Fub: To state it somewhat anthropomorphically, the internet interprets impedance to flow of any information as damage (whether true information or false information), and routes around the outage.

OTOH, sometimes it routes that outage around ;-)

Kamal:
Without a massive restructuring of the internet hosts logging access still will not provide an accurate way to identify users. See TOR

Correct. Removing CDA 230 would just mean that there’s someone to sue (the web host) for inadequate policing. It doesn’t necessarily allow one to get to the original poster.

David Thompson: I’ll disagree a bit with your analogy here. There, the telephone pole is being vandalized, and the power company will tend to remove the flyers of its own accord. Of course the company should have no liability if it has no knowledge that the content is libelous (and, in fact, actively discourages postings).

In some places, it’s acceptable to tack notices onto utility poles. In that case I wouldn’t call it vandalism (and indeed in most places I have lived it’s a matter one way or the other of municipal administrative law rather than vandalism), and I’m interested in why in those cases there would be a difference between that and a large and unmoderated online forum (think “Slashdot,” “Blogspot” or similar).

I will only note that not only can this work sometimes, this has worked at least once, in a spectacularly succesful way.

To wit: a small number of little-known (outside of the blogosphere) people and a smaller number of professional document examiners were instrumental in showing documents broadcast by a professional news team at CBS to be false–or at least impossible to distinguish from very-poorly-executed forgeries.

I submit that powerful people consider this a bug, not a feature: a bug that they want to exterminate. The CBS episode might have unfolded differently if the CBS principals had had stronger legal standing to litigate against the bloggers and examiners.

I have no regrets that Edwards, Sanford, Spitzer and Gingrich are no longer serious prospects for the Oval Office.

It happened before the Internet era, but I’m glad to have had a reasonable suspicion of what we might be getting when I preferred Bill Clinton to George H.W. Bush.

I’m with Fub.

Stephen Lathrop: I think that could work sometimes. But what will happen mostly? Reflect on how well that dynamic works if there aren’t ANY established information sources left. No acknowledged information authorities to validate the little-guy truth tellers and their discoveries. And how fast will your own critical faculties shrink when all you have to fuel them is a swarm of possibly-true assertions you found on the web?Seems to me there is an urgent need to reproduce online the facilities for information creation and validation that have until now been supplied by traditional media. No sign of that yet. And no sign of a business model to support it.

I will only note that not only can this work sometimes, this has worked at least once, in a spectacularly succesful way.

To wit: a small number of little-known (outside of the blogosphere) people and a smaller number of professional document examiners were instrumental in showing documents broadcast by a professional news team at CBS to be false–or at least impossible to distinguish from very-poorly-executed forgeries.

David Thompson: It’s hard to find attackers online because of the structure of the Internet itself, combined with what’s not in CDA 230.
Offline, it’s pretty easy to find most attackers (just call the newspaper, bookstore, or TV station). But CDA 230 doesn’t recognize that it’s substantially harder to find attackers online.

So CDA 230 acts as a subsidy with respect to speakers by failing to change the architecture and nature of the internet? That stretches the concept of “subsidy” awfully far.

You might as well say (more correctly, IMHO) that RFC0791 and lack of progress in Artificial Intelligence are also subsidies.

M. Gross: The entire world’s internet is visible from the United States, yet our libel laws don’t apply to it. What effect would changing the libel laws have other than to force posters onto offshore sites? It’s also worth noting that running a website hosted offshore is anonymous (with any attention paid to detail) and easy to do.

Exactly.

To state it somewhat anthropomorphically, the internet interprets impedance to flow of any information as damage (whether true information or false information), and routes around the outage.

The notion that CDA 230 subsidizes libel is a hypothesis that requires proof stronger than the obvious fact that some libelers will remain anonymous and thus not be forced to internalize the costs of their libel. By that metric life itself subsidizes murder, because not all murderers will be deprived of it.

Moreover, this argument:

Online, Google search drops users onto a website with no context or history of the site. There’s no indication whether a site is a parody (witness Salon being fooled by Landover Baptist), populated by anonymous trolls, a personal rant, or anything else. Of course, it is possible for users to perform this research for each and every site they visit–but the evidence is that they simply don’t (witness the Times of London being fooled by anonymous postings on a soccer website).

requires belief diametrically opposite to the conclusion that anointed offline arbiters of truth will not err because:

In the offline world, this risk of speech liability is largely internalized by the speaker and knowing facilitators: would-be authors of libelous publications know that they will be found and sued, and newspapers stop running advertisements they know to be false in order to stay out of court.

If The London Times in the offline world is sufficiently inept at fact finding that it can be fooled by false online statements, then how could any unitary regulator or arbiter of online truth value plausibly expect not to make the same error, regardless of incentive not to err?

