Thoughts on Internet Content Moderation from Spending Thousands of Hours Moderating Volokh Conspiracy Threads (Repost from 2022)

Reposting a thought I had after the Fifth Circuit's decision.

Note to readers: I first published the blog post below in 2022, after the Fifth Circuit's ruling.  I thought I would repost it in light of the Supreme Court's oral argument in the case next week.  — Orin

Reading the Fifth Circuit's decision in Netchoice v. Paxton brings me back to the old days of the Volokh Conspiracy.  A little bit of context: Back when we were at volokh.com, we introduced open comment threads. For a few years, I spent over an hour a day, every day, moderating Volokh Conspiracy comment threads.  I stopped after we moved to The Washington Post in 2014, where comment moderation was up to them.  I'm very glad I don't do comment moderation anymore.  But my comment moderation experience at volokh.com left a lasting impression.

I think three of those impressions might be relevant to thinking about Netchoice.

First: It is a strange rule of human nature that most people who are moderated in an online forum feel, with great certainty, that they are being censored for their beliefs.  Few people think they just went too far, or that they broke the rules.  Moderation is usually seen as the fruit of bias. So liberal commenters were positive I deleted their comments or even banned them because this is a conservative blog and we were afraid that liberal truths would pierce through the darkness and show the false claims of conservatives.  And conservative commenters were completely confident that I deleted their comments or even banned them because we are liberals trying to prevent conservative truths from exposing liberal lies.  It just happened all the time.  Moderation led to claims of censorship like day following night.

Second: Content moderation always reflects a message of the moderator.  My goal in moderating Volokh Conspiracy comments was just to keep discussions civil.  My thinking was that if you can keep comments civil, you will not only encourage better comments but also entice better commenters.  And I think experience proved that correct. For a few years there, moderated Volokh comment threads were pretty insightful places to go to look for perspectives on our posts.  But moderation always implies some some sort of message.  It implies some value or judgment that the site has (or maybe just the primary moderator has) that they want to advance. For example, when I was moderating out uncivil comments and commenters at volokh.com, I didn't care if an opinion was liberal or conservative.  But my moderation still expressed a value: A belief in a marketplace of ideas, where we wanted the ideas to be expressed in a way that might persuade.  That was the value we (or I) had.  It's a process value, but still a value. Moderating was always an effort to further that underlying value we had.

Third, perfect comment moderation is impossible, but you can't let the perfect be the enemy of the good.  I wrote above that many moderated commenters believed that they were being censored for their beliefs. A corollary is that many commenters had examples of comments from the other side that had remained up, apparently unmoderated, that to them proved the bias.  If you deleted a comment as uncivil, it was common to hear howls of outrage that months ago jukeboxgrad had a substantially similar comment somewhere that is still up, so that under the principles of due process and the Magna Carta it would be despicable to moderate this comment now.  The problem was scale. We might have 20 posts a day in those days, as there were a lot of short posts. An average post might get (say) 100 comments, with some getting many more. That was around 2,000 comments to wade through every day. You'd need full time moderators to try to moderate them all, with some sort of legal-like process for adjudicating individual comment moderation decisions. Moderated commenters often seemed to want that—and in some cases, to demand it.  But it was just impossible given our day jobs.  Moderation was needed to make comment threads worth reading, but the sheer scale of comments made imperfect moderating the best you could do.

Comments

[mulched]

Some of them even hold a grudge up to this very day.

[Bwaaah]

When you see life through a lens of resentment, well. there it is.

[Dan Schiavetta]

The good old days of volokh.com, with more civilized discussions. Though I do sense a return to civilization recently (is it because I've blocked the worst offenders?).

[SimonP]

Yes, it's because you've blocked the worst offenders.

[Kazinski]

Exactly, the mute button is all we need.

I don't bother muting people who are rude or uncivil, but if that's the extent of their argument then I'm not going to waste my time.

Reading through threads watching Kirkland and Drackman going back and forth doesn't shed any light on any subject.

[Bwaaah]

I can't let Drackman go down like that; not in that company. He has a sense of humor. And sometimes, a kinder, gentler Frank speaks.

Don't those count for something? I mean, compared to you-know-who?

(Or am I just racist? I don't mean to be.)

[BadLib]

In addition to the mute feature, it would be nice if a couple features were added.

First a per user option to "Collapse all comments below a post by a muted user". This would make it easier to navigate the comments. But more importantly I think it would tend to reduce the long, and rarely interesting, threads under posts by particularly annoying trolls like our Resident Reverend and discourage them as they get less of the attention they so crave.

Second, an checkbox on every comment to "Collapse this comment and all below" and to remember that status for that session for that comment so "refresh" doesn't lose it (it could be stored in local cookies). Currently I use Firefox's 'Inspect' feature to delete the HTML for subthreads that are of no interest (so I don't have to scroll searching for the end), but this is a poor stepchild to the feature I'd like to see.

[DaveM]

These are very good ideas! Maybe have THREE levels of muting: hard mute a user on all posts, hard mute a user *and* all their replies, and soft mute (collapse/expand) a post and all its replies.

I think it's great that we all get to self-moderate, it'd be nice to have a bit more control.

And speaking of control, it would be fabulous to have text-based markup for links, italic, and bold, as well as support for images.

[Stephen Lathrop]

Kazinski, unfortunately your use of the mute button is not all the publisher needs, nor is it all the public life of the nation needs.

[VinniUSMC]

So sayeth SL, the failed publisher. Take your authoritarian censorship desires elsewhere.

[Bwaaah]

If there were an AI mute capability, and you could enable it with a prompt, what would your prompt say? Mute all comments that ____________ (?).

