violating code which can be used to allow Linux users to view HD-DVDs. Once users saw that they were being censored, rather than simply voted down by their peers, a full-scale rebellion occurred, and hundreds of Digg users began submitting the code in a variety of creative formats. At its worst, the entire front page of Digg was completely populated by these articles.
Digg claims it was removing the articles as a result of legal pressure from the rightsholders to the code, but eventually there were so many submissions including the code that it was impossible for Digg to censor them all. Instead of censoring, they relented, publishing the code on the Digg Blog and vowing to fight to the end: "If we lose, then what the hell, at least we died trying."
This raises an interesting question: is the social psychology behind these group news sites such that there's no way to prevent them from subverting their initial goal? Slashdot contributer Bennett Haselton thinks he's figured out a technical solution to the problem:
The problem with sites like Digg, Haselton claims, is that it actually doesn't take many votes to get a story promoted to the front page. The system can be gamed by a group of 30 or less, who simply agree to vote up any story any of them submit. Worse, online services have appeared that allow unscrupulous publishers to buy votes for their stories:
Since Digg allows any registered user to go to a story's URL and "digg it" in order to push it upward through the story-ranking system, it was inevitable that services like User/Submitter would come along, where a Digg user can pay for other users to cast votes to push their story up to the top. User/Submitter says they are currently backlogged and not taking new orders, but they say the service will return and will soon feature services for manipulating similar sites like Digg competitor reddit.
Spending just $100, a writer managed to promote a deliberately boring blog to the top of Digg's home page and garner over 35,000 hits -- a fraction of the cost of a generating the same traffic through paid advertising. Haselton's solution to the problem is quite innovative. Instead of allowing any user to vote on any story that's submitted, users would be shown several random stories from the pool of recent submissions. If a threshold of votes out of the first 100 was crossed, then stories would be promoted to the front page, and everyone could vote on them, as before. But it would be impossible for a small group to game the system because they wouldn't be allowed to vote on a single story submitted by their friends. Similarly, paid services wouldn't work because there'd be no way to find and bribe the individuals who were allowed to vote on a particular story.
On the other hand, it seems to me that methods such as this might destroy some of the "social" aspects of a social news-gathering service. If you can't get together with your friends and decide what's popular, then what's the point? Perhaps if Digg was run under the system Haselton suggests, it wouldn't be as popular as it is now. Perhaps the point of the system is to game the system, which means that these systems are inherently unsustainable. Either they'll get popular and die because of their popularity, or they'll never get popular and die because of that.
- Log in to post comments
What happened last night is an example of a user run social site working, not failing. When management tried to apply censorship the users replied en mass and made their opinion known. Friends were calling friends and waking them up in the middle of the night, diggs were happening at the rate of 40 per second at some points, and I personally watched the number of submitted stories bloom from 1800 to well over 6000 in a matter of minutes. This was organized mass social disobedience that proved beyond a shadow of a doubt that users control digg. The network certainly has its problems, but after last night I think its fair to say that those problems only exist because Digg users tolerate them.
It seems that the problem is one of scale. As Digg has become more popular, it should increase the number of votes required to promote a story. I'm sure there are simple statistical tools that can determine a moving number that provides the threshold for the right number of stories Digg wants to show each day.
What happened last night is an example of a user run social site working, not failing.
Sure--unless they do get taken down by legal action. I'm not endorsing this particular law, but if a site becomes a center for illegal activity, then it could self-destruct...
No, it definitely failed last night. I've been reading Digg for well over a year and that almost pushed me to never look at it again. It's already pretty bad at putting good news on the frontpage (I used to click most articles, now I hardly click any), but that was just really, really annoying. The HDDVD thing was old news, and going that far with it was just absurd.
a site becomes a center for illegal activity, then it could self-destruct
A similar thing could be said for Democracy in general. That doesn't mean it isn't a relatively good system.
A similar thing could be said for Democracy in general. That doesn't mean it isn't a relatively good system.
