The problem with the tech hearings
zeynep.substack.com
zeynep.substack.com
The hearing yesterday was always going to just be a political show due to the current election but remember back to Zuck in 2018 regarding Cambridge Analytica or earlier this year with Zuck, Bezos and Cook. These congressmen are just clueless about tech entirely and apparently aren't seeking expert help in advance of these hearings.
As a big tech employee myself, I absolutely believe that a lot of these big tech companies should be broken up and I can clearly see things that we are doing that aren't good for the consumer. Congress just isn't asking the right questions and isn't asking for the right information.
The subsidiary problem, which can be remedied as long as the people want to learn and think they aren't too good to learn new things, is familiarity: It is hard to know what regulations will mean in a practical sense if you leave all your social media accounts to interns and suddenly have to bootstrap your way up from email all the way to Facebook, TikTok, Twitter, and Instagram.
Being old can be a handicap when learning about technical fields, but the old have no monopoly on smug, self-satisfied idiocy.
For example, Orrin Hatch asking Zuck in 2018 how they make money makes you assume that he doesn't use Facebook. Pretty sure anyone born in the last 30 years could immediately tell you that it's through selling ads.
The younger generation has grown up as big tech has and has a better understanding of the control and influence of these companies on their day to day lives that the older generation doesn't always understand.
I don't trust our politicians when it comes to complicated tech issues. I may not agree with how they handle current social media issues. But I don't think tech ignorance is limiting their understanding of this issue. It's a social issue and a media issue. It's not a tech issue.
In short the velocity of everything has increased by many order of magnitude. Anyone can publish anything quickly and share across aggregators instantaneously. The distribution of that data from the major platforms is completely opaque, unlike newspapers or tv which had known distribution and self-evident editorial decisions that apply to the entire audience all at once. In the new world, there can be infinite content and infinite eye balls all seeing different slices of it. It's all done by algorithm and no one except data-savvy internal staff have any idea how many people are seeing what and when.
I agree this can all be explained without much technical depth, but two problems with congress: first is that they are out of touch and still think in terms of the disappearing era of broadcast network news having massive monolithic reach and corresponding journalistic standards and ethics conferred by their unassailable moat of limited media options. Second, you can't legislate this if you can't reason about the second order effects due to the economics of how tech companies are run (ie. editorial in any old-school sense of the word simply does not fit any social media companies unit economics) as well as the information economy (ie. people prefer watching things that confirm their biases, and there are unlimited media options with many young people having no idea what a television schedule is).
They mostly seem to give the impression that Dorsey/Zuckerberg, etc, could just tell their companies to do a thing and then it magically happens.
They at least need to understand the tech well enough to recognize that fair, effective content moderation for billions of people is a more difficult and more complicated problem than any of them have attempted to solve.
I think social media companies could do a better job of being transparent and working to reduce political bias, but I do appreciate the challenge that they are facing.
They are fully capable of getting experts involved before hearings to ensure they ask the right questions-- even if the Congressperson isn't all that technical themselves. And they know this. They aren't unintelligent people.
We DO have a problem of leaders not being well informed on many topics they make laws about. But this is also true for finance, healthcare, foreign policy, and yeah tech.
When people complain about how old congress is, it's not like saying a 50yo can't code. It's more complaining about 75, 80 year olds who live off of name recognition rather than anything they actually do anymore, and some of them are clearly losing it. They could retire and live a great life, nobody's trying to take bread out of their mouth.
I can't comment on whether he was tech-savvy or not.
That being said if congress had any teeth at all they could instantly solve this technology problem by simply taking back the legislative protections that allow online advertising to thrive. If any of these companies had to operate without these special protections and thus compete with print media with equivalent protections they would immediately whither away.
Section 230 and DMCA are special protections that provide blanket immunity from any lawsuits. The protections intentionally exist to shield service providers but indirectly allow online advertising where it could not exist otherwise, there they are special protections.
And the staff are typically in their 20s or 30s, and use tech like anyone in their age group.
The elected members may not be great questioners, but the recent antitrust committee report had staff fingerprints all over it.