[Rossami]

Unfortunately, blocking the trolls doesn't teach them to not be trolls. It's too universal and too permanent. (Do you ever unblock someone to see if they've finally learned to be civil?) Moderation is a lot harder and blocking has the advantage of being an individual reader choice, getting the moderator off the hook for individual decisions but I'm not sure that it's the right long-term social policy. Moderation gives immediate and more personalized feedback about which comment was over the line.

If you believe that redemption is possible and that people can learn civility (beliefs that I fully concede are strained by commenters like Kirkland and Drackman), then that argues for some less permanent solution that the binary blocking offered by the Reason commenting system.

[Absaroka]

The list I help moderate lets us either just remove someone from the list, or set it so all comments from the offender are held pending moderator review.

With the caveat that we don't do hot button issues, it works well. If someone goes flame on, we set them to always-moderated, and give one explanation of why ('don't say penis breath'). Then when they send acceptable content, it's approved. If it is unacceptable content, it gets round filed. The majority learn to self moderate and we remove the restriction. The few who don't get tired of sending screeds into the void.

Either way, everyone knows the rules, and generally does an admirable job of self-moderation. Most people slow down when they see a speed trap, and here your speed is constantly recorded.

Dunno if it would work here or not.

(our only rules are 1)posts must be vaguely on topic for the list and 2)they must be civil)

[Bwaaah]

I wondered why you reacted as you did recently about a post that you found to be too close to inappropriate (my word, not yours). It perhaps reflected a sensitivity you got from having to be a judge of comments?

I thought your concern was right. I would have missed the problem of the post you criticized without having taken some time to consider it carefully. (The criticisms of that same post by others got lost in my People-not-much-worth-listening-to bin.)

There's a lot of malice around here, rhetorically speaking. Malice alone is clearly insufficient cause to censor. That's kind of ugly, to think that the borders of concern begin well after the presence of mere malice.

[Nige]

'Unfortunately, blocking the trolls doesn’t teach them to not be trolls.'

It's nobody's job to teach trolls not to be trolls. It's everybody's job to read and respond to who they want to and to block those they don't. Then you don't have to worry about whether the trolls are learning to be better people or not. They usually aren't and its nothing to you one way or the other.

[Brett Bellmore]

Preach it! We agree on that much.

[Bwaaah]

Even a broken clock...

[EscherEnigma]

I'm not sure the comment section for an opinion blog should be overly concerned with whether or not it's mute button is the "right long-term social policy". Kind of like expecting a kindgergarten playground to operate as if it were the UN.

That said, a mute button that had a sunset provision would allay your concerns. "Mute for a day/week/month/year/forever" would allow you to mute someone that is *currently* annoying you, but allow the possibility that when their time-out is over they won't annoy you again.

That said, perhaps the biggest flaw with the Reason commenting system is that when you block someone, they aren't blocked from seeing *your* comments, or even aware that they're shouting at someone that has dismissed them. Hell, even just having a "this user has muted you" message next to people's user names would be preferable then the current system.

[ReaderY]

Lack of moderation has caused the Conspiracy to deteriorate significantly.

[Krayt]

Some of the posts are just so rotten, I wonder from time to time if they aren't troll plants, trying to poison the well against civil libertarian-esque discussion.

Some seem to be just this side of Heil, Hitler, whose with me boys!

[AmosArch]

Is that really an issue? I just notice one or two of those types coming in a year and usually they are banned fairly quickly.

[Bwaaah]

Are you sure it's Heil Hitler, or maybe just a storied hyperbolic nexus between Taylor Swift and the Third Reich?

[AmosArch]

Eh...the worst this place got was when it was part of WAPOO and a random Ilya post got featured in a prominent spot. I mean the regular leftists here are no great minds but they were/are Goethe in comparison to the riffraff WAPOO commentators who would flood in.

[Estragon]

“WAPOO … Goethe”

The jokes just write themselves at this point. The old days were no picnic but it’s better than whatever this is now.

“Third, perfect comment moderation is impossible, but you can't let the perfect be the enemy of the good.”

Yes. Amen.

[AmosArch]

what do you mean 'whatever this is now'? I don't know what you guys are talking about when you act like this place is filled to the brim with heil hitlers and ascii dicks and links to malware. I find some of the people here not that bright and there is an occasional troll coming in every year or so but this place is pretty tame compared to other places I've been.

When the general WAPOO audience flooded in from the occassional featured Somin post this place turned into juvenile name calling and ad hominem attacks until they left again for their usual haunts of staff editorials of how Republicans spit roasted babies and kittens. At least the regular leftwingers here usually attempt to have a reasoned debate.

[Estragon]

“this place is filled to the brim with heil hitlers and ascii dicks and links to malware”

Where would that be?

I hate to push everyone’s buttons but in the words of the Rev.— it’s an ostensibly legal blog. Or, it was.

Orin Kerr asked for cites to specific pages yesterday. All we got was Lars Larson rants.

Steven Calabresi does whatever he is doing… would you want to learn con law from him— or David Bernstein for that matter? Wait, don’t answer that.

The chutzpah to decry the level of discourse here in the comments while flinging wapoo about is hardly surprising or noteworthy, frankly.

[David Bremer]

There was a sudden and massive change when things came over to Reason. Sure, there were times at WaPo when a flood of new readers would show up and turn it into a standard newspaper comment section. But they were generally civil, even if not particularly educated.

The moment the blog arrived at Reason, we saw a new group of commenters who had very little, if any, substance, but loved to spout nonsense sprinkled (then covered) with racism and other bigotry.

As time has gone on, that's become more the norm. It also doesn't help that the posts have gone from being dominated by circumspect measured writers like Prof. Kerr to political axe grinding by Blackman. As the balance skews more and more toward the Blackman type, the quality of comments has also deteriorated.

