The Dream of the Internet

April 4, 2023

The Dream of the Internet
{== On screen: Citations available ==}

{== Talk NORMALLY. Consult last bit of Pay to Win 1 for cadence. ==}

The Archive

The Internet Archive and the Wayback Machine are the best tools for anyone who does independent research. They are pretty much the only legal, free source of information an average person has access to if we take the teacher's advice not to cite Wikipedia. The Internet Archive has such a wide range of stuff: trade magazines, game demos, and most importantly books--and all of it is indexed so you can find what you're looking for.

My ongoing series Pay to Win could not exist without Internet Archive. The bizarrely specific and niche stuff they host has been really important for understanding how people in the past actually felt and wrote about games. The Wayback Machine records the entire culture of the early internet, and archived websites and posts are good for accountability. If you look at any website that's cited on Wikipedia, you'll also see a link to archive dot org.

This is kind of an emergency video, because the Internet Archive is in danger. It's not strictly about games, but the Archive is an important source for anyone who writes about, creates, or plays games. This is a video about digital copyrights, and how they run counter to the nature of computers and the internet. {-- You'll be seeing a selection of games I found in the Archive. --}

The Library

To set the stage I need to explain how the Internet Archive sparked a revolution in the way libraries handle digital books.

The Internet Archive hosts a lot of copyrighted books, but they use a clever system to avoid being sued [1]. It's called Controlled Digital Lending: the library buys a physical book, scans it, and then loans it out to one person at a time. It's a logical extension of the way brick-and-mortar libraries work, that both co-operates with copyright law and allows digital access. One book becomes one file. The International Federation of Library Associations--which is an impressive name--came out in support of the system in 2021 [2], and it's become a pretty popular option for smaller libraries.

Publishers are not too keen on it though, here in the great white north the Association of Canadian Publishers warns libraries that CDL is not protected by copyright laws and it is in fact not the library's right to scan books [3]. Publishers would rather lock libraries into their own systems, where they can charge as much money as they want.

A library can buy physical books from anyone and lend them to anyone because of something called the first-sale doctrine [4] [5]; once you buy a physical object you can lend it or give it to whoever you want. Data doesn't work like that. The law recognizes that computers are very good at copying, so when the laws were adapted they knew that the idea of "lending" or "giving away" data was meaningless. Because of this, the only legal way for libraries to get ebooks is to license them from sources the publisher approves of, and licensing is not the same as ownership.

There's a whole constellation of different licensing options for ebooks, but as a rule they are more expensive than physical books and are just worse versions of the one copy, one person rule. The license might have a limited time, or a limited number of loans--HarperCollins pioneered a model where an ebook could only be lent out 26 times before the library would have to buy it again [6]. The marketplace at work. Publishers have also experimented with putting embargos on audio- and e-books, so that somebody who wants to read a book the day it comes out will have to go straight to the publisher.

The people paying for these ebook licenses are largely public libraries, one of the few institutions that pretty much everybody likes. When Controlled Digital Lending came on the scene around 2011, it offered a huge benefit for libraries, saved the public money, and was a return to the way libraries are actually supposed to work.

Preserving knowledge is a major function of libraries, and CDL restores the library's ownership of its books. A scanned, backed up copy of a rare book will practically survive forever. When copyright law was being adapted for the internet, there was no way to actually control digital book lending, but CDL is exactly that, a way of making ebooks behave like physical books. It's also way cheaper, because physical books are a lot cheaper for libraries [4] [5] [6].

A lot of big publisher organizations will tell you that CDL hurts writers [7] [8], but in reality publishers have just gotten away with over-exploiting libraries. Requiring publishers to treat their customers fairly might hurt writers in the short term, but it's absolutely the publisher's fault.

This is all to say that Penguin Random House is not happy with the Internet Archive, whose Open Library project is a prominent example of CDL.

The Case

The trouble with CDL is that, while lawyer Michelle Wu did come up with a legal argument for it [9], based around fair use, it had never been tested in court.

When the pandemic hit, students were stuck at home without access to class materials, and all the libraries were closed. With CDL, if you have one physical copy of a book, then you can loan out one scanned copy, so long as the physical book remains at the library. The Internet Archive launched a National Emergency Library, which waived lending limits on many of the books in their collection [10]. It was a really cool thing to do at a time when there was a lot of uncertainty.

So obviously the major publishers--Hachette, Penguin Random House, Wiley, and HarperCollins--sued them for copyright infringement. And they won.

