The User Generated Content myth that dates back to the early days of the Internet and must die

A friend was into one of those popular self-help books a long time ago and she ended up going to a meetup for the book and I guess one of the stories in the self-help book somehow involved a couple who got rich because the community around them helped out.

Ergo, one of the couples at the meetup asked if you they all could help out? They promise to help the rest of you out next, once they are rich and all.

It sounded like there was a whole lot more to the book than that, but this maps to reality. There was this whole buzzword and thought process around User Generated Content, UGC for short. A bunch of companies, some of them kicking this off before anyone had even bothered to invent the term, have raced to take advantage of the UGC marketplace because the idea of making money off of something you can get for free is so damn appealing. Except that, lately, the whole lot of them have been doing wacky things as they realize they can’t subsist on money-losing repeated VC investments and are actually failing to make money off of your UGC.

In order to explain what I think happened, well, let’s imagine Spez getting mobbed by a mob of angry redditors, the frame freezes, you hear the record scratching sound everybody always uses, and spez says “Well, you probably wonder how I got here…”

Let’s go back to the days before the modern version of the Internet. There was a version of the Internet that was explicitly non-commercial things, where even the Digital Equipment Corporation, one of the first computer manufacturers to connect to the internet, having a support forum for their computers was a bit of a stretch for some people. There were some very large companies with per-minute billing and exclusive amazing things available that grew out of the Teletext services. And there were eventually BBSes and a bunch of the BBSes had networks associated with them like FidoNet and those networks were, like the Internet, explicitly non-commercial. And you pretty much needed to be part of the rarefied technological priesthood in the days before the 80s, but in the period between people having modems at home and the Internet being huge, computers became accessible to some people who were just… normal people.

We’ll start with the non-electronic real world.

In the non-electronic real world, it was totally reasonable to swap all sorts of information and goods, but the cost was actually really really high compared to how we’d look at it today. I’m going to talk about coins of value because the actual value changed over time but at least that should help you envision things without thinking about what was twenty five cents plus or minus inflation and everything.

You could make a fanzine to swap fanfics or a newsletter to share rants but at the most economical version of this, you were going to end up with a few pages worth of information that could fit in a first class envelope and get sent for the smallest stamp value that was a few coins worth of value. And if you wanted to copy it, you were also looking at at least a coin per page or so, depending on how you managed to score things. Obviously some people were able to take advantage of the work’s copy machine or mimeo machine, etc. But, regardless, even a minor little thing spread between a small group of people in a distributed fashion… that’s a handful of money just for infrastructure and it’s not that much data.

There was, of course, the professional publishing industry with magazines and books and media and stuff. But just on the lowest-end of people mailing mimeographed fanzines and stuff was enough to create whole distributed real-world movements. We talk about fanfic and slashfic now, and there’s AO3 to host it, but slashfic was created through the paper medium of mimeographed fanzines. Slashfic from the days of yore is actually very important, culturally speaking, because it was from that audience of Trek fans that modern Trek was created.

Even in the early-ish days of modems, you had spent some money on the hardware. Although that cost went way down quickly, even when things were fairly pricey, you could transfer a lot more data, especially the written word, for a lot less than ever before, down to probably a few coins of currency for megabytes worth of data, in aggregate amortized form. And, obviously, the comparison is not really fair, but if you were trying to run your Comcast connection to the limit and then break that down into how big of a Kirk/Spock archive you could collect with a handful of coins, plus or minus inflation, etc. you realize that one thing the Internet has done is cut dramatically the cost to transfer that juicy scene where Kirk finally owns up to Spock, to basically nothing. The amateurish artwork? Also basically free. Your epic mash-up of a Nine Inch Nails song and a bunch of Kirk and Spock scenes? Well, actually, the data and transfer cost might be something that would start to add up, but it’s peanuts compared to what you’d pay if Nine Inch Nails or Paramount decided it wasn’t fair use. The important thing, however, is that in the span of a few decades, it became dramatically cheaper to transfer all kinds of information and effectively free to transfer words.