[Sarcastr0]

It really feels like a 2-step function - unmoderated till it's lightly moderated till it's heavily moderated. Within each regime it all feels like the exact same effort is being put in.

I used to be part of all 3, but the heavily moderated places have all died off. It's too much sustained effort.
For all the 'you're cutting off vital avenues of discourse' I learned more new things in those places than anywhere else. More people willing to take you by the hand if you come in with a wrong assumption.

[Kazinski]

Well, god knows I tried.

[Bwaaah]

LOL. You actually do (often enough). It impresses me. I look inside myself for that understandingness, and the well tends to comes up dry. (As Sweet Brown said, "Ain't nobody got time for that.")

[Brett Bellmore]

I agree that your moderation at the original Volokh site was superb, and certainly unbiased so far as I could see. I've seen biased moderation first hand, at sites like Crooked Timber, and you can't even see that from where you were.

Really good moderation is pretty clearly superior to no moderation, or Reason's minimalist moderation. OTOH, I'm not so sure that mediocre moderation is better than none, and I'm quite sure the regrettably common biased moderation is worse, unless your idea of civility is the civility of the choir all singing from the same page.

I don't think you can do an A-B comparison between Volokh moderated by itself years ago, and Volokh on Reason. It's a different internet today, in a different society, one bubbling quite close to boiling over.

If you go back and look at Reason even 8 years ago, circa 2015? It was a rougher crowd than Volokh hosted by Volokh, but nothing like today. If you were still doing the moderation yourself these days?

You'd be doing nothing else, and still not keeping up.

[AmosArch]

I think the minimalist moderation here is just fine. Just take care of obvious spam and extreme troublemakers. Most places suffer from overmoderation. You can always mute individual users on your own accord.

[defaultdotxbe]

As a personal rule I don't block/mute people, but I agree that seems to be the best way to handle moderation other than spam and maybe some of the worse troublemakers. Beyond that simply let people curate their own experience.

[Brett Bellmore]

Oh, I agree, this is just fine; They seem to be on top of actual spam, and if somebody is persistently obnoxious, I can just mute them. It takes a pretty high quality moderation to beat minimalist plus the option of blocking.

The big problem is that people willing to put the work in for quality moderation, who don't have an ideological bone to pick, are in really short supply. Moderation is a choke point, an ideal place to situate yourself if you're inclined to be a censor, and so it tends to attract the sort of people looking for opportunities to censor.

[wolfefan]

Jukeboxgrad! Jeez, that's a callback.

[Stephen Lathrop]

What encourages censorship complaints is to let comments get published, and then take them down later. That feels like censorship because it works more like censorship than like prior editing.

If blogs instead made it standard practice to edit every contribution before publication, and to publish only the ones which passed muster, very soon you would have a commentariat which was conscientiously trying to get over the bar, with much less complaint. And you would shortly attract more high-quality commenters.

The old VC was the best VC. I miss it.

[Brett Bellmore]

"If blogs instead made it standard practice to edit every contribution before publication,"

They wouldn't be blogs anymore, they'd be newspapers, just online, and everything doesn't have to be a newspaper.

And, frankly, the few sites that actually do that sort of moderation? It's because they're so ideologically censorious that they can't tolerate dissent even being seen for a short while before they erase it.

[Stephen Lathrop]

Bellmore, what you describe would be terrible for business. If the blog were mine to manage, I would fire any editor who did what you describe.

[VinniUSMC]

Thankfully, nobody is keen to let SL near any levers of power.

[tkamenick]

SL, do you seriously think that there's no market for purely tribal gathering places online?

[Stephen Lathrop]

tkamenick, an interesting and worthy question.

The first-order answer has to be, “Yes, obviously the market exists.”

A follow-up question, to which I would also answer yes, is “Would it be possible to organize, monetize, and crucially, satisfy that market in more constructive ways than current online practice accomplishes?”

But in that second instance, I confess that it would be impossible to accomplish that goal while Section 230 directs so much of the necessary revenue to competitors who value the tribalism only for the money it affords. As you can tell just by looking around, in present arrangements there is too little expressive security for tribalists.

Also, to be forthright I confess that the kind of expressive security press freedom aspires to cannot be achieved at all on a scale to match the quasi-monopolistic business models which underpin the current internet. To suppose that could happen would be to suppose particular tribes empowered to impose dominance for their views—which if I am not mistaken probably matches your own criticism of the present situation.

[EscherEnigma]

With your interest in "tribalism", have you actually spent any time looking at and studying actual minority groups and how they formed communities (both locally and online) without the need (or in many cases, desire) for corporate sponsorship?

This idea that these kinds of spaces can only exist online if they're sponsored is very... well, entitled.

[Bwaaah]

Oh god. That's it? That's what you got?

[Absaroka]

"Section 230 directs so much of the necessary revenue"

Could you flesh out your theory of how S 230 directs revenue?

If I cancel my subscription to the local paper and just look at cnn.com and craigslist.com instead, sure, that redirects revenue from the local paper to cnn and, well, not craigslist. But 230 doesn't have anything to do with that.

[Stephen Lathrop]

Absaroka, to understand the issue, stop thinking about what you do at preference, and start thinking about what the publishing businesses which distribute your content must do to stay in business.

Those publishers typically need to make a profit. Free speech comes free. Press freedom comes at considerable expense.

Consider two business models, A and B. In Model A, every contribution gets private editing prior to publication. In Model B private editing almost never happens.

Owners of both kinds of business want to increase revenue and profits. As their principal means to become more profitable they strive to increase sale and display of advertising.