It's no secret that the Archive has an ambiguous relationship with the law. There are any number of things hosted there that could have gotten them in trouble, but the publishers chose an issue that made them look absolutely evil.

Don't get me wrong, it was naive if the Archive thought they could get away with doing something like this, but it's still a cartoonishly bad look for the publishers. If this was just some squabble over Internet Archive doing something nice I wouldn't be making a video about it, but the ruling effectively makes it illegal for libraries to scan their books. The inciting incident, the Emergency Library, is barely mentioned in the judge's opinion, which is instead devoted to tearing down the Archive's fair use argument for scanning and lending books [11]. If they just wanted to take the Emergency Library down, the publishers could have dropped the case; the Archive shut the program down 15 days after the publishers filed their complaint.

The judge's argument is a solid one, and if the judgment is upheld, CDL doesn't have a leg to stand on. It will be definitively illegal to scan and lend books, and every library will be beholden to the whims of publishers. If they don't want to make an ebook for some old out of print book, too bad. If they refuse to license best sellers to libraries, too bad. I hope you like traveling to the one library that still has a copy.

Not only would this screw regular libraries, but damages are still on the table [11, p. 46], and the Archive is guilty of reproducing 127 books illegally [11, p. 1]. {-- Copyright infringement is supposed to max out at a $30,000 fine, but what constitutes an infringement is not really clear. It could be 30-grand every time they lent out a book. --} <-- not true

This whole situation has a very similar tone to the war on music sharing in the 2000s, where regular people's lives were ruined for sharing a few songs. A lady in Minnesota was ordered to pay 1-and-a-half-million in damages for the crime of distributing 24 songs. Thats more than $62,000 per song [12]. The RIAA, a group of American record labels, used to just sue thousands of people every year [13], some of them children, or dead.

That whole conflict ended with Spotify, where publishers now have more control and can pay artists peanuts. Almost like it wasn't about paying artists fairly in the first place. Ebook licensing is the same, an opportunity to get more control and extract more money. They can be even more tightly controlled than physical books; because you don't have to deal with any pesky ownership laws if everything is a license.

Internet Archive is an enemy of middlemen everywhere and I don't think it's a stretch to say that this case is opening the door to basically suing them into oblivion. The case doesn't directly effect the Wayback Machine or their more straightforward archival work, but the CEO of the Copyright Alliance gave away the game in a piece for Publishers Weekly [14]:

>Even more troubling, there is nothing limiting Internet Archive's scheme to only books. The same ruse could be applied to all types of creative works: music, films, photographs, video games, and more. The result would eviscerate copyright law and upend the legal framework that supports and protects the livelihoods of artists, authors, editors, musicians, and everyone who works in creative industries.

Why do these people even act like they're protecting creative workers? They're profiteers and everybody knows it. If you're going to put on your armour and defend copyright, at least be honest: copyright protects the livelihoods of publishers, who systematically fuck artists in every field.

If you're anything like me, you might have a weird feeling that providing easy access to a massive library of books is actually one of humanity's crowning achievements. The evidence on whether file sharing even cuts into artist profits is pretty mixed [15] [16]. Visual artists make most of their money from commissions, musicians make most of their money from gigs, movies make their money from screenings.

The Computer

I'm not going to make an argument that file sharing, whether it be digital lending or bittorent, is legal. Because right now it's a crime. What I want to give you instead is an argument {== or arguments? ==} for the end of copyright as we know it. So: why do publishers exist?

Before the internet, a publishing company's job was to handle the production, promotion, and distribution of physical books. That takes a lot of money up-front and a lot of industry connections. Authors don't have the capital to get their books in stores, so they go to publishers, who bear the financial risk in exchange for a huge cut of the book's profits. Publishers want to pay authors just enough that they keep writing, but the publisher ultimately has the power to dictate the author's advance and their cut of the royalties. A big name can negotiate a better deal, but publishers definitely take advantage of first-time authors.

The publisher's costs are then recuperated through the sale of books, records, or whatever else. Physical things that need to be created, moved, and displayed. Copyright makes sense for physical goods: if someone is producing bootleg copies of a book, the publisher can't profit from them, and since producing books has a real cost, that could be a problem. Publishers certainly exploit authors, and small-time bootleggers have never been a real threat, but in principle the arrangement makes sense, even if publishing is a scummy business in its own ways.