When you have a new medium that changes things, there’s always an opportunity to ask questions and try things out and shape how that new medium ought to work. This has gone in all kinds of different directions. Allegedly, the earliest radio news was basically people reading the newspaper at you and it took a while for people to figure out what radio news ought to look like, only to get disrupted again with TV where you had a person… and then “Films at Eleven” is still a cliche phrase today but it’s been a very long time since some newsperson had to run out with a 16mm film camera, shoot the event, then run it back to the TV studio where a team of people would develop it as fast as possible, which is hard because these are chemical processes that must be run at a specific speed and at a specific temperature and literally anything you do to shortcut them ruins them, probably running it through the machines still wet from the final wash, so of course the earlier news broadcast would tell what happened and then the films would come out at the late news.

Ergo, there was this period for the Internet as well. And it’s recent enough that a lot of folks who were alive then are still alive now, and it’s important to remember that these were decisions that were made based on what people knew at the time and maybe even that we’re not actually stuck with them, although some of the implications still hold.

For example, there was this idea then of a micro-penny-meter. It’s still probably a great idea. Basically, you’d run a little app in your browser and it would keep track of how much Internet you were using and through some fair and reasonable process, would divvy up micro-pennies for all of the little sites you used. Maybe people could denote premium content that got a higher rate. Or maybe it would say “Okay, now you gotta pay real pennies”. The important part to this idea was that you’d put maybe a few small bills worth of value into the machine every month, tops. And some people would make some serious money, whereas a lot of people wouldn’t even make enough for hosting.

And, unfortunately, there’s not necessarily a way to make this actually work, in the real world, with fraud and attack surfaces and insurance and everything. There’s way too many ways to attack this and siphon micro-pennies and it turns out that, while we were able to make improvements on small credit card transactions between the start of the Internet and now, there’s still a pretty hard lower bound. But keep the idea of a micro-penny-meter in the back of your head, because I’ll get to it later as a metaphorical device.

There were other ideas also proposed, because it was clear that at least some money was required to do things. I guess the other popular idea was some version of subscriptions and walled gardens, except for everything. But, especially at the time, everybody was totally fine to take the advertisements as the solution, which turned out to be kind of a problem later on.

One of the things that makes us human is communication. Deep communication, like stories and myths and imagination and even how-to guides. Obviously there’s always been the fully-formed well-educated version of this, like books and movies. And, while we’ve completely screwed things up for the actual professional creators as well, I’m going to just leave that alone because that’s pretty much an entire separate subject and focus on the rest of things: Newsletters, letters, conversations, updates, stories, slashfics, half-assed guides about how to do something not quite the right way… and so on. The long and sometimes fluffy tail of content.

It is this level of communication that was dramatically better able to be served by the nascent Internet, by pushing the marginal costs down and potential audience up. And yet I would propose that this level of dialogue has always been a critical part of humanity and our communication. Either way, once it became possible, it became addictive.

The first generation of “making money on the Internet” was basically outright scams. For example, the two lawyers who spammed a bunch of people, then wrote a book about it.

But, at the same time, once details of secure payment started to get worked out, there were a bunch of things that were frankly super-obvious, that being turning both regular old catalog merchants as well as the silly specialized merchants you’d find the hard way in things like Computer Shopper into something much easier to access, even if ordering your groceries online turns out to require a carefully constructed labor abuse scheme.

But returning to the thread of the early internet, however, there was already a huge amount of UGC on the Internet, it was just generally fully isolated from the commercial version of things. Usenet, for example, was all UGC and no advertising and for a while the Usenet feed was part of your Internet bill.

But the next thread of conversation rapidly became UGC or, ”how can we take those things that people do all of the time that have historically had hugely marginal value that are now extremely cheap to transmit, and turn that into our own money“.

I should digress about value. Understanding that, in a world where the micro-penny-meter were to be feasible and copyright extensions hadn’t occurred such that Star Trek was now part of our cultural mythology in the same way that Shakespeare is, there’s an admittedly impossible parallel world where we can fiscally appreciate the value of a well-written slashfic. Although right now, you can still kinda profit when you come up with a new way to handle the pairing is to file off the serial numbers and explain that it’s some other Space Ship Captain who happens to look a lot like Kirk and another highly logical Science Office who happens to look a lot like Spock, except maybe not with pointy ears because that’s too obvious.