In both models audience attention will falter if the balance between content and advertising becomes too heavily biased toward advertising. Thus, in both models, as advertising increases, content to attract audience attention to the advertising must increase in proportion. It is that practical aspect of publishing—the need to maintain proportionality between content quantity and advertising quantity—which opens a competitive distinction between Business Model A, and Business Model B.

Consider Business Model A first, the one which edits everything prior to publication. If it wants to increase profits, it must sell more advertising, and that means it must also recruit more content to keep proportional the balance between content and advertising. That means that to increase advertising and thus profits, Business Model A must also scale editorial effort to scale proportionally to the increase in advertising. That happens because Business Model A has undertaken the burden to assure that all the increased content gets edited prior to publication.

Thus, under Business Model A every increment in advertising requires an increase in editorial expense. The publisher manages the business with an eye to keep ad rates high enough to leave something extra in the till after each increment in content has been paid for.

Contrast that with what happens with Business Model B, where no editing gets practiced. Advertising increases, and content increases to match. No costly friction gets generated by need to read and edit more content as the amount of content published increases to maintain the crucial balance. The new increment in advertising revenue goes straight to the bottom line without paying for more staff to edit and recruit more content—because that content does not get edited. It just pours in for free. That creates a decisive cost advantage in favor of Business Model B. The publisher using Business Model B is free to leverage that cost advantage to undercut business model A on ad rates.

By now, you ought to be able to see what happened to Business Model A when government policy abolished by passage of Section 230 the practical requirement to edit content prior to publication. In short, practitioners of Business Model A confronted the choice to give up prior editing, or to be driven out of business by competitors exploiting a cost advantage handed to them by government policy. That reversed one of the longest-standing regulatory requirements in commerce.

To some, no doubt, that result looks like an increase in overall publishing efficiency. It is only possible to conclude that if you discount to zero the benefits to society and to the legal system that private editing prior to publication delivered. Those included not only excellent protection against widespread defamation, but also nearly everything to do with publishers competing with each other on the basis of content quality. Those were enormous benefits, and the public life of the nation has been greatly stressed by their ongoing disappearance.

There are yet other deleterious results from Section 230 to be accounted for. Platform giantism, with all its attendant ills is one of them. To explain the linkage between that and Section 230 should not be necessary for anyone who understands what I wrote above, but I know from experience that for most it would require a similarly lengthy explanation. I will not attempt that now.

Hope that helps. Feel free to ask more questions. I never think of this subject as complicated until I attempt to explain it to someone who never had occasion to grapple with publishing practicalities. I do understand that the notion that Section 230 created notable interaction with publishing business models is utterly baffling to many.

[Absaroka]

This doesn't make any sense.

Once upon a time I would pay the Daily Fishwrapper to place an ad like "2003 Civic, 147k, new tires, runs good". Today I would put that on Craigslist or Facebook Marketplace. Section 230 isn't involved in that change: the Fishwrapper never sent out a reporter to look at the tires and edit my ad for accuracy, and neither Craigslist, Facebook, nor the Fishwrapper ever faced liability if I had actually stuffed enough bananas into the transmission to barely make it through a test drive.

Joe's Family Shoe Emporium used to advertise in the Fishwrapper, but they don't anymore. That's not because they liked the Fishwrapper's editing, or the Fishwapper vetted their claim to friendly service, it's because they can reach the same eyeballs cheaper through their business Facebook page, and because they went out of business because people started buying shoes online. Again, Section 230 isn't involved.

You are doing your standard thing: assume the conclusion ("S230 bad") and weaving a tangled skein of implausible deduction to try and support your conclusion. It's not a persuasive form of argument.

The internet has done both great good and great harm to society. The ability of people to feed their confirmation bias in media bubbles is terrible for reasoned discourse, and the profitability of viral rage clicks feeds those bubbles. But Section 230 doesn't have anything to do with those dynamics.

[Stephen Lathrop]

You are doing your standard thing: assume the conclusion (“S230 bad”) and weaving a tangled skein of implausible deduction to try and support your conclusion. It’s not a persuasive form of argument.

Those were not deductions, or any other kind of rationalism. They were the lessons of successful experience. They were also observations commonplace among others with similar experience.

What you refuse to learn is in fact bog-standard publishing business insight, proved and bolstered again and again by experience. But it is even better than that, because it is insight so consistently applicable that it enabled predictions ahead of time describing business and social manifestations which arrived years later exactly as predicted. Those were not commonplace predictions. They were laughed at as outrageous at the time they were made. Now the results get acknowledged, but by folks who remain in denial about their cause, like you.

That accounts entirely for why you remain unpersuadable. You are the personification of the guy who cannot understand a fact because his livelihood depends on not understanding it. Except that in your case I doubt your livelihood is at all at stake—only a cherished faith that technology somehow endowed you and everyone else with publishing power so great that no one on earth had ever enjoyed the like of it.

That is power no one has ability to deliver; it is imaginary, and you will not see it happen. I will put you down as an internet utopian.

[Stephen Lathrop]

The ability of people to feed their confirmation bias in media bubbles is terrible for reasoned discourse, and the profitability of viral rage clicks feeds those bubbles. But Section 230 doesn’t have anything to do with those dynamics.

There are folks who comment here for whom that remark would be only typically stupid. Your standard has been higher. How could the existence of prior editing at every publisher with capacity to advance and transmit what you describe do anything except dampen the spread of those vices?

[Absaroka]

The partisan media (Fox and Vox, if you will) do edit their slanted coverage, which is well crafted to be both inflammatory click bait and also short of defamatory.

