Politics broke the internet, and it can still break the blockchain
Sometimes history really does repeat itself
A number of posts on my Substack deal with the economic structure and social consequences of the centralized, hub-and-spoke design of web2 — the ways that money flows around in it, the behaviors it incentivizes, and the perverse effects of its major platforms on our politics. But this is all a hot topic, right? Everybody has opinions on what’s wrong with the internet and how to fix it.
The angle that’s woefully under-discussed, though, is how web2 got to be the way it is. This post doesn’t aim to be anything more than a first draft of a history of the rise of what I’ve called the adversarial web. This is just a place for me to put some initial thoughts, which I’ll elaborate on in interviews and subsequent articles over the course of this project.
If you’d like to help me out, please pitch in with comments and suggested bibliography. And, as always, shares and subscriptions are appreciated.
Contrary to popular belief — even in technical circles where people should know better — the internet architecture we’re currently struggling with is not the inevitable result of the blind march of technological progress. The internet of mega-platforms, mass surveillance, data leaks, cancel mobs, deplatforming, and intrusive advertising did not find its present shape purely through “innovation” or market forces. And, of course, it’s also not quite the decentralized, nuclear-attack-resistant network envisioned by DARPA and the academics who pioneered it.
No, the hidden architects of today’s internet are lawyers and lobbyists. It’s weird how many of us have forgotten the lobbying, lawsuits, and post-9/11 security state fearmongering it took to create today’s online experience. Even many tech types my age don’t seem to remember the story of how an alliance of major intellectual property holders, telecoms, and governments were so (justifiably) threatened by the decentralized, peer-to-peer nature of the internet that they went to war against it and won.
But I remember this story, because not only did I watch it happen, but I covered it from the trenches of the DRM and P2P wars. I co-founded Ars Technica with five other folks in the summer of 1998, and by about 2003 a big part of my reporting was focused on the intellectual property (IP) wars that shaped the architecture of the web we’ve all come to love and hate.
This is why, when I talk about the difficulties of digital sovereignty, or the fight to keep crypto legal, I keep circling back to just how much of the present way the web works is downstream of past political fights we largely lost to deep-pocketed, well-connected incumbents.
I do have some hope that crypto can avoid the fate that befell a previous era’s attempt at a peer-to-peer repristination of the internet. But every time I see some whippersnapper of a young programmer suggest that some bit is inherently unstoppable by the powers that be, I worry.
As Punk 6529 says, we do have to win politically. And we have to win politically because network topology is always and everywhere dictated by political realities. This has been true of the internet, and it will be true of the blockchain.
A short history of the P2P wars
There was a brief moment from about 1999 to about 2008 — a little under a decade — when the web-centric, client-server-based internet seemed to be headed back to its decentralized, free-wheeling roots by way of the peer-to-peer (P2P) revolution.
To know what that was like back in the day, close your eyes and imagine a software experience with the following qualities:
You download an app, install it, launch it, and point it at a directory full of files you want to make available anonymously to the world.
You don’t necessarily need to create an account anywhere new (beyond whatever local account you’re already using on your computer) in order to serve up your files to the world or to download remote files from complete strangers.
The app is both a client and a server, in the same package. This means you can sort, filter, and search, and generally find all sorts of media, and software to download, and you can serve up the same to others.
Even though this app functions as a server, you don’t need any sysadmin skills to run it. No editing config files, or setting up users and roles. It’s literally just an app that you launch and use.
You can quit the app at any time, and start it back up at any time, and quit it again, and restart it, etc. There’s no real expectation of “server uptime.”
If you want a more centralized, curated experience, you can add any amount that you like. For instance, you can use a version of this software that requires an account, or you can stay totally anonymous. You can subscribe to a centralized feed of TV shows so that whenever a new show airs it automatically shows up in your TV’s media software.
We actually had all this, once upon a time. BitTorrent, Napster, Limewire, and similar peer-to-peer apps were easy-to-use, and for many years the pirate peer-to-peer media ecosystem actually provided a better media experience than anything you could get on the white market.
