Ask HN: Is it technically possible to evade big-tech censorship?

78 points by sequoia ↗ HN
I am Trump opponent. The implications of Trump being disappeared from the modern internet, however, are very troubling. He's a sitting world leader, but Zuckerberg, Dorsey, Bezos, Cook and Pichai snap their fingers and he is effectively erased from the internet almost overnight. Of course Twitter should be able to block him, they're a private company, he can go to another network. But when he goes to another network (Parler) it's shut down with 24hrs notice by Amazon. To anyone who's the least bit of a political dissident, alarm bells should be screaming in your ears. If a handful of tech execs can (and will) do this to the US president, what chance would you stand against them, or anyone, if they want to silence you?

My technical question: What would it take to stay online in a situation like this? Twitter/Facebook/Reddit etc. are out, running your own network on AWS is out, Cloudflare DDOS protection is probably out.

Would it be possible to host your political views on any commercial hosting vendor in this case? Is running a server at home enough, assuming you can avoid DOS somehow? Would you have to start your own ISP, and if so, could your ISP be kicked off the backbone?

I'm interested in this as a technical (not political) question: is there any way to speak online if big tech decides they don't want you to?

89 comments

[ 3.4 ms ] story [ 175 ms ] thread
It's not just tech companies which are cancelling. By Parler's own admission from their CEO, banks and payment providers, logistics, and law firms are cancelling on Parler. And friends and acquaintances too, I imagine.
I've been looking at this situation in practical terms. There are all sorts of ways to communicate online even when big tech does not want you to. The problem is that these solutions do not scale. Big Tech alternatives are also notoriously obscure and/or hard to use for the average person.

Additionally, running any service in the modern day involves dozens of third party companies. Even if you solve the technology problem there is nothing stopping the bank from closing your account. What about your accounting firm, choice of payroll software, or even the landlord? All of these links in the chain are easily intimidated into cutting you off. What happens if someone sues you? We've already seen legal firms refuse to represent Parler.

I really don't see a way out of this situation, to be honest.

The President of the United States has the same venue for disseminating information as his predecessors: White House press conferences.

As for the five people you mentioned, the CEOs of the "big tech" firms, are not one entity- you've got to go well beyond the pale (as has happened in the past week) to get them all to agree that it is their company's best interest to deplatform you.

>The President of the United States has the same venue for disseminating information as his predecessors: White House press conferences.

Pretty much. The President has had access to the bully pulpit for over a hundred years. The decision to rely so heavily on Twitter instead of traditional channels is a strategic failure of the Trump administration. Normalizing this and ceding control of presidential communications to private parties might be a lasting legacy and mistake.

Would he have gotten so far in politics without Twitter? Trump’s style of direct communication is very powerful for maintaining the attention of a passionate base which is the source of his political power.
I think this discounts the power of Fox News and Sinclair Broadcasting. I agree it's a bit of a chicken-and-egg problem, though.
Yes, that's the tradeoff he chose.

Neither Twitter the company nor society at large is obligated to play into the game where a candidate figures pit how to make a non-traditional platform work for him and then expects help from others when it no longer works for him.

You can't live by the sword and then passionately advocate pacifism when a sword is raised against you.

The press conferences require you to go through "the press". If big tech censors press they don't like, then the press conferences will only include journalists approved by big tech.
NYT and FOX are not big tech. They are big news.
Big tech can control the flow of the population to news sites (one of the reasons the two industries often fight each other). Twitter suspended NYPost account for instance. And what if tech and news industries join forces to erase someone they both don't like?
The Sinclair media group is one of the biggest players in US politics and basically owns the local media landscape.
Haven't the press switched away from at least one of his conferences? I think it would be interesting for him to try to hold a press conference to see if the broadcast media would even carry it
They switch away when he begins spouting unmitigated lies, which is a responsible form of journalism.
Do they? When he spouted the nonsense about "disinfectants" they didn't switch away - they made it front page news for days.

Be consistent, either refuse the cover the lies, or cover all of them. Don't just cover the lies that make you look good. Or more accurately: Don't just cover the lies that get lots of outrage (AKA views), without getting "too much" outrage of course, because that gets people killed.

Yes, they do. While it took years, the falsehoods he spread during the pandemic and during the run-up to the election convinced many newsmedia outlets to cut him off. In other words, they learned. Consistency is not to be desired above all else---it is healthy to change one's mind and behavior when presented with new evidence or circumstances, which is what those outlets did.
Harder to blame "big tech" in that case!