Having long ago been subjected to legally cognizable defamation online (and having known others subjected to it), some of which is still available, I still have no problem with CDA 230 “subsidizing” it in the weak sense defined in this article. The far greater danger to everyone’s liberty and reputation, including my own, is the abusive application of laws effectively prohibiting anonymity.

One need only observe the abusive and draconian application of Digital Millennium Copyright Act (DMCA) takedown notices against non-infringers (as well as against technically arguable but near frivolously petty infringers) to understand the chilling effect on free speech that any effective prohibition on anonymity would bring.

David Thompson: The structure of the Internet makes it difficult to find the original attacker, and CDA 230 makes sure that hosts have no incentive to do make it easy.

Without a massive restructuring of the internet hosts logging access still will not provide an accurate way to identify users. See TOR

David Thompson: Anonymous speech is incredibly important to politics — I have no disagreement with you here. I would like to provide some solutions today that preserve the right of anonymous dissenters to stay anonymous, while helping to protect private individuals from harm

Sound’s great.. do you mean in a follow-up post?

Aultimer:
So CDA 230 is the reason there’s little chance of being found? Which provision is that, exactly?

It’s hard to find attackers online because of the structure of the Internet itself, combined with what’s not in CDA 230.

Offline, it’s pretty easy to find most attackers (just call the newspaper, bookstore, or TV station). But CDA 230 doesn’t recognize that it’s substantially harder to find attackers online. If it were always easy to find the original attacker, I would have a much smaller problem with CDA 230. Attacks would be deterred simply because nobody would be foolish enough to libel or invade privacy if they knew they’d get caught. But CDA 230 was written with the assumptions of offline media in mind. The structure of the Internet makes it difficult to find the original attacker, and CDA 230 makes sure that hosts have no incentive to do make it easy.

Google creates nothing. Any judgement the ten results are meaningful is done by the reader, not Google.

I will be interested to see if the prescriptions offered tomorrow constitute a subsidy for tort lawyers.

Jason Orendorff:
I had never heard of JuicyCampus until today. Turns out, it is out of business and the site has been taken down!

There are at least two JuicyCampus imitators still in business today. I chose not to link them so as to not give them any free advertising. JuicyCampus is a useful example _because_ it is no longer in business.

Chris Travers: So, suppose someone goes around putting libelous flyers on telephone poles. Is the utility company to be financially responsible?

I’ll disagree a bit with your analogy here. There, the telephone pole is being vandalized, and the power company will tend to remove the flyers of its own accord. Of course the company should have no liability if it has no knowledge that the content is libelous (and, in fact, actively discourages postings).

Imagine instead that the telephone company put up a sign that said “post your libelous materials here!” and then also sold advertising space on its poles. It would profit from the attention gathered by the libelous materials because they would draw attention to the advertising. And it would have actively encouraged the posting of libelous materials in order to attract more tabloid attention. That’s closer to the online reality.

Abhishek: You are missing two important points here.1) The reactions of people to an news/revelation partly depends on the source where it is found. I am far less likely to take seriously material posted by some anonymous commenter or blogger than I am to take by a professional journalist. Google may elevate NY Times and a random blog post to its front post, it won’t equalize their credibility in my eyes.

I absolutely agree that nobody takes 4chan as seriously as the New York Times. But they still take anonymous Internet content somewhat seriously. I linked to examples of the Times of London (a major newspaper) being fooled by anonymous comments on a message board, and by Salon being fooled by a parody site. The problem is that for private individuals, even the rumor can be damaging. They don’t have big PR teams to set the record straight, often don’t know how to push their side of the story to the top of a Google result, and often suffer personally as a result.

2) You focus on the privacy of those who are being written about. What about the privacy of those who are writing? CDA 230 promotes free speech by providing a safe harbor that helps protect the privacy and anonymity of those with opinions. This is especially relevant if you are criticising a powerful figure, something that previously was possible mainly for those with money and standing.You are right that Google has changed privacy and reputation forever. Unlike you, I think that is mostly a good thing.

Anonymous speech is incredibly important to politics — I have no disagreement with you here. I would like to provide some solutions today that preserve the right of anonymous dissenters to stay anonymous, while helping to protect private individuals from harm. You and I see exactly eye-to-eye on the power of the Internet to bring down corrupt and abusive regimes–it’s not a coincidence that I linked to the Iran protests on Tuesday.

Any search engine that could actually work in practice will have identical effects, and indeed all of Google’s competition does.