I disagree that we have lost anything substantive here though, or that new regulation around this might even be necessary; When in American history would all of the major book publishers refusing to print someone’s words mean the publishers should be made legally liable?
I'd say that tech companies a. have more power than book publishers, b. are more concentrated than book publishers historically have been, c. are covered by section 230 safe harbor, unlike book publishers
Moreover, why are we constrained by American history given that the challenges we face clearly are not?
The question then is, does the increased power and concentration move the internet companies to the category of something like a utility that needs public oversight, w.r.t. moderating content under section 230? To me that seems strange, other people’s attention does not seem like a utility. There may be other things large tech companies do with their power that mean they need to be regulated like utilities, but moderating user generated content so that users don’t leave or so that ads can be better sold over it doesn’t seem like it fits the bill. It’s not clear to me what is broken on section 230 specifically; I can’t separate a legitimate social ill from the politically motivated attacks.
The same way I buy gasoline, knowing that I am indirectly complicit in resource dispossession just by participating in that market.
> actually accept liability
My end goal is not in making HN liable for comments here, but I do think platform regulation is needed and if 230 has to be used as the "stick" to incentivize compliance, that might be necessary.
Major book publishers are legally liable, which is fine because they decide what they publish, e.g. they refuse to print some people’s words.
The tech industry was given immunity from that liability, with the idea they acted more like a common carrier than a publisher.
This is pure invention.
Information services' closest publishing analogue is letters to the editor. In print, the publisher has a far lower inbound volume of letters, and has to satisfy a less stringent SLA (letters are published daily or weekly). Publishers are able to moderate submissions effectively. Thus they can be held liable for user submissions they publish, even though they didn't write them.
Information services cannot do this. If they adopt the moderation model of a traditional publisher, their usefulness is lost because:
a) you'll need way more manpower to moderate (or the service cannot handle a high volume of submissions)
b) you'll lose the low-latency of an information service (i.e. nobody wants to wait 10 hours for their tweet to be posted).
Given these facts, treating information services as publishers by making them liable for what their users posted would have killed an entire industry in the cradle. Which is why Congress wrote the safe harbor into law. It has nothing to do with them being a "common carrier". You can read the law and it explicitly allows them to perform as much moderation as they wish.
Section 230 explicitly allows the tech industry to perform moderation. It was needed because ISPs were being sued for user-generated content and if the service had not moderated content they were found not at fault (Cubby, Inc. v. CompuServe Inc), but if they had moderated their user content they were found to have editorial control and thus be a publisher and legally liable (Stratton Oakmont, Inc. v. Prodigy Services Co.), this is directly relevant to whimsicalism's analogy.
A Wikipedia summary: "The court held that although CompuServe did host defamatory content on its forums, CompuServe was merely a distributor, rather than a publisher, of the content. As a distributor, CompuServe could only be held liable for defamation if it knew, or had reason to know, of the defamatory nature of the content. As CompuServe had made no effort to review the large volume of content on its forums, it could not be held liable for the defamatory content."
Section 230 was Congress clarifying which of those ways internet services should be treated. My phrase "with the idea they acted more like" is descriptive of what ISPs are / how they are classified, not prescriptive or Congress demanding how they act. It does sound like I should have said the idea was they're "more like a distributor than a publisher", instead of "more like a common carrier than a publisher".
I see it as a one-or-the-other kind of thing. Choose one of the 2 options: Censor absolutely nothing thats not patently unlawful OR censor things to keep yourself out of legal trouble because you ARE liable for the words distributed by your infrastructure.
Choose one or have a choice made for you I say.
More likely and what I personally believe happened; Hillary lost the 2016 election and the Democrats went nuts about Russian interference via social media. It HAD TO BE THE RUSSIANS as its simply impossible that our system of elections could produce a result where she didn't win. The social media elite realized "HOLY SH*T OUR PLATFORM CAN SWAY ELECTIONS??!!! COOL!!!". On that discovery they decided it was their moral obligation to sway our elections THE RIGHT WAY THIS TIME.