To be clear, given that a Twilight fanfic with the serial numbers filed off became Fifty Shades of Grey, it’s not just silly idle things. The success of Fifty Shades, while a little… something… is a good very profitable example of how just because something is written within the context of not being able to make money at present, like being fanfic, that does not mean that it shouldn’t be earning money on a micro-penny-meter, platonically speaking. Literally everything that people are willing to spew on the internet that you are willing to read has some sense of value, even if it’s marginal.

Looking at it another way, I’m not a credentialed or degreed electrical engineer and I’ve never worked in the field. I’m totally happy to give advice, but I also recognize that I’m neither a substitute for an actual education nor am I any sort of actual expert, so I get that when I write about electronics, it’s much lower value than something I might have a degree in and work professionally at. It’s still fun to talk people through MOSFETs and LED drivers because it’s fun and maybe my approach will make more sense to someone who just wants their damn cosplay to light up.

The important thing I want you to really see here is that UGC does have value, real value, and it is only because a person ignores the value out of a sprit of fun or sharing or because we’ve forbidden them to actually make money off of it or just because they like to run their mouth that it is free. After all, if I say “Penny for your thoughts,” I don’t reach for my wallet.

This conflicts with the “Information Wants to be Free” Internet meme, mind you. If you want information to be truly free, there needs to be an accompanying model where a person who is comfortable giving out not just their marginal content for free but their core professional energies for free, I think you need to make sure they don’t starve first. However, another subject for another time.

Historically, the first version UGC was Amazon. See, you could buy books on Amazon, but you couldn’t browse them in the same way you could in the store to flip through and find out if the romance was squick or if the ending sucked. So reviews were added. It turns out that, again, because of this innate urge to communicate, people are actually pretty happy to write reviews, even if they aren’t at the same level that you’d see from a book professional.

But Amazon was one of the first real actual success stories of the Internet and everybody wanted to go for that level of gold and so it was picked apart. This meant that a lot of early experiments happened with the earlier websites around the idea of building a “community”. If you go to a site that’s been around forever and not updated, you might spot the “Guestbook” or the “Forum” for example. And, unfortunately, the community part is actually hard. Even then, people rapidly discovered the need for moderation when their Guestbook had a bunch of griefers and spam links and the forum became useless.

If you go back to the book sales from that era, there’s a bunch of books about how to build your internet brand by building a community. Thus, lot of people, most of them more internet-savy than the average business person because they were making an Internet play in the days of Web 1.0, made serious efforts to build a community to sell products on their site, thus even before folks were saying “User Generated Content” it was the same idea: taking a person’s spare-time energy and turning it into money for a business person, without the person getting any of it.

Eventually, the first web boom died and a bunch of people, largely through luck, ended up actually quite wealthy and a bunch of other people ended up broke. The second web boom, Web 2.0, was arguably something the O’Rilley and Associates constructed as a way to drum up business that was otherwise in decline, but this was the actual “community” boom. Sites like flickr, tumblr, reddit, and many others were created.

At some point, Stewart Butterfield, founder of Flickr, accidentally got caught saying ”Every time I hear the phrase ‘User Generated Content’, I die a little inside.” Which roundly got criticized and he had to dance around what he meant for a while… but yeah, I guess I’m coming back to that phrase lately.

See, there’s an informal pact, but it’s not like anybody sat down to write down the “User Generated Content Generators Bill Of Rights” or anything. So it’s got a few relatively easy points to grasp. For example, were providing you with free hosting because, frankly, completely non-technical people self-hosting their own content is not happening. And then they could put advertisements up or do other things, maybe even pass the hat for the hosting bill, in order to survive as a business.

Obviously because we’re still really thinking about content where a lot of people’s stuff is micro-pennies on the micro-penny-meter, if there is passing of the hat it’s done in such a way to re-create most of the aspects of the micro-penny-meter in quantities and timings and shapes and forms that are actually achievable with the financial system. Albeit, not actually that fairly.

