Parler Loses Web Host
Amazon is using its dominant position to shut down a right-wing social platform.
In yesterday’s post commenting on President Trump’s ban from Twitter and other social media sites, I noted that Amazon and Apple were both banning the Parler app from its stores, effectively taking the ability to use the service on the vast majority of smartphones. Now, the other shoe has dropped.
Buzzfeed (“Amazon Is Booting Parler Off Of Its Web Hosting Service“):
Amazon notified Parler that it would be cutting off the social network favored by conservatives and extremists from its cloud hosting service Amazon Web Services, according to an email obtained by BuzzFeed News. The suspension, which will go into effect on Sunday just before midnight, means that Parler will be unable to operate and will go offline unless it can find another hosting service.
People on Parler used the social network to stoke fear, spread hate, and coordinate the insurrection at the Capitol building on Wednesday. The app has recently been overrun with death threats, celebrations of violence, and posts encouraging “Patriots” to march on Washington, DC with weapons on January 19, the day before the inauguration of President-elect Joe Biden.
In an email obtained by BuzzFeed News, an AWS Trust and Safety team told Parler Chief Policy Officer Amy Peikoff that the calls for violence propagating across the social network violated its terms of service. Amazon said it was unconvinced that the service’s plan to use volunteers to moderate calls for violence and hate speech would be effective.
In light of Wednesday’s riots which overtook the Capitol—and could have been much, much worse—and reports that more violence is planned for the inauguration and beyond, this is a perfectly reasonable move. Indeed, the may even have faced legal liability if they knowingly hosted a site being used for violent criminal conspiracy.
Still, this ought to further highlight the dangers of having a handful of private companies with this much control over not only our economy but our discourse. Amazon, in particular, is not only our most important online shopping venue but it’s software and app store serve the lion’s share of smartphones, it hosts massive video and music streaming services, and is by far the dominant web host. That’s simply too much power.
Parler will likely have to go offshore at this point to remain viable even as an online platform. But without access to the iTunes or Android store, they’re effectively shut out of the smartphone market in the US. And, again, to the extent that it’s where dangerous people are going to plot an insurgency, that’s a good thing. But there’s nothing to stop Apple and Amazon from simply shutting down political speech whose content they dislike—including that which is perfectly legitimate but is critical of their business model.
AWS had 32% of the market in Q3 2020, so there are plenty of alternatives for Parler. There are Russian cloud services, I be they would be happy to have them as a customer…
Edit: You can sideload apps on Android, so they would just have to add a link to the app on the website…
This type of thing has happened before on right-wing platforms. In 2018, Microsoft threatened to withdraw the use of its software Azure to the platform Gab if they didn’t agree to take down posts by an anti-Semitic Senate candidate in California’s jungle primary. After the synagogue shooting by a Gab user later that year, they lost the services of several domain providers, which caused the site to go offline for several weeks until they started using Epik, a domain provider for the far right.
@Kylopod beat me to it. This has been the trajectory of a number of “unregulated speech platforms” on the web — from Gab to 8-Chan. In fact, tech columnists have felt this was a matter of “when” not “if” with Parler for a while.
So to some degree, there is a lot of pearl-clutching here by folks who don’t understand server/web hosting stuff. And to the continued existence of 8-Chan and Gab, the reality is these platforms typically survive–so again, I’m having a hard time understanding all of the “it’s chilling speech” arguments.
That said, I think the question of Amazon’s reach is one that is wholly separate and warrants deep consideration. Especially given their well documented predatory business practices.
@PJ:
Right. As noted in the OP, they’re likely going to have to go offshore. I can’t imagine a significant US cloud host—one that would have sufficient infrastructure to support the service—will take the political risk of taking them on after Amazon has dumped them.
@mattbernius:
Again, I support Amazon’s move here given the circumstances. In an ideal world, I’d prefer that they force Parler to police terroristic speech rather than go after the whole platform, but it’s clear that Parler is incapable of doing that. But if AWS bans your platform on the grounds that it’s dangerous, I can’t imagine who else in America is going to take it on. It’s just a lot of power in private hands.
There’s a lot of dangerous content that is hosted elsewhere. I’m guessing that someone hosts Daily Stormer’s website, perhaps they’d be amenable to hosting Parler’s garbage.
I’m not crying over this, I hope they fail hard and fast.
@James Joyner:
Agreed. And as we have seen with other platforms like 8-Chan, DailyStormer, and others, they ultimately find hosting from servers that are based outside the US. Again, I could also do that today with my own websites.
What I’m missing here James is that I think you find this non-American hosting as an issue. But I’m not sure exactly why? Is this a cybersecurity thing? Again, I’m not seeing the underlying harm or threat. Can you unpack that for me?
Also, for the record, they could also set up their own server staff and host themselves, inside the US.
From the user agreement for Parler:
Not sure how long a leftist joining and being active will last until the ban hammer hits, but the social network is quite happy to ban contrarian users comments. Not the free speech heaven it advertises itself as…
@mattbernius:
My issue is with a handful of tech execs being in a position to determine what speech is permissible on American platforms.
@PJ:
So, I think this is a different issue. If Twitter had started off explicitly as a safe space for, say, LGBTQ people and allies, it would be totally appropriate for them to ban anti-LGBTQ speech. But, as basically THE platform for political speech in the English language, I find it highly problematic for them to censor speech on basis of content. Not because I think the First Amendment is implicated—it isn’t—but because they have become the defacto town square.
@James Joyner:
I feel like we’re talking past each other here. I’m honestly not trying to be obtuse. Accepting that by “American Platform” you mean something like Parler, I’m at a bit of a loss to see how this is different than any other media platform that runs on a broadcast platform. There have been plenty of examples of talk show hosts who have become “to toxic” for networks to carry. And again, that’s a determination by a handful of media executives.
In fact, again, the argument is that if Parler wished to, they could set up their own server system, here in the US to broadest their own content. In fact, that’s far easier than someone who wants to set up an alternative traditional media broadcast system.
Futher, you also have demonstrated that they have been in repeated violation of their hosting TOS. I fully admit that the bigger issue here is timing — i.e. why are they now being held accountable versus all the time in the past when they were still out of line. I do think that is a much more important thing to focus on (and again, to what degree Democrats taking control of regulatory structures had to do with these companies suddenly rediscovering they had TOS).