I guarantee you that if he called into Fox, his comments would be carried by every media outlet and discussed exhaustively on Twitter.

Are you saying the solution is to make sure you have a large and powerful media organization on your side?
Of course not. Where do I say that? Please don't attempt to slide me into a position I didn't take. That's bad faith.

For the record, if Trump called up Maggie Haberman and gave an on-the-record interview, I'm sure that would get covered too!

I think the test is probably whether White House statements are reasonably available to people. For instance, lots of media don't cover any White House press conferences, never mind carry them live and so on.

The discussion around social media then gets split in two; what is currently legal, and what should be legal.

Trump probably doesn't have any current legal right to force Twitter (or CNN) to carry his message. But maybe there should be some laws about what Twitter can't moderate.

A: No he does not. The White House can talk as much as it likes, but without the press to actually disseminate the words no one will hear them.

If Newspapers, TV, and YouTube (etc) all refused to carry the words what other avenue does the President have? An emergency alert to mobile phones?

And B: That doesn't answer the question posed. The question is not about the President, it's about a political dissident in general.

The First Amendment's Freedom of the Press protections align with this outcome. It is intentional that the President isn't able to force the President to cover him. The government isn't supposed to be able to force the press on what they must say any more than it may force them on what they cannot say; these two compulsions are two sides of the same coin.

It's an extraordinary circumstance when it's not in the press's best interest to report on press conferences (i.e. it's generally newsworthy information being freely handed out), and this was an extraordinary presidency.

Yes they are not one entity, and yet they all more or less simultaneously decided trump (and parler) “violate our TOS.” It’s precisely this sort of collusion, with the figleaf of “our TOS” (give me a break, these were acts of political expediency not previously unknown TOS violations that all suddenly came to light) that I find alarming. If he/parler were kicked off one or two places it wouldn’t be as scary. It’s the blackout from social network to app stores all the way down to web infrastructure that I found shocking.

If it’s hard to get upset about trump/parler getting banned, just sub in your favourite political or artistic group and gauge your reaction.

The problem with that of course is exactly the same as with Twitter: the press needs to a) show up in the first place, b) cover the story at all even if they do, and c) relay what was said without adding or removing context.

Which, just like any other private entity, if they want to not enable violence then they don't have to do any of those points. I.e. just like twitter.

Honestly I doubt there is. So if you co-located in a large ISP, you will still sign a contract which most of the time they put it very generic morality clause for types of content so they can pull your service at a moments notice. So even assume you had enough money to get the infrastructure you need to plug into the net, and you usually do this via co-locating and buying net access. Only way I can think of right now is you'd literally have to go the the length of setting up your own first tier ISP (like Google, Facebook etc do), which is doable but not cost effective. Cause outside of that you are either running on someones cloud (AWS/Azure/GCP) or you are running under a contract of an ISP that has similar terms. Maybe I am missing another avenue, though, always possible.
This is an excellent point. Every ToS has clauses in it which can easily be interpreted against you.
> your own first tier ISP

And if other ISP's refuse to peer with you?

Maybe we need another kind of law, above and beyond 230, requiring large ISP's and cloud providers to act as common carriers in exchange for certain protections.

I don't have a fixed concept on the details here, it's just an idea I had.

Think about this: Does Google have a legal or moral right (or responsibility) to delete emails calling for specific and actionable violence against someone? Or is email too private for that?

> Does Google have a legal or moral right (or responsibility) to delete emails calling for specific and actionable violence against someone? Or is email too private for that?

This is the rathole we're in. If someone does plan over gmail, does all of gmail get shut down? What about the ISP that served it? If someone plans a violent attack over the phone, does a company's phone network get shutdown?

How far are we willing to go for our rights and privacy?

It's very simple and nothing new. If the company receives a court order, they have to act on it. Or they get shutdown. In most countries it's been like that forever.

What you're talking about it preemptively policing and taking actions. I'm sure politicians would love that but we're not there (yet).

If a spammer tries to use this law to say they're providing valuable information to the public and they're being censored, what's the result?
> So if you co-located in a large ISP, you will still sign a contract which most of the time they put it very generic morality clause for types of content

I don't think that's true. Hurricane Electric for example is a major provider and the AUP [0] only prohibits illegal and/or technically damaging behaviour.