[David Nieporent]

If blogs instead made it standard practice to edit every contribution before publication, and to publish only the ones which passed muster, very soon you would have a commentariat

…that didn't exist, because one can't have an active discussion when it takes hours for posts to appear.

[Stephen Lathrop]

They manage it fine at the NYT. It isn't as hard as you think. Short comment, pure opinion, no factual assertions. It runs. On to the next.

[David Nieporent]

They do not manage it fine at the NYT. Because they're the one national paper in the U.S., they still have a lot of online readers, and therefore they can still get a large number of people to post. But they do not have discussions. (Also, they don't turn on comments on many of their stories, and quickly close the comment sections on many others.)

[Stephen Lathrop]

[Stephen Lathrop]

But Nieporent, characteristically for you, because publishing practicalities are not your thing, you missed a stronger objection. There is no way to pay for such comprehensive editing if you compete against other bogs which just publish everything without even looking at it—which is possible because of special privilege extended by government to internet media.

[David Nieporent]

I'm well aware of how you don't understand the economics of the Internet. And making it more expensive for blogs to have comments would not make it possible for other blogs to afford to pay for editors. It would just lead to a lot less Internet posting.

[Stephen Lathrop]

Nieporent, it is simply not possible that you could be as careless in your legal practice as you are with your conjectures about internet publishing business models.

[Bwaaah]

What a practically horrible suggestion that is. And it's so *you*.

[Stephen Lathrop]

Why is it horrible? Why does every blog have to be craptastic? Nobody would have to contribute who did not want to. The people who did want to contribute would self-select, and be grateful for better discussions. Why be against that?

[Bwaaah]

There's a blog, produced by its editor(s), and there's a discussion under the blog, produced by its discussers.

Your proposed elimination and modification of comments effectively negates any spontaneous notion of person-to-person dialogs.

Can't you just find a place that runs daily writing contests, and submit your stuff there?

[David Nieporent]

Right: remember, Lathrop only cares about jobs for, and speech allowed by, professional publishers. He doesn't want ordinary peons to be allowed to speak.

[Stephen Lathrop]

Nieporent, I have empowered far more ordinary Americans to publish their own opinions than you can imagine. My guess is that you have empowered none at all. I do not think you are well positioned to disparage public service so much better than your own.

Prove me wrong with specifics if you think I am wrong about you, and I will apologize.

[VinniUSMC]

How many "far more ordinary Americans" did you silence in your pursuit of only letting opinions you like get published at the shitty local newspaper that fired you? Piss off authoritarian thug.

[Stephen Lathrop]

Vinni, the various publications I managed have always been profit-making businesses. Controversy serves various purposes, but any competent publisher knows that controversy is profitable.

For that purpose and others, I went out of my way to recruit views I disagreed with. I cannot recall ever getting a complaint about coverage which I did not reply to with an invitation to write a letter or an op-ed.

To do that in response to complaints from owners and managers of my most important advertising accounts was a particular pleasure. It reinforced for them that they could trust my publication to be fair-minded with everyone, and thus a sound medium to trust with their advertising. I thus found myself on cordial terms among an unusually conservative business community which opposed my politics, sometimes vehemently.

Anyone who demonstrated the slightest hint of justified criticism suggesting mis-coverage (it happens from time-to-time; sources who ought to be reliable sometimes turn out mistaken) could count on a personal visit from me in response, with an invitation to publish counter-point at whatever length they pleased, in the same location in the newspaper as the publication they disagreed with had been published.

Those practices worked so well that they provided what amounted to cost-free libel insurance. When agitated phoney-baloney would-be defamation clients turned up in lawyers' offices, they got appropriately deferential treatment for their complaints—but at times with a bit of expectation management to say our publishing practices were known as generally fair-minded and conscientious, so don't count on a lawsuit. Not infrequently, lawyers referred angry would-be clients to me, and I was happy to receive them. In such cases, almost all of them learned they were not getting even a retraction, but none ever left unsatisfied that they had been treated within reasonable standards. It was common knowledge in the community that folks who hated our editorial policies frequently threatened to sue. Nobody ever did sue, not once.

Except in instances of would-be defamation, or in a few instances of apparent derangement needing medical attention, I do not recall even once turning down a proffered letter to the editor. We were known for publishing letters so shockingly expressed that our audience had never seen their like in any other publication. That too became a selling point.

One satisfaction I got from all that was a warm inner conviction that I had a tool to advance my own preferred causes. I could simply let authors of scurrilous attacks have access to the best medium available to give them the attention needed to discredit them—which was my medium.

Vinni, it would have been on that basis that I would have accepted anything you had to offer.

[Stephen Lathrop]

Your proposed elimination and modification of comments effectively negates any spontaneous notion of person-to-person dialogs.

Bwaaah, check the comments section of the online the NYT. There are delays, but not long delays. There are person-to-person dialogues. No one intending excellence, publishing success, and profit-making business on the internet needs to be as obtuse or as inefficient as you describe. The standard of responsiveness set by the NYT would be easy to beat, and there would be a business payoff for keeping dialogue sufficiently current.

However, I question whether your objection is fully captured by your dialogue complaint. I think it probably centers more on typical internet utopian demands for unlimited commenter scope to lie, publish fraud, subvert elections, and practice defamation-for-profit business models—all totally unedited, cost-free, anonymous, and world-wide.

[Stephen Lathrop]

Bwaaah, point to where I have ever proposed eliminating comments, or for that matter modifying them except with an eye to avoid defamation?

By the way, in your view, is avoiding defamation a bad outcome for the internet?

[EscherEnigma]

Your proposed elimination and modification of comments effectively negates any spontaneous notion of person-to-person dialogs.