Personally, I dont care who, I dont care why, I dont care if its "a very good reason". All forms of censorship have no place in a free society. Not by the government. Not by private enterprise. Sure as hell not by the unelected likes of Jack Dorsey and Mark Zuckerberg.
Put another way, why are you discussing this here and not on 4chan?
^^ Your words - not mine.
When HN takes action to remove unproductive or incorrect comments on posts its called "Moderation".
When a media organization "Moderates" a US newspaper by locking their account over a political story its called "Censorship".
“Unproductive and incorrect” is not patently unlawful, so censoring it would open YC to liability for everything posted here, under your proposal.
No, the reason they have it is so that they can moderate without fear of litigation; before 230 unmoderated content generally did not expose companies to liability unless they knew of unlawful content (distributor model), but any attempt at moderation made them liable for everything on the site, whether or not they had specific knowledge (publisher model); 230 specifically was passed to negate liability as a publisher for user submitted content even if the platform engage in moderation.
https://www.businessinsider.com/jack-dorsey-ny-post-remains-...
Moderation is a long way from outright ban-hammering a US newspaper over their posting of a news story. What Twitter specifically did here more or less makes them an unofficial editor at NYPost. Thats not acceptable.
Ask yourself this simple question - if the roles were reversed and Twitter was run by a Republican and they locked the Washington Post's account because it printed an unfavorable article about Donald Trump - what would you call that action?
Yes, companies aren't bound by it legally, but it'd be a great moral guide to follow if that's what they're after.
The real action to take here is antitrust type stuff to limit the size of any single media outlet or platform anyway.
Did 230 say anything about that or did the law have no stated reason for that difference?
Why can't a print magazine do that, but a website can? Did 230 say anything about why those are treated differently?
The law itself doesn't, because laws generally don't. The discussions around the law offered a number of reasons; one basic deontological idea, as I recall, was that a website moderating user content was functionally more like a choosing which third-party publications to offer than actually publishing. But probably the more significant argument was consequentialist, that if companies were liable for everything if they tried to moderate at all, the into viable large scale sites would be completely unmoderated, with no attempt to preemptively identify illegal and offensive content. And that's why it was included as part of the Communications Decency Act.
On the other hand, a print magazine hires employees to produce a bespoke product sold to the public, so whatever they publish in the magazine is understood to be a representation of the magazine company. In fact, the owner of a website would indeed be liable for posting infringing content if it were understood that the site owner was representing themselves (e.g. if I posted stolen photos from your laptop onto my self-hosted blog).
> Choose one or have a choice made for you I say.
This is silliness.
There's a ton of legitimate reason to have moderation standards, community guidelines, etc, for user generated content beyond "patently unlawful."
To say that any forum that wants anything stricter as a moderation standard must be liable for any malicious actor's behavior is to doom so many online communities.
You're so focused on your extreme "any censorship is bad" "principle" that you've lost sight of the nuance of the real world. A church, for instance: should they be able to have a discussion forum with standards around profanity without becoming liable for anything anyone in the world might post? It's not illegal to swear at everyone...
This is what Twitter does. Permanent ban for even the slightest offence.
No, the solution is to make all censorship illegal on large platforms, because otherwise the unsolvable question of "whose standards" it is that determine "offensive" arises. Right now Twitter employees consider themselves this high court and they will ban you based on their own political leanings, which only 50% of the country agree with.
People pay good money with good reason to have things curated for them. The New York Times decides what to publish and your local TV station decides what to air. That they choose to not print your article or letter to the editor or air your content is not out-of-bounds censorship. Why must Twitter or Facebook or YouTube be held to a looser standard? Or should the NYT print it all as well?
The problem is the grey area, and people's political views and biases strongly affect what they think is or is not following the rules. That makes the rules extremely hard to enforce in a manner that keeps people satisfied.
There are many examples of this in the recent past, but perhaps the most surprising is those Facebook accounts who posted material indicating that Kamala Harris's record on gun rights in California is not encouraging for gun owners (not difficult material to fact check -- she activated the handgun roster) and found themselves suspended because, according to Facebook fact checkers, "Kamala Harris is pro 2nd Amendment".