We don't live in that world anymore. Ever since computers became affordable, publishers have been afraid of digital piracy. Because for all their problems, computers made it cheap and easy to share information. Copying is a fundamental part of computing; running a computer program is the same as copying code from, say, a cartridge into the CPU, line by line.

Once the code gets to the CPU, it executes instructions as the programmer wrote them. Every time you add two numbers, the sum is copied somewhere. If you use the stack, data is being copied to and from it. Even moving data is just copying it and deleting {== probably dereferencing ==} the original.

If you wanted to make a CPU with exactly one instruction, it would have to be none other than copying [17] [18]. You could make a full computer, capable of running any program, whose only ability is to copy a number from one place to another.

Copying is fundamental to the way computers work, and I'm not just making a spiritual gesture toward the nature of things when I say that.

Since the very beginning, software has been something freely exchanged and remixed for different machines and different purposes. It's no accident that the concept of open source came from the software world. And since the beginning profiteers have been afraid of it. Bill Gates, who was a billionaire in spirit even when he was broke, wrote a combative letter to computer hobbyists in 1976 because they were sharing a programming language that Micro-Soft made for the Altair 8800 computer [19]. Those details don't really matter... and actually Micro-Soft didn't design the BASIC programming language, they ported it to the Altair without paying the people who invented it.

The source is a dead link, but Wikipedia says Altair BASIC was at minimum $150, $755 in today's money, and although computers were also very expensive, I seriously doubt most hobbyists could even afford the program [20].

Let's ignore the morality here for a second: what Gates and others were rubbing up against was a revolution. Computers, by their very nature, undermine the idea of copyright. People may not have understood this, because programs were being sold on physical cassettes or disks, in physical magazines. It sucks that people who put up the time and money to produce those goods were getting their software pirated, but their whole existence was a contradiction. Scarce goods don't make sense in the world of computers.

As computers became even more popular, we started to see ridiculous Digital Rights Management or DRM schemes. Some games would have you consult a physical codebook for a password in the game, while registration keys were a popular option for software. These are both attempts to hold on to the old world of copyright, to remove your ability to copy the data.

The internet made the contradiction even more intense: by the 2000s you could download programs online, so distributing software became practically free. Books, music, and movies were being digitized too, and it really seemed like the old ways of doing business were doomed. Copyright was not only archaic and antithetical to the internet, but it was practically impossible to enforce.

The technology would naturally lead to art and knowledge proliferating, freely, everywhere.

The Dream of the Internet

I mean, it just had to. The internet offered no resistance to the flow and exchange of anything anybody wanted to share. It was mythologized with phrases like "information superhighway" because people knew that it was really going to change everything.

The early web you find so many people pining for embodies this spirit. People were making websites about topics they enjoyed, or just about themselves and their friends. These were made possible by the fact that you can just copy any website, and you can save cool GIFs. Computers are very good at stealing, and stealing is the first step to making something new.

The internet definitely posed the question of how people would get paid for their work, but that question has answers. Everybody can produce and publish whatever they want now, and you can support them directly. Sure the work is open to anybody, but so is the donation link.

That system isn't perfect and it can be difficult to make ends meet but niche projects are able to find a lot more success than they ever did in the past. Maybe your lo-fi furry indie rock record will never get pressed on vinyl, {== Twin Fantasy ==} but if you practice and work hard you'll make a few bucks. Writers' and artists' lives have always been sort of precarious, and they don't need publishers censoring their work and taking 90% of the money. They can just work directly for their audiences, and make similar money releasing stuff for free.

There would be a new cast of middlemen and parasites skimming fees off the top, but the lumbering, centralized world of publishing had no place online.

For audiences, this new world promises access to a limitless well of cool stuff. You can get any book, album, or game you want and support the people who made it directly with money or popularity. It's a system with pros and cons. An independent writer can't get an advance, and authors have to take all the risk of making and distributing their work onto themselves. But the old system had its problems too, especially for anybody working outside the main stream.

Most people can't afford to buy a whole bunch of new albums or books every year, and while exposure doesn't put food on the table it's a lot easier to find an audience now than it was before the internet. The obvious argument against this model is some variation of "human nature is bad, everyone would steal," but reality proves that wrong. People are paying millions of dollars a month for podcasts, videos, and games that they could easily pirate.

People do steal artists' work and pass it off as their own, but this could easily be fixed by any platform that cares enough to do something about it, and communities have already done a really good job self-policing when it comes to tracing, bots, and art-theft.