Incorrect. What it negates is the ability "spontaneous [...] person-to-person dialogs [sp]" among people that the moderator (be it the author or someone else) does not like, and in so doing it turns it from an open(ish) forum to a private(ish) party.

And that's okay. Not every party needs to be open to everyone, nor is every party obligated to welcome everyone.

[VinniUSMC]

Why does every blog have to be craptastic?

If you would leave, it would be less craptastic.

[Randal]

SL, if your suggested business model for blogs were a better one, people would be doing it.

They aren't. It's that simple.

[Stephen Lathrop]

Randal, no, it is not that simple. The suggested business model gets precluded by internet giantism enabled in turn by the combination of Section 230 and algorithm-driven personalized content feeds.

Absent Section 230, that artificial, adverse market condition for prior-edited publications would recede. Both the internet giants and their smaller competitors would have to pay alike for editorial effort, including prior editing, and in both cases scale those payments proportionately to growth of their content and advertising revenue.

Thus, assessment to sort winners from losers among internet publishing business models requires a level playing field as a premise. Level competition is not possible under a congressionally mandated regime which short-sightedly empowers one side at the direct expense of the other.

If like so many others you remain unclear on any of that, try a thought experiment. Whenever someone suggests repealing Section 230, an outcry ensues. Folks scream that the cost of compliance to major platforms would drive them out of business.

But we know that to practice publishing on a smaller more-local scale—even on an urban-area-wide scale—has become far less expensive on the internet than it ever had been in bygone days of ink on paper. Thus, there can be no doubt that smart internet publishing business, practiced on a level playing field, could pay for the cost of prior editing. We know that because it did pay for prior editing, and enabled profits, under a notably more expensive prior cost regime.

What insight does it take to boil that down to a comparison? Simply this: on a level playing field which insisted on prior editing for all competitors, what folks insist would bankrupt the social media giants will not do that to smaller more diverse publishers by the thousands. Those thousands previously proved they could handle prior editing, and profit by it, even in a more expensive prior regime where the cost of printing tended to amount to ~ 50% of all costs.

That comparison makes nonsense of any claim that current internet business conditions are solely the result of more efficient competition by the giants. Only an arbitrary cost advantage delivered by government policy—by passage of Section 230—enabled unedited social media giants to function more efficiently than other smaller publications which practiced prior editing as both a matter of course, and as a practical legal requirement.

So make it a point to notice. It was not internet efficiency which delivered today's publishing regime. It was a decision by government to advantage a new business model, by eliminating from publishing law a requirement wisely regarded for centuries as indispensable.

[Libertatem]

It's always going to be difficult to balance comment moderation with open discourse. Sometimes I wonder how much of the problem is because the internet 'tis a silly place.

[Bwaaah]

Nicely speaking, 'tis.

[Jonathan Affleck]

There should not be any content moderation although a social medium platform can remove content that is unfit according to state communications common carriage regulations.

Technology Overview

11. 47 U.S. Code § 230 states that “the term ‘Internet’ means the international computer network of both Federal and non-Federal interoperable packet switched data networks.” Non-Federal interoperable packet switched data networks include state and private interoperable packet switched data networks. Government packet switched data networks consist of Federal and state packet switched data networks. End user computing devices, which are within and attached to the Internet, can be government or private computing devices. Packet data traffic flows between web or cloud servers and end user computing devices. Government and private web or cloud servers download programs to an end user computing device for the purpose of transporting messages within the packet data traffic to and from web or cloud servers to end user computing devices. Government and private message traffic is mixed and entwined within mixed, interconnected, and entwined government and private technology.
12. 47 U.S. Code § 230(a) declares that in 1996 the Internet had become “a forum for a true diversity of political discourse, unique opportunities for cultural development, and myriad avenues for intellectual activity” to the “benefit of all Americans.” This declaration states that the 1996 Internet is a government-designated public forum. By 1996 the government had spent a tremendous amount of public money in the development of the Internet. 47 U.S. Code § 230(b) declares (a) that continuation of this development of the Internet is a policy of the US government and (b) that it is another policy of the US government to “maximize user control over what information is received by individuals, families, and schools who use the Internet.” The US government continues to spend tremendous amount of public money on government networks within the Internet, on government technology within the Internet, and on government end user computing devices within the Internet. Even though the US government has subcontracted a small amount of the management of the Internet to private entities, which the US government often funds and in which the US government often participates, a substantial part maybe most of the US Internet belongs to the US government or is funded by the US government. The US Internet remains a government-designated public forum until the US government repeals 47 U.S. Code § 230.
13. A 1996 Interactive Computer Service (ICS) like AOL, Prodigy, or CompuServe created its own public forum inside and outside the government-designated public forum of the Internet by means of common carriage of messages[1] from a user of the ICS to another user of the ICS. Such a 1996 ICS also “[included] specifically a service or system that [provided] access to the Internet.” A 1996 ICS also provided service that was not common carriage.[2] While a 2023 social medium platform creates its own public forum within the government-designated public forum of the Internet by means of common carriage of messages from a user of the social medium platform to another user of the social medium platform, a 2023 social medium platform does not provide access to the Internet. Because a 2023 social medium platform does not “include specifically a service or system that provides access to the Internet,” a 2023 social medium platform does not meet the definition of an ICS. A 2023 social medium platform provides an email service that has a niftier user interface than an ordinary Internet email service.[3] Like an ordinary Internet email service, a 2023 social medium platform has bailment of a user’s message in a backend server (or a user’s message can be considered bailment of the social medium platform). Such bailment is not speech of either the Internet mail service or the 2023 social medium platform, but the bailment is valuable to the 2023 social medium platform because the user’s message can be used to attract the eyes of another user of the 2023 social medium platform to a web page of the 2023 social medium platform. “Eyes-on-a-page” is a valuable item to the 2023 social medium platform. A 2023 social medium transports messages throughout the Internet, but little of the Internet belongs to a 2023 social medium platform. Most of the US Internet belongs to the government and to other private entities or individuals.
14. A state becomes inextricably entwined with a 2023 social medium platform when it uses the public forum of the social medium platform to create a state public forum, to make state announcements like job postings, or to discuss state legal procedures or rules. The 2023 social medium platform transmits its messages in state networks and runs its software on state end user computing devices. The 2023 social medium platform provides a substitute for a state web or cloud site. Bailment of the state’s messages in a backend server is used to attract “eyes-on-a-page” to a web page of the 2023 social medium platform. Such involvement of the state and the 2023 social medium platform in each other’s business and activities is much more than the grant of a liquor license and represents inextricable entwinement to the point of symbiosis.
15. The state public forum exists within the public forum of the 2023 social medium platform. The 2023 social medium platform’s public forum exists within the government-designated public forum of the Internet. Not only is the Internet a government-designated public forum, but the Internet also belongs to the class of “Establishments affecting interstate commerce or supported in their activities by State action as places of public accommodation.” The preceding clause does not require the Internet to be a place of public accommodation but only requires the Internet to be like or as a place of public accommodation. The same logic desegregated a public drinking fountain, which is an establishment supported by state action as a place of public accommodation for drinking water. A public drinking fountain is a device (a valve on a water pipe) and not a place that one can enter. A 2023 social medium platform is located within the Internet because it is located by connection and by IP address within the premises of the Internet. Premises includes grounds and appurtenances, which include wiring.