Unless you have human beings refereeing content (ala. the old Yahoo hierarchy they seem to have abandoned, quite reasonably), you can’t do anything else. And if you try to have humans checking everything, you can’t afford it, and you can’t keep current enough to get people to bother with your results.

The “problem” here is that there’s too much content for any centralized body to judge accuracy, fairness, or anything else.

(And I think the other issue is that this reads as a static analysis – comparing the new [and unfortunate] reality of people being able to be jerks with no reasonable way to stop them with old expectations. Remember, though, that in practice people also know that the internet is full of jerks who can defame and stalk – and this knowledge becomes more ingrained and widespread every year.

The top results for someone being full of crazy rants won’t mean anything, soon enough, other than to tell anyone that searches “You have a crazy internet stalker who needs to graduate from junior high”.)

(And contra Stephen, I don’t think the traditional media actually produce that much “reliable information” in the first place – and they never really have, we just didn’t have an easy way to check. Every time I see reporting on a topic I’m knowledgeable about, I see how wrong it is in detail and often in general. The mistake is thinking that’s not true of every topic.)

The entire world’s internet is visible from the United States, yet our libel laws don’t apply to it. What effect would changing the libel laws have other than to force posters onto offshore sites? It’s also worth noting that running a website hosted offshore is anonymous (with any attention paid to detail) and easy to do.

…some sites have encouraged tabloid anonymity (think JuicyCampus) rather than thought-through content (think VC). Maybe the market will correct the imbalance, but so far it has responded to the subsidy for libelous speech by producing more of it.

I had never heard of JuicyCampus until today. Turns out, it is out of business and the site has been taken down! So you should change this to, “The market has corrected the imbalance…” or else pick a better example.

Abhishek: I envision a world where an anonymous little-guy truth teller has by default, far less credibility than a established journalist with a history of reliability. If little-guy comes out with damaging allegations, few will heed him unless there’s further evidence or proof. But if what he says is really true or worth taking note of, there will usually be someone else who does more research and unearths it. Gradually more people will catch on, including more established media. In short truth will often prevail, falsehood will usually die.

Granted, the above account is simplistic. The dynamics in any particular case may differ, things may not always turn out as rosy.

I think that could work sometimes. But what will happen mostly? Reflect on how well that dynamic works if there aren’t ANY established information sources left. No acknowledged information authorities to validate the little-guy truth tellers and their discoveries. And how fast will your own critical faculties shrink when all you have to fuel them is a swarm of possibly-true assertions you found on the web?

Seems to me there is an urgent need to reproduce online the facilities for information creation and validation that have until now been supplied by traditional media. No sign of that yet. And no sign of a business model to support it.

tfkw: The “subsidy” is real, but it’s enormous. Requiring any kind of affirmative policing would be spectacularly expensive for any website with a decent amount of user-generated content, and a take-down notice scheme akin to copyright might not be much better.

Put that another way and you get right to the heart of the matter: It is far more expensive to produce reliable information than to make stuff up. The current business model for internet commentary is to let traditional media bear the cost of producing reliable information, then put that information on the web and basically talk about it for free. It’s exhilarating, but it can’t go on forever.

The traditional media are going out of business because of the web. On this course, there soon won’t be much reliable information to talk about. Then almost everything on the web will be stuff people make up. How could it be otherwise? Of the news you know about today, how much did you see for yourself?

Unless someone figures out a way to produce reliable information for free, there seems to be an unavoidable fork in the road ahead. Either the web is going to become less free and notably more expensive, or it is going to become far less useful as a forum for public discourse.

gs: In that context, I wonder about the full range of interests and entities that will be subsidized (and damaged) by the fix that David Thompson will propose.

ReputationDefender will be among the biggest recipient of such subsidies ;-)

Several commenters have suggested reliance on the discernment of readers as a plausible defense against falsehood. The notion, as expressed by one, is that when you compare an anonymous post and something from the NYT, you don’t have much trouble distinguishing which is likely to be reliable.

Others have suggested that anonymity is a great help in getting out the truth, in a world where the powerful have the ability to retaliate.

Two problems:

1. Those are contradictory ideas. Your anonymous little-guy truth teller isn’t going to be heeded if everybody is using anonymity as a proxy for unreliability.

I think that is a little black and white. I envision a world where an anonymous little-guy truth teller has by default, far less credibility than a established journalist with a history of reliability. If little-guy comes out with damaging allegations, few will heed him unless there’s further evidence or proof. But if what he says is really true or worth taking note of, there will usually be someone else who does more research and unearths it. Gradually more people will catch on, including more established media. In short truth will often prevail, falsehood will usually die.