More controversially, it is now being argued that Twitter's recent suspension of the New York Post was entirely baseless: that there was no reason for the social network to conclude that the news agency was reporting the products of disinformation.
The way property rights usually work, you aren't able to force people to treat you fairly in a particular situation, but you can take them to court over it and be made whole. The system certainly functions as an aggregate encouragement to treat other people fairly but only in rare situations are you going to see people forced to perform in a particular way. People would still lose their accounts but expanded rights would give them a way to pursue a remedy.
Why should these belong to the users when the software hosting them, the servers running the software, and the infrastructure running the servers are all owned, operated and maintained by a third party?
That's like saying I should own my table at a restaurant simply because when I walk in and sit down, they serve food to me there. Using a service has never entitled you to an ownership share of the establishment.
It's kind of like how you own the assets in your bank account and have a right to the returns on them even though the building, the accounting software, the IP and internal best practices supporting it are all (usually) owned by the bank.
We need to recognize the property interest of users in what they built -- in their content, following, online store, &c -- as a separate matter from the from the platform's ownership of its infrastructure, best practices and organization.
We "need" nothing else for shouts to the void as nobody gives enough of a shit to set up and use more guaranteed terms and any arguments are just as timely, sane, and useful as saying we need to make blood oaths legally binding and have notaries for them. No we don't we really don't.
And on every social media platform with which you have an account, you agreed to terms of service which very likely grant the platform a "worldwide, non-exclusive, royalty-free license (with the right to sublicense) to use, copy, reproduce, process, adapt, modify, publish, transmit, display and distribute such content in any and all media or distribution methods now known or later developed." (copied from Twitter's TOS, but every other platform is basically the same.)
data and infrastructure are not separate matters. Just as, with a bank, your account is a liability unless it makes them money, with a social media platform your data is a liability unless it makes them money. And in exchange for a free platform (or free checking and a pen) you gave certain rights of ownership away. If you want to "recognize the property interest of users" you have to recognize it in that context.
Yes, because the money being deposited is already owned by the account holder, banks would not exist if the bank became the owner of the money deposited, such an arrangement makes no sense. However, banks are within their rights to close your account at any time for any reason, they just have to make sure you get your money back. If you're suggesting that websites should be required to "return your data", that seems reasonable to me.
I would be able to choose the provider that I trust the most (or self-host) and even move to another provider, while still being able to participate in the global conversation.
This way we'd also have different fronts (apps, websites, clients, etc.) to this network, each of them editorializing it however they please.
I'd own my account, my posts and I'd be able to choose which front I want to use to interact with the network.
Users would be responsible for what they post and fronts would be responsible for what they show, if they choose to filter and/or promote content.
At that point you are not at the will of one centralized company. If you feel like your rights are being violated, you can change provider.
Following your parallelism with banks: if I have my money in bank A and I don't like how it is operating I am free to change from bank A to bank B.
Channels, accounts, and online storefronts don't stay online of their own accords. They require hardware, software, and occasional power.
If I decide to unplug an old PC that was acting as say a dedicated game server would I be depriving many people of property? Which would make my own ownership of the server a lie?
The only way that would be sustainable would be a contract to pay the host. That is basically existing contract law with a self-aggrandizing name attached to it. It contributes literally nothing new conceptually.
My biggest concern is that the "they're private companies, they can do whatever they want" argument applies to ISP's, too - they're private companies as well. We've already seen Epik being "deplatformed" by PayPal (also a private company) for allowing controversial content to pass through their infrastructure. I'm not sure I agree that Twitter should be forced to reinstate the New York Post, but I'm having trouble disagreeing that they should carry more liability for what they do allow if they refuse to.
This isn't true, that's just your impression from within the twitter bubble. Even if it were true, you can easily contact journalists through many avenues, being easy to reach is part of the job description. Getting banned from twitter is a totally unimportant internet triviality.
There's literally no way to get more accurate news faster, by following sources you trust.