For me, easy access to books, especially out of print books, has been a game changer. If there's something I need, there's a good chance it's on the Internet Archive. I can get all these books that I could never afford to actually buy, and use them to put together videos like this, which I hope are helpful or at least informative.

Even with its problems, this dream of the internet is one I believe in. It's a little slice of the world where people give what they can, and where getting a book for free doesn't actually hurt the author, because copying a file is free. The internet gives us a digital commons, a canon that we can freely interact with and build on. On the whole, it can only increase the pace of innovation and the quality of work, even if there are a lot of dead-eyed algorithm-chasers out there. There always were.

The Internet Archive is just the beginning of what we could have online, an archive of sites and works that anybody can use. While places like sci-hub, Anna's Archive, and libgen are allegedly really awesome, the Internet Archive is uniquely allowed to do its work legally, out in the open. It offers a sort of compromise, or a transitional state between the dead world of copyright and the future.

The Reality of the Internet

If the internet existed in a vacuum, that might be the end of it. But here's a very good test for whether some idea is just a fictional utopia: does it require everyone to peacefully give up their power? If so, you're doomed.

Publishers and their advocacy groups are not going to just throw in the towel and quit. Institutions, be they Penguin Random House, the local library, or the World Intellectual Property Organization, are always recreating themselves, always fighting to make their vision of the world real. The war between pirates and rights-holders has been raging since the dawn of the internet, through the recording industry suing random people, laws like the Digital Millennium Copyright Act, and through more and more intense DRM.

DRM had to get more intrusive over time, because stopping a computer from copying something is hard to do. In the world of games, Denuvo causes problems with performance [21] [22] and probably reduces the life of your SSDs. Websites use tons of tricks and systems to restrict access and stop you from saving files, at the expense of making every website into a slow, bloated "web app." Bitcoin is the logical extreme, it guarantees that every link in the chain is unique and in the correct order. And it wastes a small country worth of energy to do so, because it is a complete contradiction with the way computers work.

But DRM's final form is what book publishers have been trying to push onto libraries, and you might recognize it as the Spotify model. Music is licensed to Spotify en masse; they collect revenue from ads, the sale of personal data, and subscriptions and pay out a fraction of that money to artists. The major record labels all own shares of Spotify [23], they make the playlists that the site constantly shoves in your face [24, p. 14-15], and they get sweet deals because they're powerful [24, p. 13]. Musicians, who publishers definitely care about, got less than 6% of the 31 billion dollars the music industry made in 2022 [25]. I don't know how radical you are, but I'd say they did at least 50% of the work.

Spotify can't even be called DRM because you don't have any rights to manage. You are renting your music now, and paying for it by listening to ads or subscribing, and this generates money for a bunch of blood sucking record labels. Remember, distributing files has basically no cost and the only reason publishers have a foothold online is because they sued options like Napster into oblivion.

If you care about musicians, there is practically no difference between pirating a song and listening to it on Spotify. In fact, when Spotify started it was actually just piracy, the service was peer to peer and the company didn't own any licenses [23, p. 12].

OverDrive is a company that negotiates with book publishers and sells ebooks to libraries. There are other firms doing similar things, but OverDrive is the big one. They are not quite the same as Spotify since their main customers are libraries, but both services are a movement away from ownership and toward rentals. While these ebooks do offer some extra revenue for authors, OverDrive's main job is to help publishers hold libraries hostage to whatever stupid, exploitative licenses they come up with. And obviously the ebook apps spy on you too, just like Spotify [25].

The dream of the internet is ownership in common, the creation of a canon that we can all access and use. What we're getting is a huge asterisk after that dream. We get access to services, where we can consume a large but incomplete catalog of things in exchange for personal data and ad views. The platforms don't support remixing in any interesting ways, and usually make remixing or even notetaking more difficult.

OverDrive might make authors a little extra money for now, but it's part of an ecosystem that hurts everybody. Amazon's Kindle Unlimited subscription is a sort of Spotify for books that pays authors by how many pages people read, adding up to two or three dollars if somebody reads your whole book [27] [28]. Amazon feeds on authors by underpaying them, OverDrive feeds on libraries by overcharging them.

This shit is bad for artists, it's bad for readers, and it means the government or some corporation can take inconvenient books off your Kindle. The endpoint is to make all of the alternatives, like Controlled Digital Lending, illegal. Internet Archive is already the only semi-legal source for free books. Laws like the DMCA, SOPA, and PIPA were early attempts to lock the internet down, and the upcoming RESTRICT Act will allow the American government to ban or censor any site that it deems a threat to national security [29]. They want you to be scared of TikTok so they can pass it, but the reality is no more sci-hub.