There really was a time, up through at least 2011, when broadcatching was actually the best way to stay current on TV shows. There were no good white-market alternatives that could push high-definition, ad-free shows from all over the world to your media center software immediately after they aired on whatever network or channel.
During these years, I remember regularly thinking of the combination of BitTorrent, TVtorrents.com, and Plex: “I’d love to pay for this if only Hollywood would offer a service that’s this good.”
Of course, the reason the above combination of decentralized apps and services worked so well for TV was that this was the one media niche that had managed to avoid the litigation that had tamped down much of the peer-to-peer traffic in movies and music just a few years earlier. I covered some of these suits at Ars:
IFPI launches largest legal action ever against P2P users. Will they drive us all to darknets?
P2P v. Big Content: Opening arguments in the Supreme Court case to watch
There’s a lot more on this page and further back in the archives. Just one example after another of lawsuits or of ISPs caving to copyright holder demands.
So it was that eventually, the emerging peer-to-peer ecosystem was deliberately crushed — sued and throttled back into obscurity. It never got the chance to evolve into whatever it might have become.
All this suing and legislating was done in the name of “protecting artists and creators,” but really it just protected middlemen, incumbents, and rent-seekers. As I put it in this 2005 article, people at the time were thinking hard about ways to ensure that creators got directly compensated in the emerging P2P ecosystem:
This editorial goes off the rails in so many places, but I'll content myself with pointing out only one fallacy that underlies the NYT's entire argument. The NYT's editors assume that peer-to-peer file sharing is by necessity a content distribution system in which content creators cannot and will not be payed. It appears that the NYT thinks there's something inherent in peer-to-peer as a distribution mechanism that magically dictates that creators can never be compensated for any work distributed over p2p networks. This is, of course, complete and total baloney.
There have been any number of fantastic proposals floated for building artist compensation into p2p networks. One of the earliest and sanest was Napster's proposed settlement with the RIAA, and one of the most recent and cleverest is the Weed distribution system. The problem with all of these proposals, as I argued in an editorial on the original Napster proposal, is that none of them involve the RIAA getting paid because under such schemes the distribution channels that the RIAA controls and profits from are eliminated.
We never pursued any of these possibilities, though. We weren’t allowed to.
My point: An entire alternate branch of the evolution of consumer internet hardware, software, and services was simply wiped out, like a meteor wiping out the mammals instead of the dinosaurs. Again, this didn’t happen because the experience of running your own file server sucked or required specialized expertise — it absolutely did not suck, and in fact, the nascent P2P ecosystem was quite painless from a user admin point-of-view. Rather, it happened because without a centralized authority that the RIAA could sue or the feds could threaten with prison, people would have just shared whatever the heck they wanted with whoever the heck they wanted, and we just couldn’t have that.
It wasn’t just Big Content, though, that greatly preferred an internet centralized into as few hands as possible. There were at least two other forces at work that I’ll only briefly mention, below:
The post-9/11 security state was deeply worried about people sharing state secrets, or jihadi inspirational materials, or bomb-making instructions, or infrastructure vulnerabilities, etc. The Wild West of P2P was a nightmare for them; they insisted on centralized points of control for information flows.
The telecom industry was incredibly threatened by people using the internet for voice and video communication, thereby taking a chunk out of their core business. They wanted the ability to do traffic shaping and to dictate what consumers could and could not use their networks for, and they also wanted to charge the large platforms for access to their customers (in violation of net neutrality rules). They certainly didn’t want customers running file servers.
With the government, the telecom industry, and every last corner of the content industry (not just Hollywood and music, but mass-market and academic publishing) fighting against unregulated P2P, the heavily centralized internet is where we all ended up, whether we wanted to be here or not.
Can good legislation undo what bad legislation did?
The centralized, ad-supported internet is why we have mass surveillance, de-platforming, and the moderation wars. These are all problems that various legislative bodies have tried to solve by passing new laws, but the solutions are all destined to backfire because they’re treating the symptoms, not the disease.
You can’t legislate privacy and consumer rights back into the current architecture of the internet, as the Europeans have tried to do with GDPR. The end result of these efforts is usually worse than if politicians had just left things alone because the only ones who can comply with the onerous privacy laws are deep-pocketed incumbents. But a takedown of insane EU anti-tech regs is a topic for another day.