Perhaps this is a “generational thing” (and I know there isn’t a generation between us), but I just don’t see how the platform argument as you are presenting it holds water.
@James Joyner:
Except that is exactly what AWS did here. If you read the actual letter, they have been bringing up this requirement to parler for months, with examples each time. Parler proved they couldn’t or wouldn’t moderate content and therefore violated their TOS with AWS.
@James Joyner:
I grew up in a time where there were three television networks and a Fairness Doctrine. A handful of television execs decided what everyone saw, though the FCC got to put its oar in.
The time we are in now seems a lot more like the 1890’s in that there was a multitude of newspapers and magazines – which were the news delivery technology of the day. Now we have websites and chat channels, and so on.
Amazon has a lot of clout via AWS, but there’s also Google Web Services and Microsoft’s hosting platform, and … lots of stuff. I have a cousin who works (in Seattle) for an AWS competitor. (I hope they would take the same stance with Parler, by the way). AWS does not dominate the market in the same way that Amazon’s retail operation dominates its space.
You can load apps onto Android from a link – I’ve done exactly this sort of thing with an app we developed for the purpose of testing. The Google store is not necessary. The Apple store is. They have their platform locked down, so that’s the most troubling – and also the reason I don’t own an iPhone.
I’m bothered more by Apple’s situation, but still not all that much. I think there are other practices of Amazon (in retail) that bother me a lot more.
James, as I’m thinking through your stance, I had a question about an assumption I may be making that’s worth getting your perspective on…
@mattbernius:
Do you see hosting as more of “public utility” than a “broadcast platform”?
@Jay L Gischer:
Yeah, it is also worth noting that the idea of “American Hosting” also gets complicated in the era of virtual hosting where the organization providing the services can be HQ’s in the US, but the actually servers are located overseas. Daily Stormer and 8-Chan, for example, are still “hosted” by an American Company. However, the servers themselves reside outside the US (more details:
https://news.netcraft.com/archives/2020/10/23/exploring-8chans-hosting-infrastructure.html).
There’s a greater than zero probability that’s where Parler will end up as well. So in that respect, it’s still going to be hosted by an American Company.
The obvious answer here is that they should turn to one of the Chinese web hosts.
Steve
@James Joyner:
While at the same time legislators from both the left and the right seek to remove/limit Sec 230 protections from the tech companies and make them libel for what is said on their services. Tech execs can’t be both the policer of speech and neutral observers.
In the case of Paler they agreed to certain Terms of Service and haven’t lived up to that commitment. Being suspended is a natural consequence of their actions.
The defacto town square concern isn’t new, but is now being applied to the internet. In the 70’s and 80’s there was much concern about the efforts by the owners of enclosed shopping malls to limit political speech. The argument of the malls opponents was the defacto to town square argument. At the time there was litigation around this issue and if memory serves me correctly, the courts came down on the side of the mall owners. Those precedents are valid here.
The ability of Apple and Google to effectively be judges of who should and shouldn’t be able to market aps is really more salient than the hosting question. There are plenty of hosting resources and if you can’t find one that you like, it is possible to create your own. In the 80’s, some techie friends who shared a ramshackle house in Northeast Minneapolis, set up their own ISP in the basement. A half dozen servers and the needed routers and switches, open source SW and bingo they were in business. They did require a gateway connection to the internet from the local telco, which they received via a T-1, but the telco, nor any other outside service provider hosted any data. So Paler and others could go that route. Though given the traffic, it is an expensive option.
Mistermix at LGM has a good post up on this subject. He notes that, surprise!, Rebekah Mercer is one of the founders of Parler.
@mattbernius:
Realistically, not everyone can have a show on a major network television show. Or any given network. Conversely, everyone can have a Twitter account.
@Sleeping Dog:
I honestly don’t remember that debate or its context but, offhand, it’s demonstrably silly. A shopping mall is, well, a shopping mall. It was erected purely as a place for people to gather to shop, eat, etc.
Twitter and Facebook were established as and did a lot of encourage as many people as possible to sign up, build communities, etc. so they could be monetized. Eventually, those two platforms pretty much took over given the nature of the enterprise. Now, being deplatformed from either seriously damages our ability to engage in political speech.
@James Joyner:
I am gobsmacked by the characterization of Twitter as “THE platform for political speech in the English language”. That seems like an outrageous exaggeration to me, and I’m not that old.
(He said, participating in an English language political discussion on a non-Twitter platform…)
@Jay L Gischer:
I’m not sure why you find this troubling. To me, it’s one of the iPhone’s biggest pluses. If you own an iPhone, every app has had at least basic checking to make sure it does not contain malware. Android apps have no such guarantee. Heck, there isn’t really “an” Android. Every manufacturer tunes and alters Amazon to their own commercial benefit. When I briefly got my kids Android phones when we lived in China, you could only download apps through the manufacturers store, which only contained items approved by Chinese sensors. You couldn’t install Google Maps or Facebook, even if you found it somewhere else.
@James Joyner:
Not only wasn’t it silly, but for a number of years the side for “public square” was winning. I very much remember tables set up in malls where the fusion or anti-abortion people distributed their literature. Airports were even worse. In fact, there was a movie that began with a bad guy character punching out a Moonie or Hare Krishna or some other such type who was pushing their literature on him as he walked through the airport. I remember thinking, sure, a bad guy, but he’s got a point….
As a broader take, I do not think that a “one size fits all” model for social media works in the long run. In any final reckoning, people who operate platforms are going to have differing opinions about what is acceptable on their platform. This is a good thing, I think.
I would like to see this lead to a trend toward smaller, federated social media platforms, serving different constituencies with varying terms of service. For some swearing and sexual material are off limits, for others not. And so on.
I rather despise “the algorithm” in its current form. I want to be the one deciding what is before me. I don’t mind help finding stuff I might find interesting, as long as I’m still in control. One of the most infuriating things is when stuff keeps showing up, even though I’ve seen it, and realized I don’t like it and don’t want it. Just go away already!
I’m not confident that this can be accomplished by legislation. Maybe it won’t happen at all.