Twitter is the most important journalistic tool? Not e-mail? Not end-to-end encrypted chat services? The entertainment website made up of mostly celebrity gossip, memes and culture war propaganda? I think you're the one living in a cave my friend, a cave built, owned and operated by twitter. Just because it's popular with many journalists doesn't mean its important. There's no doubt there are some amazing journalists who use twitter at some points during their day, but the overwhelming sum of what goes on twitter can hardly be considered "journalism".
> There's literally no way to get more accurate news faster, by following sources you trust.
That's your opinion, not a fact. There are many people who would say using twitter as your primary news source is a terrible idea. News on the internet is a billion dollar business, twitter is just one of many innumerable sources of real-time news on the internet.
Point #2) Twitter itself is not a news "source". News Networks however ARE. News is reliable on Twitter because you follow only who you trust, and you make the assumption that you're feed isn't being monkeyed with by Twitter. You're out of touch if you don't realize you can get more accurate news faster on Twitter than thru News Networks. Trust me on that. Plus what you get from the networks is all propaganda anyway, and generally not worth my time.
Google end-to-end encryption journalism. e2e encrypted services keep journalists and their sources alive, it far outranks the petty trivialities of twitter.
> News is reliable on Twitter because you follow only who you trust
Every social media website has a "follow" feature, nothing is special about twitter that makes the publishers more trustworthy than any other social network.
> You're out of touch if you don't realize you can get more accurate news faster on Twitter than thru News Networks. Trust me on that.
I don't need to trust you, we can agree to disagree on your opinion that twitter is a good place to get news.
Every legislator, politician, celebrity, and public figure is now using Twitter is their primary way to reach audience for written communications...except the banned ones.
More important however is the issue that Twitter can only block access to Twitter. That it is difficult to switch ISPs only matters because with few alternatives an ISP blocking a site, or you canceling your contract, would lead to that site being inaccessible. The general policy 'companies shouldn't use their position to alter access to other companies' is more defensible than 'popular social media has to pretend to be a utility for a given value of popular and social media'.
You don't think Twitter is analogous to phone service. I do. That's the disconnect here.
I think both of them should have the discretion to do whatever they want while the government should constantly work hard to ensure there is always more competition. It is the government which has made it impossible to create credible alternatives to Google, Facebook or Twitter. Firstly, US government will try to kill any foreign competitor in the name of national security or some other straw-man. Domestic level the regulation is darn too complex for anyone to succeed with reasonable capital.
Do you really want nose-ring kids in Silicon Valley who're still mad at their parents having the power to shut down Presidents, Heads of State, and Media organizations?
Come on man.
Re: the change in reach that happens, I don’t see how this is different from a television network canceling a host.
Thru a series of events no one planned we just evolved into a situation where a public figure's main way of reaching out directly to their audience is over Twitter.
The correct analogy here is the phone (not TV networks). When the telephone system was originally a monopoly (or even today) imagine if the phone company had started disconnecting lines or censoring calls? Alexander Graham Bell would've been hanged from the nearest tree.
But if you want to use a TV Network analogy it's not like canceling an Anchor, it's like canceling a 'viewer'. It would be like if CNN had the power to stop any individual from either consuming CNN content or creating CNN content. Total Godlike control over the individual.
Here's the crux:
Q) Can AT&T censor texts or cut service if an AT&T employee gets triggered by the content of speech?
A) Not just no, but hell no.
The same applies to Twitter for precisely all the same reasons.
...so people who want to live in an online safe-space can run with "Safe Space" mode on.
But AT&T can't cut service, nor cancel texts after sent, etc, and for all the same reasons you'd give for phone companies not censoring I'd claim apply to Twitter.
As an aside, the constant bashing of "twitter kids" and allusions to political speech like "all lives matter" being censored (which is not happening, as far as I know) seems counterproductive.
But hey, easy for me to say since I've pretty much abandoned twitter/facebook/etc nearly a decade ago and avoid things like snap/insta/etc for things like a mastodon instance.