Notes

1. 47 U.S. Code § 153 (11) Common carrier not 47 U.S. Code § 153 (51) Telecommunications carrier. The FCC does not define a service to be a common carriage service. The common law definition of common carriage determines whether a service is common carriage. The FCC decides whether a communications-related common carriage service is a telecommunications service that the FCC should regulate.
2. 47 U.S. Code § 230 seems only to extend the traditional rights and immunities of a communications common carriage service to non-common carriage services of a 1996 ICS so that the 1996 ICS could without incurring possible future liability deny transport of content, which was unfit in a way that was neutral with respect to point of view. Because a 2023 social medium platform provides no non-common carriage services, this extension of rights and immunities is irrelevant to a 2023 social medium platform.
3. An ordinary Internet email service like Gmail and a 2023 social medium platform both meet the traditional definition of a telegraph service: a service that transmits a message electrically by wire or by wireless means. In 1901 the Supreme Court reversed itself and ruled that a telegraph is a common carrier. The transmission error problem, which worried the Supreme Court in the 19th century, does not exist in the context of the Internet. Since 1869 (approximately six years before the invention of the voice telephone), Massachusetts has penalized denial of common carriage of a message by means of MGL c. 159 s. 1 & s. 2. See also MGL c. 159 s. 10.

[Dan Schiavetta]

Does that bring back memories. I haven’t seen the term “packet switching” in 25 years.

[David Nieporent]

I guess you missed Martillo's extensive spamming of this incompetent legal and technical analysis in support of his farcical court case last year.

[Jonathan Affleck]

State regulation at the end-user point of presence by a communications common carrier was always explicit and implicit in the FCC framework that Title 47 created. The Telecommunications Act of 1996 did not change this two-tier framework but only made explicit communications common carriage that was not regulated by the FCC. Thus the Telecommunications Act of 1996 introduced the concepts of telecommunications service [47 U.S. Code § 153 - Definitions (53) Telecommunications service] and of information service [47 U.S. Code § 153 - Definitions (24) Information service]. Nothing in the legislative record indicates that Congress intended to exempt a social medium platform, which is an obvious communications common carrier, from 47 U.S. Code § 202 - Discriminations and preferences.

As far as I can tell, state regulators continue to use the traditional definition of a telegraph, and no federal statute has overridden this definition: "A telegraph service is a service that transmits a message electrically by wire or by wireless means." See Easylink Servs. Int'l, Inc. v. State Tax Appeals Tribunal, 101 A.D.3d 1180, 955 N.Y.S.2d 271, 2012 N.Y. Slip Op. 8366 (N.Y. App. Div. 2012). The operative rule or regulation is N.Y. Comp. Codes R. & Regs. tit. 20 § 527.2 (d). A social medium platform meets the definition of a telegraph service in NY and probably in every other state.

Since the start of US telegraph service in 1845, a state has been regulating what a communications common carrier must and must not transport. See attached article. If someone believes there is much difference between the original telegraph service and the service, which a social medium platform provides, he must believe a social medium platform must operate by magic.

The Texas and Florida laws, which regulate a social medium platform, are completely in line with the legal history of communications common carriage.

A lot of law must be clarified by full litigation. I hope the Supreme Court will remand both cases back to the trial court and order the Texas and Florida laws to be put into effect until Netchoice can show why they violate Title 47 or the US Constitution.

[Bwaaah]

X.25 baby! None of this IP tomfoolery!

[Jonathan Affleck]

Do you seriously believe the Internet is a circuit switched network? Of course, the Internet is a packet switched network.

[Bwaaah]

No, I do not. That was just jest.

Did you think there was any likelihood that a terse reference to X.25 and IP would include that level of ignorance? Where would that even come from? A network IT person from the early nineties who just came out of a closet 30 years later, landing here at a discussion page but not yet having considered the transport layer?

Not likely. lol

[defaultdotxbe]

Sometimes I suspect people who write legislation get paid by the word.