Do you know who else thinks this kind of control in a handful of private companies is dangerous, James? That woman whose policy prescriptions you‘ve claimed are too radical for the US. From Elizabeth Warren‘s 2020 campaign position paper on Big Tech:
@James Joyner:
Wait, I’m losing the thread here. Are we talking about hosting (the issue with Parler) or are we talking about content moderation (the issue with Twitter)? Because, while there are similarities, I see them as being different issues.
@mattbernius:
Yes, in that the Internet is supposed to be a neutral space and is essentially unlimited. But, really, that’s more my argument with the power of Twitter and Facebook. Here, it’s even bigger in that Amazon is a vertically- and horizontally-integrated goliath. It’s simply more powerful than any private entity ought to be.
@MarkedMan: ” In fact, there was a movie that began with a bad guy character punching out a Moonie or Hare Krishna or some other such type who was pushing their literature on him as he walked through the airport”
You may be thinking of the movie based on Charles Willeford’s Miami Blues, in which Alec Baldwin breaks the finger of a Hare Krishna, who then dies of shock. But Robert Hays also does it in Airplane…
@James Joyner:
It was just a bit before your time as an adult James. And that “silly” argument was unsettled enough that the Supreme Court took it quite seriously:
https://en.wikipedia.org/wiki/Pruneyard_Shopping_Center_v._Robins
And it’s been revisited a few times by the Supreme Court. The decision has been used to shape policy, including online moderation debates for years.
@DrDaveT:
OTB is a tiny walled garden. We have a few thousand visitors a day, and maybe a few hundred regulars—maybe three dozen of whom actually participate in the discussion.
On Twitter, I have ~8000 followers but can engage pretty much everybody from government officials to industry leaders to the intellectual elite if they’re interested. At the very least, I can follow what they have to say and be connected in that manner.
@mattbernius: Ah. That happened while I was in high school but never made it to my Con Law books. But this seems key:
Because, ordinarily, I’d think shopping mall owners would have the right to ban demonstrations, soliciting, etc. on their private property. I see that as different than a platform that’s explicitly a venue for speech.
James and Matt,
I think I understand where your disconnect is. If James is saying what I think he’s saying I agree with him. Here’s how I would explain it:
The problem isn’t so much with private companies deciding – for whatever reason – to ban Parler or anything else. Yes, private companies have every right to moderate content on their platforms. The problem is that the internet has developed such that a handful of these companies are effectively the gatekeepers for what’s become the national public square. That gives that handful of player immense power in shaping not only what people see through their black-blox algorithms, but also in terms of who gets banned and who doesn’t. And I think that is what James and I see as a problem.
Those who are cheering these actions against Parler and Trump ought to consider that there is nothing stopping this same handful of companies from banning BLM or any other movement or organization.
And that handful of players tend to be insular and when one of them does something, the others usually follow. It’s effectively a cartel in some ways.
So the problem isn’t that private companies are banning content they don’t like on their platforms, the problem is that this handful of companies has too much influence on the national conversation.
Consider this scenario: Supposed Rupert Murdoch bought 90% of the nation’s newspapers and news outlets and then decided he would no longer allow liberal or progressive views on any of those websites, papers, and platforms. Since it’s a business, “free speech” in the limited sense of government control and regulation of speech does not apply. But I think we would all object to that situation. And we’d object not because of Murdoch doesn’t have the authority to determine what information appears on platforms that he owns, we’d object because Murdoch controls too many platforms.
I think that’s what James is getting and that would be my view as well – the issue isn’t so much that Amazon and others have blocked Parler, Trump, and other content, the issue is that the decision to block content effectively rests with a small number of very rich and powerful companies. And furthermore, these companies have demonstrated time and again that their moderating decisions are – to put it charitably – not consistent or transparent. In fact, they appear to be quite arbitrary. That’s the concern.
@mattbernius:
I also forgot that Amalgamated Food Employees Union Local 590 v. Logan Valley Plaza (1968) and Lloyd Corporation, Ltd. v. Tanner (1972) were also key Supreme Court cases on the topic of speech rights within malls (my 1st Amendment Law professor is looking down on me with a sad head shake at this moment.
It’s worth noting that while Twitter is a quasi-public space, the presence of an affirmative Terms of Service arguable differentiates it from a Mall. That’s been a critical differentiator — you have to explicitly enter into an agreement to speech restrictions to participate in the platform.
The politics of Moderation… that’s something different and I highly recommend Tarlton Gillispie (among other’s) writings on that.
More details on those other two cases:
https://mtsu.edu/first-amendment/article/582/lloyd-corporation-ltd-v-tanner
https://mtsu.edu/first-amendment/article/463/amalgamated-food-employees-union-local-590-v-logan-valley-plaza
It’s pretty typical in US regulation habits: if your organisation polices itself sufficiently, then you are more or less left alone. If your organisation can’t keep control over bad actors, especially if there are externalities to society from their bad behaviour, then you get control imposed upon you by a bigger, less indulgent entity.
I weep no tears for Parler and the hosting problems that they are now confronted with.
@Andy:
Apologies for the snark, but if you list a few more buyers and switch that from newspapers to “am news radio stations” that’s essentially the media landscape we have been living in for almost 2 decades. So, I’m not sure what the point of this thought exercise is. We are essentially already there.
(ok edit is working again, so I’m combining these as they are on the same topic).
The newspaper example doesn’t really work because there are no limits–beyond economic–on the number of newspapers that can be in a location. That was why they proliferated during the era of print. So that really doesn’t work.
Radio is a better example in that there are a limited amount of Stations based on bandwidth. Given the long standing right-wing take over of AM radio (yes, I realize sports radio still exists, but that isn’t a real alternative to right-wing radio… and then we also AM Christian stations which are almost exclusively right-wing in ideological alignment), the reality is that it requires someone buying an existing station rather than trying to get a new FCC radio license which are extremely difficult to come by (even for low-fi community bandwidths). So that’s a case where there has been a defacto takeover of a limited resource and honestly, everyone has been cool with it as far as I can tell.
@James Joyner:
And if Twitter vanished tomorrow, that would still be true, because they have their say in some different venue(s).
You are arguing that the union of all non-Twitter venues and channels, plus whatever competitors to Twitter would arise in the event of Twitter becoming objectionably censored, would not be sufficient to support ongoing national political discourse. I think that wildly overstates the importance of Twitter.