Some of the others have language like that in their AUPs but it doesn't seem to be universal.

[0]: https://www.he.net/aup.html

I'd argue that in Point 1 of the AUP you linked to gives them all the out they need: "...adversely affects Hurricane's legal interests." and then the two sub bullets, #6 & #7 both give them lots of grey area to terminate hosting. Not to mention they state you will enforce an at least equally strict and enforceable agreement with any of the users utilizing your system.

To be fair, I think most reasonable people would say sub points #6 and #7 seem pretty fair, at least on the face of it they do to me. But what one person considers a threat another considers banter and hence we have scenarios where people can easy run afoul of some morality judgement by an employee or exec at the company. It doesn't have to be as blatant as a death threat it could be some banter via an innuendo and suddenly they ban your account. Sure, you can sue, argue your point, but either way your business would be offline. Just the first part I quoted, all they have to say is your content or users affect their legal interests in some way and again subjective and your done.

I'm not sure "don't incite a violent mob" qualifies as a technical approach, but it seems likely to meet your requirements here.
The "dark web"?
I was going to mention tor, but it too isn't a very scalable service and can be (almost) just as easily taken down by disgruntled hosts or government sanctions.
Hacker News needs a black bar across the top while this assault on Free Speech continues. First they came for the Trump supporters, I said nothing...
First they came for the Nazis, and I said "acutally this is great for the Jews, socialists, trade unionists etc" and we don't have to do the rest of the poem.
Using Nazi tactics to defeat Nazis makes you consider the rest of their fascist ideology: total state control. 1984
(comment deleted)
First they came for the Trump supporters, and we all got downvoted :)
When they came for the nazis, I said nothing for I was not a nazi. Then they came for the ku klux klan but I said nothing for I was not a member. And then they never came for me because I do not use violence to achieve my political goals aka the definition of terrorism.

A society of tolerance can only exist if it's intolerant to intolerance. It's really simple. Our grandparents would be ashamed of those of us who say we should make sure the nazis have a platform in the name of free speech. I know at least my nazi killing grandfather would be.

Here's something even simpler: Trump supporters might try to do the same to you if they get power again.

These are Nazi tactics being used by the Left.

What would they do? The government is doing nothing right now. For the next week, they're still in power. It's only private companies. The government isn't forcing these private companies to do anything.

And you obviously have no clue what Nazi tactics mean. If you want to talk about suppression, we should talk about Wikileaks, Snowden, Swartz, or the NSA. When presenting my research on Snowden's impact on encryption is 2015, we focused on the legitimate efforts of individuals to circumvent government overreach into privacy. This is so different my head is spinning.

Yes, by remaining small-tech.

It's basically like how the US keeps airstriking "the leader of Al-Quaeda" every six months. If you're high profile enough, and causing trouble to enough people, eventually it will come back to you.

Trying to overthrow the Senate is evidently the line you can't cross.

> But when he goes to another network (Parler) it's shut down with 24hrs notice by Amazon. To anyone who's the least bit of a political dissident, alarm bells should be screaming in your ears. If a handful of tech execs can (and will) do this to the US president, what chance would you stand against them, or anyone, if they want to silence you?

Your question is weirdly lacking in context. There was an attack on Congress, whose nastiest bits were openly planned on Parler. It's not like these companies just suddenly decided to cut ties because they didn't like Trump's views. They've actually been extremely tolerant of him, giving him policy exceptions left and right (e.g. if he wasn't the president, he'd probably have been banned from Twitter years ago).

> Would it be possible to host your political views on any commercial hosting vendor in this case? Is running a server at home enough, assuming you can avoid DOS somehow? Would you have to start your own ISP, and if so, could your ISP be kicked off the backbone?

> I'm interested in this as a technical (not political) question: is there any way to speak online if big tech decides they don't want you to?

Obviously, yes. Donald Trump has a website (https://www.donaldjtrump.com/), and it's still up. He can speak online there.

Here's an idea I've been playing with: you can buy your own AS and small IP block for a few thousand dollars. There are many data centers that will give you BGP access to then announce that IP range and route it to your servers. The result will be that whois tools will not actually report which data center you're in, but will report "Uncensorable Inc" or whatever you choose to call yourself. No one will know which company to tweet at to try to have you deplatformed.

I'll be trying to do this to see if it works, if I succeed I'll post a story or two here with a demo and how-to instructions.

Wouldn't people just tweet at the companies offering you peering?