[Jonathan Affleck]

I did not write "packet switching". I wrote "packet switched". I quoted 47 U.S. Code § 230, which someone has not read recently if he has not seen the phrase "packet switched" in 25 years.

(f) Definitions
As used in this section:
(1) Internet
The term “Internet” means the international computer network of both Federal and non-Federal interoperable packet switched data networks.

[Voize]

most people who are moderated in an online forum feel, with great certainty, that they are being censored for their beliefs

I notice that the text following that observation didn't contain an actual denial.

[Rossami]

Actually, it did since you cannot simultaneously be a liberally-biased censor of conservative views and a conservatively-biased censor of liberal views.

[Voize]

Whoosh!

[defaultdotxbe]

The argument isn’t that people are simultaneously both, but one or the other (and posts that match the moderator’s bias go effectively unmoderated)

The response and denial is his point that moderators aren’t omniscient, and the seeming unmoderated posts are simply missed due to the volume of posts.

[Theendoftheleft]

Sure you can.

1. Have different censors at different times, with different views.

2. Censor an expressed 'conservative' view or a 'liberal' view if and when that particular view doesn't comport with the censor's particular stance on the matter.

The idea that there's ideological uniformity, for example, on this blog alone would be preposterous.

[Absaroka]

Well, the list I moderate is a hobby list, and we certainly don't censor people because of their views on the best pruning scheme for roses or whatever. We stop a few for being off topic, e.g. starting a thread on football or some such. Those people never seem to object.

But the ones that are moderated for being uncivil ... let's face it, they self select for incivility. When we reject their posting that calls someone else on the list a poopyhead, some say 'Ooops, sorry, got carried away', but more of them go off on what horrid imbeciles the moderators are. Most of those do that for a few days, and when they don't get a response, they go away for a few weeks, then start posting acceptable content. Others stay away, which is OK too.

What's really funny are the ones who send three paragraphs of obscene insults, then supply their phone number and suggest I call so they can explain the error of my ways in more detail. I don't take them up on that opportunity. There are some odd characters out there.

[John Rohan]

Couldn't you at least auto-moderate comments with lots of dollar signs in them? Those are really the only problem here at reason.com.

[David Nieporent]

You know the difference between Reason and the VC, right? While those posts do occasionally pop up here, they're mostly polluting the Reason comment threads, not the VC ones.

[mse326]

I actually find that the biggest issue with comment sections here aren't that individual posts are over the line (though some are) but that on any post that has even the most miniscule tangential of politics or politicians will lead to someone, or a few someones, coming in with a horrific comparison or jump in logic and that becomes the main part of the comment section to the point that the section is only like 25% actually discussing the post at hand and 75% the same tired political talking points and rally cries. And that absolutely crosses ideology, neither side is immune from it.

[Eric Rasmusen]

espeically on this site, I'd like to see quality moderation as well as civility moderation. Could you do something like check off a "Good" box when you moderate comments so we see the Good comments at the top? I don't read VC comments much because the gems are obscured by the nice but worthless drivvle.

[Social Justice is neither]

There is that progressive need for censorship again. So pathetic.

[Archibald Tuttle]

We all owe you hearty thanks for the effort. I've done moderation at another site. I know how difficult and tedious it can be. So, I understand well that you're not able to keep up with perfection.

Thanks EV.

[Vryedni]

EV also did an admirable job of moderating in his day, but you do know you’re replying to Orin, right?

[SRG2]

VC and Reason are the only two unmoderated comment venues I regularly go to. I generally follow a tit-for-tat model for my responses, which works well enough on VC.

On Reason, however, tit-for-tat simply doesn't work. It's dominated bv vocal and commonly ignorant crackers who are pissed off that Reason doesn't share their white grievances. I suspect that the reason they still hang around is that it's the only unmoderated place where they can still find a few non-right-wingers to shout at.

I think it was months before I received any kind of "moderatable" insult on VC. It took about 3 posts on Reason, coupled with the assertion that I was someone else's sock.

So I stopped bothering to remain polite over there.

[ObviouslyNotSpam]

I would say the moderation is so brilliantly done on VC I haven't actually noticed it.

The Reason comment section is just for mindless fun. Reminds me of the mosh pits of my mis-spent youth...

[jdgalt1]

There seem to be a lot more comments than when the blog was on volokh.com, and the blog seems to have become both more welcoming of them and more responsive to points raised in comment threads. This seems to reflect a change of attitude by the moderators, away from the "Mr. Ed" simile I remember them making some time ago, where "we only post when we've got something to say" (explaining why there were no open threads then).

I like to think we've become enough of a community that low quality comments are mostly driven out by peer pressure. If I could change one thing I would bring back the ability to vote comments both up and down (and see the tallies) that we had when volokh.com used Disqus.

[Brett Bellmore]

But in practice (consequential) downvoting has led to the existence of downvoting cliques and lists. At sites that use it you'll see coordinated up and down voting used to silence dissenting views.

[DaveM]

Self-moderation is the only way to go for viewpoint-based discussions. It is the only option that preserves both the freedom to speak and the freedom to not be bothered. This can NEVER be outsourced. It would be a contradiction.

There is a place for automated moderation of comments that harm the ability to have a discussion, and are not expressing a viewpoint, but rather seeking to subvert the platform. For example, spamming, denial of service attacks, impersonation, etc.

The best discussion boards have both. The VC does have one self-moderation tool, but they could do much better.

[Bwaaah]

I find this post by Orin, and the comments, to be very interesting.

(And there's an unusual level of civility, perhaps due to circumspection, that seems somewhat inherent in an enterprise of commenting about commenting.)

[EscherEnigma]

Point 2 sure sounds like an open admission that the comment section here is awful.

Which is correct.