@Andy:
Ok, I found the issue at hand “block content.” In my opinion, that isn’t an accurate representation of what just happened (or at least not in Parler’s case).
Were the people in question deplatformed? Yes.
In Twitter’s case, that is a permanent deplatforming. And I agree there should be real discussions about moderation policies and whether others (like the Chinese government) should be deplatformed as well. To the degree that this constitutes a blocking is I think a fair issue. I personally have big issues with the fact that suspending a user’s account also suspends all of their past tweets (and that I definitely see as a very specific type of “blocking.”
But in Parler’s case, there is no content blocking going on here. Again, as the ongoing odeous presence of the Daily Stormer proves, they will find another hosting service and there content will be back on line soon enough. It may have to be accessible by other means, i.e. a mobile browser interface versus an app. But the reality remains that its still accessible. (BTW, this is part of the reason why I’m saying that Twitter Moderation is a fundamentally different issue than what happened with Parler and collapsing the two into each other isn’t helping us have substantive conversations).
Pull all of this a different way, a lot of this also comes down to a definition of what a platform is and whether or not we are considering Web Hosting (AWS) or Twitter (Communications Platform) as public utilities and therefore need to be regulated as such.
I’m definitely open to that conversation. But then we should do directly at that versus dancing around the edges.
@mattbernius:
Sinclair and local TV stations, the main news source for many people.
@charon:
I thought about that, but I think a key difference there is that they are still opting to run network content during primetime that tends to be far more liberal (especially if they are running news show content).
You really don’t see that same thing with talk radio that tends to run only Right Wing content (and occasionally some syndicated lifestyle/speciality programs that might drift a little more to the left).
@Andy:
If Joe Biden starts calling for the cancellation of an election and urging violence against the government of the United States I hope the gatekeepers cut him off. They did the right thing, they did it at the right time and for the right reasons. Might the power be abused in the future? Sure.
And BTW, did it not help that Trump was frantic to cancel section 230? Yeah, that was some stupid politics. But it’s no big reveal that Trump is a fucking idiot. Twitter did the right thing.
@James Joyner:
Our current posture on antitrust law was largely defined by Robert Bork in “The Antitrust Paradox” and implemented by Republican politicians and Republican judges of the Law and Economics strain. And, of course, big bucks corporate lawyers. As I understand it Bork’s book was largely a bait-and-switch. Antitrust law was about power, Bork somehow made it about prices, and then argued that economy of scale meant monopolies were good for consumers. I read the students in his anti-trust law course referred to it as Bork’s pro-trust course.
Think how much more damage he could have done had he been confirmed to the Supreme Court. He really was a RW nut job. But very skillful at it.
Here is a thread on a lot of the technical difficulties Parler is faced with:
https://twitter.com/QuinnyPig/status/1348117921965297667
etc. etc.
@mattbernius:
TL:DR; at least in Parker’s case creating friction to make it more difficult to host or access content != the same as blocking.
@charon: Great thread on the technical issues of switching servers/cloud hosting.
I did find this entry particularly on point:
The point was the repeated and unaddressed calls for violence on Parker are what lead to this (and Amazon’s rightful concerns about the liability that housing those calls could lead to). Not to mention that this is well within the terms of the contract.
@mattbernius:
This from the thread:
Which I take as a sign of how toxic Trump and the people who attacked the Capitol are now.
@mattbernius:
Sure, but I was just an example to illustrate the actual point I was trying to make.
But the radio example is interesting. And, not to be snarky, but the left has been complaining about it for years with some wanting direct government intervention when it comes to content. So it seems no one likes it when their side is disadvantaged yet don’t care much when it goes the other way.
I would just go back to first principles which is where I try to root my viewpoints. And so I would say the same thing about radio – the problem isn’t that the content skews heavily in one direction, the problem is there are too few players making content and editorial decisions. The same problem exists with the tech giants as discussed above but to an even greater extent since tech is orders of magnitude larger in terms of audience and influence than radio is and tech and social media are growing much faster.
Also, I don’t think the government has any business trying to moderate or otherwise enforce viewpoint diversity on any platform. The government’s role is to ensure there is adequate market competition more generally and promote and incentivize transparency.
BTW, the edit function on the site is very weird lately. 90% of the time it isn’t there and comments are uneditable, but every once in a while it is. I can find no pattern.
Also, it frequently takes a page reload or two for the format buttons to show up.
Not a huge deal, but I just wanted to point it out.
@Andy:
I agree with this point. But it is also the case that no substantive action has come from it.
And additionally you are continuing to side step issue of friction versus blocking.
Beyond all of that this diametrally comes down to a question of whether or not these are akin to public utilities or not. If you are not willing to have that conversation, then really it’s a case where, despite your and James libertarian free market tendencies, you don’t like the results that same market has led to.
Unfortunately, you both also have a habitual aversion to discussion of regulation and therefore your first principles are utterly unfettered from reality.
Which again is cool, but really unproductive for a grounded discussion.
@Andy (and James), if the edit button was working, I would remove/alter “habitual aversion” as I think it’s reads as way more snarky than intended. Apologies for that.
The underlying point still stands in that this really feels like an issue where a moral panic (big tech censorship!) Is intersecting with competing and conflicting ideologies around free market actors and regulatory frameworks. And the moral panic aspect “all unpopular speech is in danger at any moment” send to be blocking us from substantive conversations about how best to operate within these points of contradiction. Or what are policies to investigate at the national level.
@mattbernius:
I think that’s only a semantic distinction. Parler was removed, according to Apple, Google, and Amazon, for not blocking content. The implication is that if Parler moderated Trump and his followers, or banned them as Twitter did, then Parler wouldn’t have been banned from the app stores or kicked off AWS. But they didn’t, so they were.
Blocking content (or attempting to block content) is entirely what is going on.
It will be interesting and instructive to see how far this goes. Will any platform or app that allows Trump to say what he wants to be removed from the app stores? Will every AWS website that does the same be closed for the same reason as Parler?
Again, this isn’t only about one app or the mad ravings of Donald Trump and his followers – it’s also about a few very powerful companies in control of a big part of the global commons unilaterally deciding what is and isn’t acceptable based on inconsistent and opaque criteria. In isolation, banning Trump is reasonable and warranted, but what’s happening doesn’t exist in isolation.