Don't get me wrong, I'm not a fan of groups like the Electronic Frontier Foundation either; their founders wanted the internet to be censored by private corporations instead of the government [30], and they used a lot of mythical bullshit like the phrase "cyberspace" to distract us while advertising companies and silicon valley destroyed the very dream they peddled. It's called the Electronic Frontier Foundation because they thought the internet was a frontier for them to conquer, without the government getting a piece of the pie. The slavery implications never seem to come up when they talk about it.

For the time being, supporting the EFF is a good strategy. They ran successful campaigns against SOPA and PIPA, and they are representing the Internet Archive in the ongoing case along with a law firm. They're going to appeal the judge's decision on Controlled Digital Lending. For now, all you can do is sign this petition, try to raise awareness, and donate to the Archive if you can. I have no idea if politicians or judges actually listen when people talk, but maybe if we make it clear that crippling libraries is unacceptable the Archive's lawyers will be able to get something done.

The Future

What I've been implying this whole time is that it's not enough to make a sound legal argument for CDL. CDL shouldn't even be necessary, it's a workaround because copyright is an obsolete concept, just a way to keep the publishing vampires from sucking all the blood out of libraries. CDL is consistent with the way physical books are treated, but the point of copyright law is to serve copyright owners, so that doesn't actually matter.

CDL is an existential threat to digital copyright, because if it becomes legal to lend out digital books, why not recorded music or games? This is why the publishers are trying to make it illegal for a library to even scan its books, they want to make it easier to dismantle the Internet Archive. I think it would be difficult to patch CDL into the existing laws, so if the Archive's campaign gets enough traction it could mean that the law has to be changed substantially, or at least some new and useful precedent will be set.

If you want things to get better in an enduring way, adding new footnotes to outdated laws is not going to do it. It doesn't cost anything to want a better world, and you don't have to water down or focus group your dreams for a bunch of CEOs who hate your guts. The internet can be better, but just dreaming about it is not enough. You need to actually fight for what you want.

We made art long before copyright existed, and we can make art after it. Computers are made for remixing, and popular art offers us a common language to rearrange, reinterpret, and build on. This was true even before computers, and back then, like now, the people benefiting from copyrights were publishers. Copyright doesn't represent your interests as an artist or a consumer, and it never has.

In the meantime, acquire everything legally, support publishers, and please remove the gun from my head Mister Warner.


An American Tragedy: E-Books, Licenses, and the End of Public Lending Libraries?, Matthew Chiarizio. ARCHIVED. Not a great article, views books as subject to market forces; great books will be written with or without "incentives," open access only punishes mediocre profiteers. The Odyssey, for example, did not require copyright law to come into being. But the article is a nice overview.
M. M. Wu, "Building a Collaborative Digital Collection: A Necessary Evolution in Libraries," LAW LIBRARY JOURNAL, vol. 103, [Online]. Available:
S. B. Karunaratne, "The Case against Combating BitTorrent Piracy through Mass John Doe Copyright Infringement Lawsuits," Michigan Law Review, vol. 111, [Online]. Available: ARCHIVED.
F. Oberholzerā€Gee and K. Strumpf, "File Sharing and Copyright," Innovation Policy and the Economy, vol. 10, pp. 19-55, Jan. 2010, doi: 10.1086/605852. Available: ARCHIVED.
Defence Science and Technology Organisation and O. Mazonka, "Bit Copying: The Ultimate Computational Simplicity," ComplexSystems, vol. 19, no. 3, pp. 263-286, Oct. 2010, doi: 10.25088/ComplexSystems.19.3.263. Available: ARCHIVED.
Farhad Mavaddat and Behrooz Parhami, "URISC: The Ultimate Reduced Instruction Set Computer." 1988. [Online]. Available: ARCHIVED.
Bill Gates, "An Open Letter to Hobbyists," Feb. 03, 1976. [Online]. Available: ARCHIVED
M. Eriksson, R. Fleischer, A. Johansson, P. Snickars, and P. Vonderau, Spotify Teardown: Inside the Black Box of Streaming Music. The MIT Press, 2019. doi: 10.7551/mitpress/10932.001.0001.
J. P. Barlow, "A Declaration of the Independence of Cyberspace," Electronic Frontier Foundation, Jan. 20, 2016. (accessed Oct. 17, 2022). ARCHIVED.