Granted, the above account is simplistic. The dynamics in any particular case may differ, things may not always turn out as rosy. But in general, the above account is what I think will roughly happen, and in the balance that is why I think:
a) The two ideas you mention are not really contradictory
b) A world where there is “availability of unedited public access for cranks” is a better world, for freedom and for truth.

Google has elevated the ramblings of a lone speaker to the same visibility as the New York Times. This is a wonderful development for politics and freedom. It is a frightening development for personal privacy.
…
Tomorrow: Fixing the CDA 230 subsidy

Let me guess: it’s necessary to destroy privacy in order to save it.
***********
Commenters on previous posts have stated that ‘closing of the frontier’ is a dubious analogy for attempts to further regulate the Internet. Particularly perceptive, IMHO, was someone’s allusion to the Enclosure Acts.

Today’s post draws my attention to a supposed ‘subsidy’ for libel. I’m no lawyer, but it seems to me that any policy, even inaction, benefits somebody more than it benefits somebody else. In that sense, any policy can be termed a subsidy.

Federal, state, and local legislatures are spewing out laws faster than the BP leak is spewing out oil. Special-interest lobbyists influence laws. When it suits their purposes, governmental and non-governmental players aggressively try to expand (or reduce) the scope of a law’s applicability, and precedents are set.

In that context, I wonder about the full range of interests and entities that will be subsidized (and damaged) by the fix that David Thompson will propose.

tfkw: I’m curious to see how David Thompson proposes to change things, because it seems like there are a lot of small changes that would promptly put a lot of websites out of business. The “subsidy” is real, but it’s enormous. Requiring any kind of affirmative policing would be spectacularly expensive for any website with a decent amount of user-generated content, and a take-down notice scheme akin to copyright might not be much better.

I personally think this site would be one of those which would be impossible to maintain under such a scheme…..

Is Mr Thompson expecting that The Internet Archive would be held liable if it archives libelous posts?

Free speech aside, this is terrifying from a business perspective. You can make an argument for it on ethical grounds, but you have to acknowledge how expensive it would be. Mr. Thompson might really be saying that all social media should be on the same terms as the New York Times, but if so I’d really like to hear his take on the magnitude of this change.

Does unintended irony get any better than that?}

Um, this is a private venue.

Stephen Lathrop: Losing the editorial screen provided by traditional media turns public discourse into a minefield for listeners—and there are all too many cynics and opportunists willing to sow mines. One need not believe that traditional media perform the editorial task well, or even well enough, to understand that undertaking the task of vetting information for accuracy is a plus for politics.

I completely disagree with this. The traditional media is so hopelessly biased in favor of one party that they are unable to perform this function in any acceptable fashion. Vetting by the traditional media in our current environment is merely one page in the Democratic attacks book on Republicans.

In these cases, CDA 230 acts as a subsidy by removing liability (cost internalization) away from the speaker and host. [...] And there’s no incentive for users to not create libelous materials; in many cases there’s little practical chance of being found—even though it can take a victim years to clean up the damage.

So CDA 230 is the reason there’s little chance of being found? Which provision is that, exactly?

PersonFromPorlock: The freedom to post on the Internet devalues the credibility of individual postings, which draws the teeth of libel at the source.

Anyone who wants to put the teeth back into online libel will have little trouble doing so. Dates, times, names, specific allegations of wrongdoing, references to other authorities (perhaps purpose-built to further the libel), can all be piled into a mountain of apparent corroboration. Before you know it, PersonFromPorlock can become the guy who fled Porlock one step ahead of the law.

JP says:

I think you are missing an important aspect of Google’s libel. Google actually pushes libel at you through the search suggestion feature. Type in “is #” where #= any letter and one of the top 10 suggestions is probably libel. They do this completely anonymously with no sources whatsoever.

“is ob” is a fun one.

I’m not sure which side of the political divide you come down on from your comment, but “is bush” comes up with nearly identical google suggestions as “is ob”. The only reason “is bu” doesn’t is probably that there are lots of other plausible, popular words that start with “bu”, not so many with “ob”, so you just have to narrow it down a bit by adding a couple letters.

I’ve been targeted in Google groups/usenet by an individual intent on google-bombing me with false allegations that approach libel, but really – there’s nothing you can do except wait for the individual to cross the line to legally actionable statements – which a skilled stalker/troll can avoid for years while still issuing constant defamatory statements.

Google couldn’t care less, abuse complaints probably go right to the trash.

I’ve been targeted in Google groups/usenet by an individual intent on google-bombing me with false allegations that approach libel, but really – there’s nothing you can do except wait for the individual to cross the line to legally actionable statements – which a skilled stalker/troll can avoid for years while still issuing constant defamatory statements.

Google couldn’t care less, abuse complaints probably go right to the trash.