And since this is OTB and people have the tendency to come up with novel and uncharitable interpretations for what I write, I want to reiterate that I’m not saying that Trump or Parler shouldn’t have been banned; I’m not concern trolling with a secret desire to be nice to Donald Trump; and I’m not offering or advocating for any particular solutions or remedies. At this point, I don’t think any remedies are warranted beyond careful observation and skepticism. I’m not trying to create a hill to die on here.
The concerns about concentration make all kinds of sense when you are talking about Twitter and Facebook. They don’t make any sense when discussing AWS. Because AWS does not have dominant market share in their business. They are big, but not dominant in the way that Twitter and Facebook are.
And yet we keep drifting back to “they are too big and they can shut people down”. Please try to keep focus on what the actual chokepoints are. They are Facebook. They are Twitter. They are broadcast media. They are the Apple App store. (Not the Play Store, there are alternatives there).
There has been a distinction about “open platform” versus “closed platform” that has existed since I played a very small part in creating the internet. “Closed platform” was always bad, but it succeeded anyway.
On a related note, I’m still really salty about the demise of Google Reader and RSS in general. Closed platforms, e.g. monopolies, are favored by businesspeople because it makes them a lot more money. They aren’t good for the rest of us. This should be a “duh”, but it hasn’t been.
@James Joyner: How mate is this in any way different than the entire history of publication? One could have said precisely the same thing about any publisher in all but a handful of large urban markets.
@mattbernius:
Not sure how I’m sidestepping it. The stated goal of the actions of these companies is to block Trump and his content on their platforms as well as other platforms that they are gatekeepers for. Not sure how much more clear their intentions could be.
But it’s also obvious that the internet is a big place and therefore it’s impossible to completely silence anyone. After all, Al Qaeda and other foreign violent groups still use social media, including Twitter and Facebook. Maybe that’s your point about friction? If so, then I agree.
But I think one must consider first principles, consistency, and at least acknowledge that if a tool is used for something one thinks is good and righteous (banning Trump), it can also be used for something one thinks is bad (say, banning BLM).
And my point here is we should be skeptical about those who wield the tool, especially when it’s clear they are not using it based on principles or using it transparently.
@Andy:
This is why I keep saying then this conversation needs to be about whether or not AWS is to be treated like a public utility versus a content platform.
Or we need to talk about the CONTENT.
Because we either say that, like a public utility, AWS or Twitter is in no way responsible for ANY monitoring of their content or then we need to have a grown-up discussion of what constitutes dangerous and regulatable content.
Because currently, AWS is liable for things like Child Pornography and use in the coordination of Violence on both the Civil and Criminal Levels. That keeps getting left out of these high minded conversations and moral panics.
So where do you net out on those aspects? Should AWS or Twitter or Facebook be legally responsible for any content on their respective platforms?
@Jay L Gischer:
I agree with you that AWS isn’t that big of a deal compared to Twitter, Facebook, Apple and Google.
I’m 100% with you there. I still miss Google Reader. I’ve been using Feedly for quite a while now and it’s been an OK substitute.
@Andy:
This might be shorthand, but it’s disingenuous as written. The goal here wasn’t to “block Trump” or “block Parler” — it was to block specific content that those individuals were posting that, among other things, were opening the platforms up to liabilities. Additionally both platforms provided said individual and organization, with numerous warning that they were in violation of the TOS and contracts that they signed with said platforms.
So I’m sorry, unless we want to go down the true public utility route discussion, and absolve platforms of any liability (criminal or civil) for the content that they deliver (in this case things shift to a discussion of the reliable and fair delivery of the underlying service and then what governmental regulations are in place to police the content on said platforms), then this is one big moral panic that has very little basis in the reality of these situations as they’ve unfolded.
@mattbernius: My take is that Dr. Joyner is hoping that the market can police content without refusing to serve the providers. I’m not a customer for that hope–wish in one hand s
hpit it the other, I know which hand fills first.Also, following through on some of these points, I can’t help but wonder if the underlying question is “why can’t Twitter be more like Parler? Or better yet 8-Chan…. Because no one gets banned there and you can say ANYTHING!”
Which, hey, good luck with “let’s have a totally unregulated speech space…” I’ve seen way to much goatse in my life to really want to venture down that particular path.
Or are you calling for giving the Government all of the regulatory responsibility?
@Just nutha ignint cracker:
But isn’t that what they are doing right now? And what happens when the people who make up that market refuse to comply with the regulation, as in the case of Parler? Or the President?
@Andy: Well, too bad that “conservatives” have essentially wiped all anti-trust enforcement off the government’s to-do list. You know, there used to be rules governing how big companies can get, but then some clever libertarians argued that as long as monopolies don’t raise prices everything’s good, and the right wing just went with it.
@wr: “Miami Blues” was the one I was thinking of. Thanks.
As a bit of an “olive leaf” of sorts, I present this Twitter thread from a University of Chicago law professor that I think points to a better starting point for a discussion around these same topics.
https://twitter.com/glakier/status/1348280186685575169?s=19
I don’t think I fully agree, but I need to sit and reflect on it more. It does basically set up the issue that or laws and regulations are not designed for the current quasi-public spaces that we are occupying and conducting our speech acts within and that is probably the real issue at the heart of all of this.
@James Joyner: Definitely not limited to California. My comments above were about when I lived in Rochester, NY
What is Parler’s business model, meaning, how does this get paid for? Most of these platforms, especially Facebook, make their money to run through advertising; it’s how they operate for free. Is there advertising on Parler? If so, are there enough brands beyond MyPillow, etc., that can keep this endeavor afloat?
The type of infrastructure and investment necessary is expensive, and even the Mercers won’t want to keep pouring cash into a black hole. This could well be one of those things the market ends up sorting out.
Reddit tried the pure libertarian approach and their service became known for hate speech and kiddie porn. So they changed their minds, because: reality.
Gatekeepers are useful and necessary. Twitter had terms of service which Trump repeatedly violated. His fault. No one else’s.
One last one, as this conversation is helping me shape some view points…
@James Joyner:
First James, apologies for missing this sooner as I think it’s a good comment for drawing a distinction.
I’ve been trying to draw a distinction between communications *platform* and *utility* because there are arguable some key differences here that are necessary to understand my thinking on this issue.