Or put another way - if this works, why haven't spammers done it?

As I understand it, hosts tend to have AUPs that restrict speech more than the first amendment does, but peering arrangements don't do the same-- correct me if I'm wrong, though!

There are data centers that are popular with hax0rs because they don't validate source IP on outbound packets, thereby allowing people to run reflection DDoS attacks. You can visit hacker forums and get a list of these. There are still a ton of AS's in the world doing this, and no one is unpeering them even though they do huge damage to core infrastructure. On the other hand, if you try that on a well-policed data center like AWS, you'll be either blocked by automated tools or dropped as a customer (or both). That illustrates the difference between how customers are treated and how peers are.

Spam is a bit different because email providers just block IPs and AS's that are major spammers. This is done with a simple firewall policy and has nothing to do with peering.

Well, not terribly hard to find the two upstream ASNs and figure out where the datacentres are.
Yes. Self-host like how 4chan and Gab are doing it.

Although its easier said than done, but when they're constantly de-platformed from the face of the internet; they're left with no other choice but to self-host.

(comment deleted)
> The implications of Trump being disappeared from the modern internet

Trump has not been disappeared from the modern internet.

He controls, for instance, a very significant online presence here:

https://www.whitehouse.gov/

He also controls a broader presence through a number of subordinates, see, e.g.:

https://www.usa.gov and the whole list at https://www.usa.gov/federal-agencies

And, in his (unusual for a President) parallel life, he also controls a web presence through his private business (delegated to subordinates headed by immediate family members):

https://www.trump.com/

Not sure what I am missing here but the obvious answer is an offshore host right?
This is an inherently political question where the context is critical. If you are working with North Koreans on detailed plans for disruption of infrastructure for example then you would encounter a different alarms and enforcement than a biotech researcher doing gene splicing experiments. Email lists are extremely easy to set up, make robust, and transfer as necessary, so using tech like that to simply skip around big tech might be an alternative.
As a technical question: no, if everyone you want to peer with wants to kick you off, you can't be on the "internet" in the regular sense.

Signal was funded by Radio Free Asia, which historically got past the general form of this problem by broadcasting radio stations from right outside the borders of a hostile country. St. Maximilian Kolbe did the same with amateur radio.

I think the more interesting question is some blend of technical and political: how do you get across an idea that people aren't exposed to? Radio Free Asia/Europe/etc. and Kolbe all faced this problem, as did the samizdat publishers, the organizers of the Underground Railroad, etc. You have to be creative, but it turns out that getting ideas across is not an unsolved technical problem, it just requires a bit of work. If the goal is to make people aware that an alternative to your authoritarian government exists or that an organized scheme to take you to freedom exists, you can communicate it, even without anyone else's infrastructure.

Neither the purpose nor the effect of Trump's ban is to prevent people from knowing that his ideas exist. They are well-known and popular. It denies him the use of mass media (and even so, newspapers and TV stations haven't banned him), and it disrupts the ability of people to coordinate. Regardless of the merits of the ban, I think it's pretty clear this is a different sort of problem, and yes, if you want to use wide-reaching, high-bandwidth, reliable infrastructure that other people built, you need to make them okay with you using the infrastructure.

An analogy can be made to spam. Obviously everyone knows that you can lose money by wiring it to some random dude with a foreign bank account; the "censorship" of 419 scams does not in any way restrict what people know about, and spammers get denied the use of infrastructure if they make the people who built the infrastructure unhappy.

In terms of technical means, I'm not sure how to permit an existing, major political party to have access to infrastructure however they want without also, by implication, permitting spammers to have access to that same infrastructure however they want. The only distinction between those is human judgment.

> Is running a server at home enough, assuming you can avoid DOS somehow?

If your content was lawful, and if the Trump Administration had not repealed net neutrality, it would be enough legally. It wouldn’t be enough technically if your ISP opposed you.

> Would you have to start your own ISP, and if so, could your ISP be kicked off the backbone?

As a technical and not a legal/contractual matter, yes, you could be kicked off by whoever you connected to directly.

You’ve claimed that you are asking a technical rather than political question, but virtually all of your specific examples seem framed as legal questions as obviously any connection to a third party can, as a technical matter, be abandoned by the third party.

The technology you're asking about is called a blog, and there are myriad ways to host one.

The thing that's harder to do if big tech is gunning for you is to maintain an audience of millions, and get paid for that traffic. But you can put your manifesto up anywhere you want.