Sure, but the difference is that a democracy is run by humans. If you get to the point where someone goes to ridiculous lengths based on a literal interpretation of a law, a judge can rule that the "intent" of the law was to prohibit such behavior.
Computers can only behave the way they've been programmed to behave, and the question is, can they be programmed to find truly "popular" articles using a social model? Or will Digg and its imitators go the way of the trackback, and be spammed into obsolescence?
@Stuart Coleman "No, it definitely failed last night. I've
been reading Digg for well over a year and that almost pushed me to never look at it again."
It's not absurd for the people wanting to have a voice. People are sick of the mainstream media that is controlled by power brokers that do censorship, or never even publish topics because they are too controversial. People don't want stories from news media where the media companies decides what they think the people should read. They want to read what like minded individuals find interesting.
The ones that are truly against letting the people speak are the ones that usually stand to gain by keeping the media and content the traditional way. There are too many old cronies and rich people who control the media and many will stop at nothing to keep the power and money in their own hands.
Bennett Haselton's idea of proving other sites are also publishing information before allowing a story to be submitted is a complete joke. The #1 complaint of Digg users is that there are to many repeat stories and not enough unique ones. What if someone wanted to post a unique story but couldn't because there isn't enough traditional media sites publishing it to make it submittable? If anti-censorship people wanted to get a story up, they would post articles on 10 different web sites then submit the story anyways.
This idea would not work and people would quit the website in droves and would end up on a website somewhere else speaking their minds. The website might even be outside the territory of the United States because apparently thats what people have to do in order to find what they want, i.e. gambling, peer-to-peer sharing, uncensored stories. Is this the land of the free or the 1930's-40's Nazi Germany of burning books. It's beginning to sound more like the latter.
So much for America being the land of opportunity when all it contains is companies issuing cease-and-desists letters, suing over trivial patents, Microsoft underselling all if its low performing software to snuff out competition, and media outlets controlled by the corporations.
What a joke.
Sure, but the difference is that a democracy is run by humans.
Are you saying here that a "socially generated news site" is not run by humans?
It was really amazing last night when I went to Digg and saw complete chaos. It was actually kind of exciting to see - and if that's what the majority of users wanted, so it should be. Digg was made for the popular opinion to be expressed and that's exactly what happened, just as designed.
I can't believe this hasn't been done yet: 09-f9-11-02. . .
I don't think that word means what you think it means.
If it becomes a center for illegal activity it might *be destroyed* by a government enforcing that law. That wouldn't be "self" destructing any more than someone loudly criticizing Stalin would be *literally* committing suicide. Credit where credit's due.
And as others have said here, this isn't an example of the users subverting the site's goal, but rather, fulfilling it. They promoted something they found interesting. And the Internet, as the saying goes, treated censorship as damage, and routed around it. This may tend to piss off the censors, but really, let them take responsibility for their own acts.
Spam promotion might be more troublesome, but a negative-vote mechanism will suppress it pretty quickly if the majority of users bother to exercise it. (Assuming the real users aren't actually outnumbered by the spammers.)
Anyway, I don't think they're in much danger. It's pretty obvious that the managers can't control what appears on Digg - they tried and failed. It would take an uncommonly stupid court to try to hold them responsible for failing to achieve the impossible.
Are you saying here that a "socially generated news site" is not run by humans?
Not in the same way a democracy is. Without computers, Digg would not exist, but a democracy still would. If you can't program the computers to run Digg effectively, then Digg will necessarily decline.
As Chris says, "It's pretty obvious that the managers can't control what appears on Digg - they tried and failed."
And Chris, you may be right -- the courts may decide that it's out of Digg's hands. However, Napster tried to make that claim, and lost.
unless they do get taken down by legal action.
Taking down the site, while certainly possible, does not change the basic nature of the social group. Trying to solve the problem by killing sites is like playing whack-a-mole with devious, tech-savvy moles. Or something.
So even if Digg were to disappear, similar activities would show up somewhere else.