First, by a utility, I mean a network that carries a message but ultimately doesn’t retain it. For example, phone operations (though this isn’t necessarily even the case anymore with digital networks) or the post office. It’s worth noting here that while utilities has far less liability for what they carry, there is still liability and regulation of materials (see all of the great mail and pornography Supreme Court cases– the one involving “Blue Balls” and “Intercourse” PA are a particular favorite).
Platforms, on the other hand, generally speaking, maintain possession of the communications. Among the earliest example are town bulletin/message posts. And it’s worth noting that these were already regulated environments. Even to this day, posting something to your library board means that you’re following some type of posted TOS.
So in that respect, they have always already been subject to regulations that arguably infringe on strict first amendment rights. In part that’s because things like commercial or violent speech (not to mention pornography, let alone, pornography created with children) are all regulatable speech in public spaces (let alone quasi-public/private).
This also gets to the role of explicit and affirmative Terms of Service as barriers of entry to the said communications platform. Which gets to the other critical aspect of a platform — it’s a space one enters into (a semi-closed ecosystem) not unlike a physical structure. And that adds the private component to this entire thing.
Like I said, our current speech laws (including critically liability laws) aren’t really structured for a quasi-public private spaces. Again, Tarleton Gillespie does a great job of tracing out the resulting tensions in “Custodians of the Internet: Platforms, Content Moderation, and the Hidden Decisions That Shape Social Media” — highly recommended and super readable (disclaimer, Tarleton was one of my professors and advisors during my brief time at Cornell).
FWIW, I completely agree that speech on platforms is something that needs to be taken up by the courts and legislatures to come up with a better solution. There’s a case working its way through the 9th Circuit around Twitter’s TOS that may give us a sense of where that the current Roberts Court (which has typically come down hard on the rights of private institutions to police speech) stands on this issue.
But generally speaking, unless we are willing to go down the legislative route, I think this is going to be a case where it’s necessary to pick one poison or the other — either we need to take away the responsibility for platforms to police themselves (and with that some of their liabilities) or we need to accept that we’re willing to accept TOS based moderation from private entities (and the best we can push for is consistency–which, I think we have seen with AWS and honestly, the reality is that if Trump wasn’t PoTUS, his account would have been suspended within the last for years for TOS violations).
@mattbernius:
My view is regulating these companies and essentially turning them into government-sanctioned monopolies (ie. public utilities) should be a last resort and is undesirable for many reasons. And doing that would make the content on those platforms subject to the 1st amendment. If the online commons is going to have government oversight then the Nazi’s will have to be allowed to march in a virtual Skokie.
Another problem is that regulatory control would be centralized in a federal agency and I’m not a fan of centralized authority. I would think the problems of centralization should be obvious given the example of the Donald Trump presidency.
I don’t think these companies should be criminally liable for content posted on their platforms but should obviously be required to remove illegal content. There is nothing new there and what we are really talking about are Terms of Service and more general moderation. I think if companies are going to choose to moderate content, then the moderation rules should be clear, objective and consistently followed. Failing to consistently follow moderation policies would open companies up to lawsuits similar to how it works in other areas. And, of course, whatever moderation policy has to be consistent with all the usual anti-discrimination laws. I don’t think it’s as functionally difficult as you’re making it out to be, but there is the practical problem of how these companies can manage and moderate billions of pieces of content by hundreds of millions of individual users. Another reason to oppose anything approaching a monopoly or oligopoly whether it is government sactioned or not.
Like banking and other areas, my view is there should be no firm that is “too big to fail” or have too much market influence. So my view is that government should promote competition, not preference the business practices of the established players, and appropriately use anti-trust laws when necessary. I don’t think we are at the stage where anti-trust action is necessary, but we’re moving in that direction.
If that were true Twitter could have simply continued to moderate/delete Trump’s tweets as needed instead of kicking him off the platform entirely. I think the argument that Apple and Google were forced to remove Parler entirely because of a fear they might be sued for Parler’s lack of moderation is not true, considering there are plenty of apps out there that have all kinds of actually illegal content, to say nothing of objectionable content.
Sure. And I guess maybe I haven’t made this clear enough so I’ll state it – again- that I don’t have any problem with Twitter dumping Trump and the rest of it. I don’t have any issue with TOS and their enforcement in principle. Where it starts to get concerning is when:
– Market concentration is such that there is a de facto cartel or monopoly. I don’t think we are there yet, but we’re moving in that direction.
– The TOS enforcement is arbitrary and lacks transparency. That has clearly LONG been the case with social media companies. In a world where there are many options to choose from this isn’t a problem – individuals can use another service or app. In a world where most online gatekeeping comes from four companies that aren’t even in direct competition with each other in many cases, then it becomes a problem. If Twitter and Facebook are going to have ToS, then it’s not unreasonable to suggest they ought to be enforced consistently. Otherwise, they are just window-dressing.
No one here is saying there is any moral panic. No one is saying the world is ending. But, for me at least, this is a real issue that we eventually have to deal with as a society as the bulk of our nation’s political and other speech moves online.
And for those who operate off of first-principles instead of ideology, I would think AM radio parallels you mentioned are pretty plain – if market concentration is bad in AM radio, it’s certainly bad for the exact same reason in social media, banking, or any other business sphere.
Ok, at this point I think I’d just be repeating myself and my wife is asking why I’m at the computer and not doing what I’m supposed to today, so I think I’ll leave it here. I may respond to further replies later this evening depending on time.
@Michael Reynolds:
Yup, I was thinking a lot about Reddit’s evolution as we mentioned this. Also, don’t forget illegally obtained revenge and celebrity porn too.
It’s also worth noting that Reddit, among other companies, has also been sued for civil damages over said content.
In some cases, there have also been federal charges over content (especially if it’s perceived as child porn or human trafficking). I don’t believe that ever involved reddit, but the nature of reddit makes it a really hard thing to google for and I’ve already spent way too long on this discussion thread.
@Jen:
Parler intended to make its money through influencers, whom various advertisers would target.
@Andy:
Again, they were in DIRECT VIOLATION of Apple and Google’s TOS.
Look, I am not making these companies out to be the good guys. But apps get removed from the app stores all the time based on reporting. Again, chances are that the reason that Parler has been able to get away with this for so long (despite documented infringements) is because both Apple and Google were afraid of additional regulation from Conservatives.