American fantasies of implementing similar protections are likewise doomed. These ideas are all going nowhere because they’re trying to fix the problems at the wrong level.
The real solution is to unwind the previous two decades’ worth of extreme centralization by dismantling entities and incentives that created the current internet.
Another way to put my point is that today’s widely reviled social media and e-commerce monopolies are downstream of pre-existing IP and telecom monopolies. If you busted up the IP and telecom monopolies and removed all regulatory barriers to completely unrestrained, uncensored consumer peer-to-peer communication on the internet, the platform monopolies that are downstream of the IP and telecom monopolies would eventually fall, as well.
There are things legislation could do in this regard, if only we had the political will.
Legislation could bring back the Founder’s Copyright and put the IP monopolies out of business.
Legislation could empower municipal broadband efforts and bring true broadband connectivity to rural America.
Legislation could enforce network neutrality (ducks as my libertarian readers throw things at me!).
Legislation could make Sci-Hub legal and make open-access scientific publishing the norm.
Legislation could force Treasury to obey the law and forbid it from sanctioning unowned, uncontrolled (and uncontrollable) smart contracts.
Legislation could actively encourage crypto-based peer-to-peer payment systems and even debt financing that would compete with (and ultimately replace) current big banks and payment cartels.
Instead of any of the above, the hot idea is to tell Facebook which startups it can and can’t buy, or to make laws about accepting cookies, and other marginal efforts that will not make anything better and could make a lot of things worse.
A newly decentralized internet could really work, though. If only we had the will to make it happen. Think about it: Everyone in 2022 has an astounding, stupendous amount of networked computing power and storage at their literal fingertips, much of it either unused or occupied mainly with browser bloat. In a world with radically different laws than the ones we currently have, we could easily mesh these networked resources together into a censorship-proof, surveillance-free, sovereignty-enhancing network of networks, with micropayments flowing all over the place and everyone in control of their own data.
Anyway, we’re not going to fix any of this stuff because we can’t even pass basic laws about obvious stuff, anymore. So I type all this out in the hopes it will be seen by some future, post-collapse civilization that’s trying to rebuild without repeating all of our mistakes.
Probably the best outcome I can hope for from telling this story is that web3 folks will read it and realize that incumbents and politicians can and will tame any amount of new hotness, and either neuter it entirely or twist it to their own ends.
The bottom line: It really, truly is never enough to just write code. You have to also write laws — like, old-school laws that they pass in capitols, not just DAO rules or whatever. If we don’t take politics seriously, then the present peer-to-peer revolution will end up in the same obscure and mostly forgotten place the last one did.
I had a front-row (well, maybe second-row) seat for many of these development of the Internet and the Web. Let me just say in two words: It's hard.
My touchstone on net policy is Lawrence Lessig, and several of his early books cover exactly this territory. But he's hardly the only one. There have been many papers and books by various people going over this topic overall, often from the perspective that techno-Libertarianism is nonsense. Alas, my final conclusion, after more than two decades, was that this really is downstream from politics in general, and there isn't much that an individual can do - unless they want to be a full-time political player for their life, and that's a specific career. Again, touchstone, Lessig has written about this extensively in his later books, about general politics (i.e. huge amounts of money!) driving the choices of technological development.
In a way, it's a meta-problem. What's the audience for telling people something they don't want to hear? How is that supported and compensated? And translated into action?
A wonderfully succinct history.
If we are to follow the centralized CCP vs Decentralized West trajectory, do you think a natural outcome of that might be a reversion towards local servers and peer to peer hosting, as a trend away from the Spotifys, Netflixs, and YouTubes of the world? One example in hardware that is very present, is that Apple will be producing a good percentage of their phones in India, outside of China, as of this year.
I can imagine (hope) that ongoing emerging conflicts will do a similar thing to the software-IoT side of things, as there's more port wars and geo-restrictions. But to your point, that inertia is still largely at the whim of global geopolitics and outdated/unhelpful internet laws