It's almost like people don't understand that private companies don't have to guarantee free political speech, especially to nazis.
I’ve seen a lot more racism fueled violence on BLM protests than with Trump supporters.
I believe all people calling for any sort of violence on social media should be deplatformed.
That teaches them nothing and gives them a martyr complex. Put them in $social_network jail for some length of time and teach them what is and isn't allowed.
As long as we agree the correct response isn't inaction.
Exactly, there definitely shouldn't be inaction. However, after BLM protests I've seen action only towards one side, which is exactly what led us here in the first place.
I can understand this point of view, but will the same standard be applied to anyone who advocates punching Nazis?
That's up for private companies to determine as they risk finding a balance between alienating their own users either by inaction or too much action.

As for free speech in public spaces as restricted by the government, the standard comes down to imminent risk of harm. You're not allowed to scream "Fire!" in a public building if there's no fire. Private companies like Twitter and Facebook appear to try to hold themselves to a public standard but private companies are allow any standard they please. If you're in a movie theater and start yelling throughout the movie, the theater isn't restricting your free speech by kicking you out.

What about Doctors who were silenced because they said Hydroxychloroquine is effective against COVID?
It is currently very possible to host absolutely any political views on big tech platforms with no repercussions. Conservative media, libertarian media, communist media, scientology media are all proliferating like never before. What you can't do is incite violence or direct hatred at groups based on their innate character. At least, not with impunity.

Shutting down Trump is not at all an unprecedented move. Social media shuts out evil world leaders all the time.

Just like P2P bitcoin, can any big tech shut down bitcoin globally? Can someone develop some chat services based on HTTP/3, WebAssembly and P2P like bitcoin? Also Zero Knowledge Technology needs to be utilized.
When one says “big tech” I think it’s usually used to represent Twitter, Amazon, Google, Apple, etc “big SV FAANG’y tech”. This seems like controlling content the same way large media outlets always have (ie newspapers, magazines, etc), they have no obligation to amplify your message. You can still take your message to a smaller player if they’ll have you (ie a smaller newspaper) or if your message isn’t wanted by others at all you still have your own machine in a closet (ie a printing press and pamphlets).
I'd argue the only meaningful online social network is git, and maybe soundcloud/bittorrent, the rest is just somebody elses variable source of narcissistic supply.

You could use one of those p2p-blockchain messengers to talk to other people who use p2p-blockchain messengers, but then you are talking to people who use p2p-blockchain messengers.

It's really a question of how much breadth you want/need. The internet itself reduces to a politically atomized individual sitting alone at a keyboard, struggling to connect, while being neutralized from all sides by random noise to stimulate both outrage and helplessness. It really is the totalitarian dream. You'd have to leave The Internet proper as we know it now, and create the next one, just as AOL/Compuserve were walled gardens that the internet grew outside of, a new network that resembled the 90's era internet could emerge in as little as a few years.

Think about what Linux was: a new kernel for bare metal that broke out of the constraints of an OS landscape that looked a lot like today's internet giants. At the time, global communication was novel, today, privacy and regional networks are novel. Most linux users had a bsd/solaris/windows machine to connect to the legacy world, so you don't have to design the whole thing at once.

A p2p-reddit that used a blockchain that operated with a kind of old fidonet/UUCP/nntp over bittorrent as transport could be a source of new community. A bare metal blockchain client kernel for open RISC and ARM processors that uses a wireless mesh, and includes a kind of regional pub/sub pattern would do it.

The tech is there.

Move the discussion offline. The state can still get to you but at least you will mostly be out of big tech's reach.
I'm troubled as well. Here's how I see things that played out:

Twitter and Facebook banning Trump? Not surprised, they censor conservatives all the time and have been doing so since even before Trump. Also remember: it wasn't just Trump this time, many others have been banned from Twitter and Facebook, and many groups have been shut down.

Apple banning Parler? I was surprised by this. Apple is supposedly about privacy; clearly, they are not. Changes my view on their value proposition moving forward.

AWS dropping Parler? Again, this is surprising. Head scratching even. Going to make a lot of people think twice about AWS.

I'm also kinda surprised at the HN response to all this, perhaps its because many of the posters here work at one of these Orwellian companies and don't see the issue. I see complaints here about payments companies blocking criminal activity, yet very little relative objections to simply cancelling people the CEOs disagree with.