AGain, this isn’t to see the switch as noble now that liberals are in control.
But this gets back to the larger issue of consistent application of TOS or proposing an alternative governance framework.
Also…
What are you talking about?! Philosophical first principles–which yours are– raree inherently ideological…
Same and pace! 🙂
@wr:
Then I’d like to introduce you to the Dodd-Frank act and related legislation that allowed big banks to stay big and get even bigger following the financial crisis. Small banks were specifically allowed to fail (last I checked, the number was several hundred in total), with their assets gobbled up by the bigs banks with federal government approval. It was a scheme that no Libertarian I’m aware of supported, but was supported by almost every Democrat. When these bigger, systemically even more risky banks cause another financial crisis in the future, I’m sure there will be people blaming Libertarians for it.
Full Disclosure: I’m not a Libertarian.
@mattbernius:
Got time for one more:
Yes, and as stated many times, I don’t have a problem with enforcing TOS and don’t have a problem with removing Parler. The two issues I’m addressing are the lack of consistency and transparency in TOS enforcement and the fact that market consolidation is putting these enforcement decisions into a few hands.
To that I’d add a third potential issue – I’m skeptical that these big companies have the ability to evenly enforce their ToS even if they wanted to. Hence another reason why fewer, larger social media companies are problematic.
Yep, I agree, and though I said so.
Because principles matter when it comes to governance and all these other process issues. Whatever system or rules we adopt should be based on first principles and not on whose ox gets gored nor should it be subject to any particular ideology.
Too often in political discussions, the legitimacy of an act is solely based on who it benefits instead of underlying principles. We see this frequently when partisans switch sides when the circumstances change, in some cases making the same arguments their opponents were making before because now the argument suddenly serves their partisan interests. We can’t base policy on such self-serving and ephemeral whims, it needs to be based on neutral and objective principles that can be consistently and evenly applied.
Ok, out for real now, thanks for the discussion, and have a good day!
@mattbernius: I haven’t been chiming in, but I want to thank you for taking the time and effort to make your important points clearly and with references. It’s pretty clear that the people you are in discussion with are not as versed in either the case law history or the issues as you are, and you are dealing with that very capably and civilly.
In my line of work, I am surrounded by people who believe that the free market would somehow magically produce the services (and constraints) of a perfectly regulated public utility, if government would just get out of the way. This is for the most part a religious belief, not empirically based. The mechanisms for changing that belief are those of religious conversion, not those of academic debate.
@mattbernius: The other side of my point, I think. They want to have the cake and eat it to. They regret that the sanctions are necessary and wish the sanctioneers wouldn’t carry those sanctions out, because… freedom of speech. At least that’s what I see when I read the apologies. YMMV.
Wish in one hand…
@mattbernius: I see your dissonance. In my mind, the problem is that speech has become the province of the demos more and more as time goes on. Even in the 70s, speech by most people was limited by their ability to attend public meetings where comments would be heard and access to stationary and stamps based on the good graces of chairs of public meetings and editorial page editors. Now, any hairball (or ignint cracker if you prefer) can say pretty much anything and upwards of hundreds of thousands of people will hear it. By comparison to the past, speech is virtually unlimited–and a lot of it is, in fact, dominated by the voice of the demos (in the ancient Greek sense of the mob rather than the modern sense of organized crime) whom, in my expectation, no reasonable person ever expected to be heard except in times of rebellion, when their collective blood would be spilt to restore order–unless they won and changed systems.
Our systems keep out stripping our ability to use them wisely. Over. And. Over.
I am punching out on this for the day. Thanks to everyone for the great discussion/debate.
One thing also that I realized after all of this is that I haven’t touched on the dreaded Section 230. That had huge implications on these discussions as well.
I also don’t know it well enough to even start that conversation.
@Andy: Thanks for the nod. One thing…
I did not mention Google. You added them to the list. What issue do they pose, what is the threat? I’m not saying there isn’t one, it’s just that it isn’t apparent to me, and nobody has made a clear case for them. Yes, they are big. However, there is no inherent monopoly for search. They just do it better. You could switch to Bing tomorrow. It’s not like “all my friends are on Google”. But I’m starting an argument. I mean to inquire, truly. What is the issue you see with Google?
Too busy today to follow the thread.
But I wanted to say that perhaps the best analogy to the domination of a social media giant in the past, would be the Yellow Pages.
Yes, there were other outlets for advertising, but people tended to look at the commercial phone directory first.
@Jay L Gischer:
Arguably its not the search dominance per se; it’s that that leverages a stupendous revenue stream from algorithmically targeted advertising, and that in turn provides the means to be half of the mobile phone duopoly, to buy up other search-related and content provision services, to run server farms for “cloud” etc.
@JohnSF: Yes, Google has a dominant position in advertising, and is one half of the duopoly in mobile phones. Except that that half is based on a platform that is, in fact, quite open.
For instance, the Kindle Fire has it’s own ecosystem – it doesn’t use the Play Store. Even though it is based on the Android source, which is open source. Android is an open platform, whilst iOS is closed. This makes a big difference for me.
As to the advertising, I get that it wigs lots of people out, but where is the threat that is similar to the deplatforming we’ve been discussing? I don’t want to be argumentative so much as to understand the threat you see, and how it relates to this discussion, if at all. I might be argumentative in the future, but for now I’m just curious.
I see the problems that Twitter and Facebook have been bumbling through as the collision between ideals and reality. Both had a “free speech” slant at the beginning, then started to get complaints when the inevitable trash started to take over, tried to clamp down, then backed off at other complaints, and here we go again. They’ve been making it up as they’ve been going along, not just from a policy viewpoint but from a technological viewpoint as well. And I suspect that soon enough they’re never going to be able to give an explanation for why they kicked someone off further than “well, the algorithm said so.” Which gets REALLY amusing when you’re working with an AI bot that’s been trained on images and text.
The question is, how do you deal with information in a system where the craziest punk has a soapbox and can find other crazy punks? It used to be that such would be relatively harmless due to being immersed in the stolid surroundings of normal, ordinary neighbours who might listen to them politely but didn’t pass on the message. In communities, everyone was a gatekeeper and editor. It was only under unusual circumstances (famines/wars/whatever) that the paranoid view became contagious and would take effect. Now, of course, with the internet, crazy people can find other crazy people to go down the rabbit hole with and egg each other on in grand schemes.