And your reply about that would've been: "Hey people can still send information using radio and telegraph, so they should just come to terms with that and stop trying to rely on phone systems that belong to Bell?"
If people want to be useds, that's their choice.
I'm a huge fan of decentralized social media, web3.0, IPFS, ActivityPub, Mastodon, Pleroma, and the Fediverse, and all that is the FUTURE to replace Twitter. However for today Twitter needs to stop censoring.
Yeah, well that's not a battle I want to fight. Id rather twitter/fb keep pushing people off their platforms and kill themselves via death by 1000 cuts.
In your social media example, the only case where it doesn't involve easy alternatives is the size of a public audience; so it would be more like a television host being dropped by the network because they don't like their views. If you are trying to communicate with friends and family, or even with large section of the public, there are absolutely alternatives. What you want is more akin to compelling a television network to air someone because they can't reach as many strangers without that television networks large audience.
The result would be opening up any and all moderated forums to a tsunami of lawsuits. Twitter would survive, they have armies of lawyers. HN would be shut down.
A party that hates regulating business suddenly sees the light when powerful businesses start removing them from the public square.
A party that pretends to be for free speech and free journalism suddenly has a change of heart when it benefits them going into an election.
A party that purports to stand up to big business suddenly supports the largest megacorps in the history of man because their interests are temporarily aligned.
A party that thought Citizens United was terrible because it gave the rights of citizens to corporations suddenly cares very deeply about Twitter/Facebook/Google/Apple's freedom of expression when used to eliminate critical voices on their platforms.
A party gets rid of the filibuster for judge nominations and then gets mad that they can't filibuster the other party's judge nominations -- or a party deciding it's suddenly ok to seat judges during an election.
Both parties are about power. Neither party has any remaining principles.
Edit: Grammar.
Also, I don't think ideological consistency the way you define it would even be good thing. Standing up to corporations does not mean corporation is wrong in each case. One also need to severly mischaracterize objections to citizen united in order to make it incompatible with twitters right to censor.
This isn't new or unique to paypal. The Credit Card companies and banks that underly pretty much every fancy modern method of money transfer have and do "deplatform" people. And they've done it for ages. Porn companies can't use conventional credit card processors, and there's a long history of banks and credit cards refusing to do business with certain companies (https://en.wikipedia.org/wiki/Deplatforming#Financial_servic...).
Whatever your overall opinion of deplatforming, the idea isn't anything new or unique to twitter. ISPs were singled out (they aren't anymore, net neutrality was rolled back, now your argument only applies to phone companies).
The same is true for cellular providers: They lease spectrum that is owned by the public.
Free speech isn't free. It's a responsibility, and we should take care to fulfil that responsibility with caution and diligence.
To pick two recent examples, it's useless to wait for people in suits to "conserve" the gray wolf or enact fuel economy "standards." I vote with my wallet to pressure companies constantly to produce more fuel-efficient cars, and I promote wolves and wolf appreciation in my local community.
This is orwellian. I can already tell it's going to set up the argument that actual censorship is necessary to counter the "censorship" of harrassment, disinformation, trolling, distraction and piecemeal leaks of hacked materials, such as Trumps tax returns, the Panama papers, the Snowden files ...
It reminds me of the idea that "words or violence" or even that "silence is violence", justifying actual violence - in this case actual censorship - as a defensive measure.
Censorship is a very clear word. If someone wants to say something, and someone else wants to listen to him, and you interfere, that's censorship. If someone wants to say something, and someone else says something else that distracts from what the first guy says that's not censorship.
A common line to justify deplatforming is that you have no right to an audience, but it really is appropriate here: The mainstream media has no right to an audience. If people prefer to listen to Alex Jones, then maybe the mainstream media should ask themselves how they managed to fuck up peoples trust in them, rather then whining about the censorship of Alex Jones distracting from them.
That is not a recipe for good government regardless of political affiliation.
What the big tech companies are doing seems to have crossed into criminal territory.
Wait, who is this? I know representative Jim Jordan was a wrestling coach and current senate candidate Tommy Tuberville was a football coach at several places.