News coverage has failed to mention why the internet community felt so strongly about the key number. The key represents the grassroots opposition to the upcoming digital rights managements (DRM) restrictions, which are viewed as subverting consumer's rights.
What would be appropriate online resistance to a law that people dislike? How do you conduct civil disobedience in a forum like the internet?
Digg is doomed to failure for the simple fact that 90% of the front page stories are crap.
Actually, since the vast majority of the front page stories are crap, Digg is doomed to remain wildly popular.
Yes - it's called "abandoning the rule of law".
The challenge and the obligation is to write good laws. People en masse are incapable of doing intelligent things, and so the laws are not good.
The problem isn't that the systems humans create are flawed and limited, the problem is that the humans are flawed and limited. Which is why your field is so very important.
The real problem here lies in the gap in expectations. One party insists on keeping information proprietary, the other wants it open. Then DIGG steps in and screws up handling the resulting brou-ha-ha. I'm not up on the whole HD-DVD thing, but it looks to me like one group wanted to control who could use their product, but the other said "no dice".
I'm with the second group. Make it possible for people to check out what you have available, and 9 times out of 10 they'll end up buying it. Either because of the extras they can only get through purchase, or because the paid version comes in a more convenient format. (Dropping an e-reader into your bathwater may not kill you, but it'll still give you a nasty shock.)
Security through obscurity is pretty much dead beyond revival at this point. Some people just don't realize it yet.
I don't see how Digg being uncontrolled by any central authority will necessarily cause it to decline - rather the reverse, if anything. Centralization is a weakness at least as often as it is a strength. Of course it depends somewhat on what your goals are. But Digg, like the Internet itself, doesn't have any goals that require centralization - and therefore probably works better without it.
You don't need to program the computers to "run" Digg, because the users already run it. The computers just do the bookkeeping, which they are quite capable of. This means that no censorship will fly without the users' support, but that's more of a good thing than a bad thing IMO.
P.S. If there's no popular support for a law, but it gets passed anyway, in what way does the government that passed that law resemble a democracy?
You don't need to program the computers to "run" Digg, because the users already run it. The computers just do the bookkeeping, which they are quite capable of.
I can't imagine a more false statement than that. That's just completely bogus. Of course you have to program the computers. The users couldn't run anything without the computers. If the program is ineffective, then the system will collapse. The real question is whether it's possible to program the computers to give the users enough control to keep the system from self-destruction.
Perhaps all you're arguing is "yes, this is possible." But don't argue that you don't need to program the computers.
Email -- and blog commenting -- are two examples of programs that appear to be user controlled yet in fact need a great amount of central control to avoid being overwhelmed by spam. Don't believe me? Try to post a comment peppered with words like "gambling", "hold-em", and "viagra" and see if it shows up.
I agree, the "Digg Revolt" is a good example of a non- hierarchical news site working. At least Digg was originially referred to as a non-hierarchical news site, I don't think it meets that litmus test and Bennet's proposal is closer to the a non-hierarchical site. Legal jeopardy aside, and I sympathize with Kevin Rose, the Digg system "routed around the damage" in an incredibly fascinating, yet violent fashion.
-wayne
Regardless of whether Digg and sb sites are doomed to failure, I came up with a list of reasons why bloggers should avoid them in the first place. It's on Sciencetext.com it has already generated some illuminating responses.
db
Fashion Obsessed is a Retailer of Costume & Fashion Jewellery and Fashion Accessories based in Birmingham. We offer a large array of unique designs and quality products at great prices. Our range includes, fashion/casual necklaces, high end designer necklaces, earrings, bracelets, charms, fashion belts and more. Our product range is backed by thorough market research of the latest fashions and prevailing trends.
I don't see how Digg being uncontrolled by any central authority will necessarily cause it to decline - rather the reverse, if anything. Centralization is a weakness at least as often as it is a strength. Of course it depends somewhat on what your goals are. But Digg, like the Internet itself, doesn't have any goals that require centralization - and therefore probably works better without it.
thank you very nice. succesfull article