It didn’t help matters when Zuckerberg himself claimed Facebook policy takes its cues from the First Amendment–which is obvious BS for a site with a TOS that explicitly reserves the right to remove “hate speech.” They just pretend to be following “freedom of speech” in those instances when they choose not to remove something and ignore the fact that they routinely remove other things.
@Jay L Gischer:
I only mentioned Google because of the Play store. But now that I’m thinking more about it, we could probably add Youtube in there. I’m not particularly happy that Google has its tendrils in just about everything and that it dumped its credo of “don’t be evil” in word and practice. I’ve been reading about a lot of Google’s AI work and it is exciting but also quite scary.
I have tried Bing and other search engines and yep, Google is the top dog – even for “private” searches where it is not using all the personal data it’s collected from you. There used to be some great meta search engines, but the various companies have closed up their APIs which makes that difficult. nowadays.
@Kylopod:
Your comment is a very good example of what I was trying to say about how these companies arbitrarily moderate content. There’s no consistency or transparency.
And now Facebook is taking out full-page ads whining about Apple’s new privacy settings as if every other company on the planet is obligated to unconditionally support their business model and let Facebook collect user data for its own benefit.
And then you look at what Facebook has done in other countries where it’s made agreements with governments – agreements where the result has either been domestic instability (and in some cases tribal and ethnic violence), or more government control and less freedom.
Much of my skepticism about anointing one or more social media companies as “public utilities” comes from what has happened in other countries where social media is a “regulated public utility.” It doesn’t end well. Clearly, there are some cases where a natural monopoly makes sense, in which case a government-regulated monopoly/utility is warranted – social media and online communications is not one of them.
@Jay L Gischer: If it helps, I get here every day (practically) from Newsblur. You might want to check it out.
@Andy:
There is a very limited amount of radio band, which can be effectively monopolized by well-heeled interests to keep the scrappy underdogs out. Particularly conservative, business-friendly well-heeled interests, as they tend to have more money to lose. Government regulation of the use of a scarce resource is pretty much required.
Meanwhile, building out a new website or app is pretty cheap. The tendency of tech to consolidate into two or three brands per sector is something else entirely — a different issue which should not be bundled with the first, as different issues likely have different solutions.
Social networks have a stickiness that might be somewhat comparable, since you want to be where the people you want to communicate with are. Google’s forays into social networks have failed. As have Amazon’s. However, WeChat, SnapChat, TikTok, Reddit, WhatsApp, Grindr, Discord, Twitch, Ferzu, Parler and others pop up overnight, so I’m thinking not comparable in the end. I had to look up a list there are so many.
I think there might be a market for a social media aggregator, but then I realize that I never want my TikTok and my Ferzu to meet (actually, those would probably get along fine… but LinkedIn and FetLife? No)
@Andy: @mattbernius:
Re the whole anti-trust, public utility, free speech, quasi-public forum debate, I think the calculus changes depending on which level of the technology stack is at issue. For example, I basically have no problem with a consumer-facing application/platform at the top of the tech stack, like Twitter, booting a category of users en masse. Don’t like it, make your own Twitter. Of course, the alt-righties did that and created Parler. Now, however, we’re drilling down a level in the tech stack and booting them off their web host (AWS) and their app distribution providers (Apple and Android app stores).
That move down in the tech stack is something that I think should give us all pause, because the deeper into the tech stack you go, the greater the barriers to entry for a would be alternative provider are. Maybe we’re still ok at this level, though, since they can at least still go offshore for hosting and distribute their app via other (considerably less accessible) storefronts to those who really want it.
What if we keep drilling further into the stack, though? What if search engines refuse to list their site or domain registrars refuse to issue them a domain name? What if ISPs refuse to allow access? What if the owners of the literal cables and switches and all the other hardware that form the internet itself refuse to connect them or any provider who services their site? At what point, if any, do the practical limitations to finding a workaround become so onerous and the gatekeepers so concentrated that service providers should operate on a viewpoint-neutral basis, either as a matter of informal norms and/or as a legal requirement?
For productivity’s sake I am going to try to stay off OTB and social media today, but I wanted to say that I appreciate the point you are making @R.Dave about the differences at different levels of the tech stack.
I also wanted to remind folks that a lot of this isn’t a new conversation at all and there is a lot of great work out there that unpacks these issue far better than most of us could hope to. The Electronic Frontier Foundation and Harvard’s Berkman Center are two great place to start.
@R.Dave: @mattbernius: There will always be a poison pill case for any set of principles. The poison pill will be in keeping with the organizations principles…but will destroy it in the long run.
Shorter version: 2 Aspirin fixes a headache. 200 Aspirin kills you.
There will always be an inflection point that forces reimagining of out values. Technology is doing this to free speech leading to outcomes that threaten this Republic. Eventually this will become an existential threat as factionalism will overun society as you get multiple groups of people living in differing realities. Today it is 2 factons…in 10 years it could be 5.
Im just not sensitive to the slippery slope or other side of the coin arguments for free speech. Can there be overreact of a flavor I dont like? Yes. That is a tradeoff Im willing to make in exchange of not taking the poison pill of treating this movement of seditionists and insurrectionist as if their calls for violence and perpetual baseless accusations of criminality toward the elected leaders they dont like is protected.
They lead to a very bad place
@Gustopher:
I work in the cellular industry so am well aware of that. But regulation is limited to spectrum management, not content or viewpoint management.
My view is that government should not be in the business of trying to manage viewpoints – I think it’s clearly contrary to the 1st Amendment and is something that government functionally is incapable of doing. If there’s too much consolidation in radio, then that’s a problem from a competitiveness/anti-trust standpoint.
@Jim Brown 32:
Along the lines of RDave’s comment, I think there has to be a limiting principle at some point. But getting back to my original argument, I think the slippery slope only becomes a problem where there is a lack of competition and a few powerful players can exercise essentially arbitrary control and influence over the national conversation.
@R.Dave:
I think the problem is that people did tell others to make their own platform if they didn’t like Twitter. So people made Parler. Then the same people said they didn’t like that those people had their own platform and unless they run Parler with the same censorship standards, they will excise them from the internet.