There’s also a bit of an understanding on both sides that nobody is going to try and force too much verification but that people are going to be roughly who they say they are. You can’t require people to verify their ID because that’s a fairly direct avenue to direct abuse, plus it’s seldom things that users want to reveal. Except that this identity thing has been a bit of a disaster, really, because it’s far too easy to create yet another identity and it turns out that a great new way to perpetrate The Big Lie is to create a zillion sock puppet identities all spewing similar truth-y sounding stories. So, while we started fairly early on with spammers and people who wanted to hawk a product, now we’re talking about adversaries that are content to spew all kinds of mistruths in bulk to move the needle on an election a few percent.

Part of what killed a lot of the pre-Web 2.0 efforts was that it was a business person trying to create product and provide a space for user generated content, which created a lot of opportunities for sleazy abuses, which I guess turns “roughly who they say they are” into just ”Authenticity

It’s also actually hard in ways that folks didn’t appreciate earlier, because if you compare mimeographed fanzines to modern sites, the biggest change ends up being that you have an incredible amount of UGC because it’s so much easier to round up a bunch of people who care about the exact same ship or the exact same photographic subject or style or anything else in common. You aren’t fighting locality and the cost of the actual data involved is remarkably low. On a good day, the UGC site is a victim of its own success.

So, while I was talking earlier about the marginal data costs, it’s actually not nearly as marginal as I was making it out to be, although I would note that there was an explicit pivot to video in social media that involved companies intentionally screwing with the metrics they reported to wall street and if we hadn’t pivoted as hard to video, this would be a much smaller problem. Or maybe not. The Internet is made at cats and it’s hard to disagree that there’s a whole lot of cute kitten behaviors that don’t quite work as words or pictures.

And I’d also point out that moderation has always been a part of this, except that the moderation for a fanzine or BBS or a small site is very different than a large site that requires you to have people who are going to have to sort through deeply disturbing content all day. Except that on a lot of sites, such as Reddit, there are also volunteer moderators who aren’t paid either, even if they tend to be isolated from the disturbing stuff.

I’ve kinda been pointing out how these things that were informal unstated social agreements have fallen apart from their own weight, especially with the very large category-killer sorts of user generated content sites like Reddit. However, in the past year or so, money isn’t so much free, which means that all of those sites need to make money, and I think this is really where the wheels are falling off the car.

To make money, you basically have two options, one of which is advertisements and the other is packaging and selling up a user’s data and both of them are subject to a lot of newfound regulations as people are getting creeped out by them. Plus, literally everything that relies on advertisements ends up with a huge number of creepy scammer lowest-common-denominator advertisements and it keeps feeling like that’s a giant house of cards because eventually people are so bombarded by advertisements that they just get really good at ignoring all of it and/or you see equally effective advertisements for an entire industry and don’t end up preferring Coke or Pepsi. And then I don’t think people realized the degree to which companies were willing to vacuum up their data to make a buck. Essentially, costs are going up at the same time as the abilities to make money are going down.

Still, people are giving you this content for free for you to monetize. It’s got to be really embarrassing to fail at making money where you get the product you sell as well as a bunch of the moderation and curation that helps things along for free.

Furthermore, with ChatGPT, the shoe is on the other foot. See, before, it was the UGC site and their not-well-understood covenant with the random internet user, where the UGC site made money that they didn’t share. Now, we’ve got ChatGPT and other AI engines scraping the UGC site for content written by a random internet user, which means that the UGC site owners are suddenly on the receiving end of seeing their efforts used without compensation now.

It’s really quite disingenuous to talk about all of the expenses you are incurring and how you aren’t seeing a dime from AI training tools without simultaneously respecting that the world’s greatest moderated and architected UGC site is nothing if all of the users go away and, furthermore, nothing you are doing to try and squeeze money from the AI training tools is going to the actual creators of the content being scraped.

Cory Doctorow talks about “Enshittification” and while that’s happening too, there’s this unique aspect of UGC that’s separate from that. Even paid services with nothing to do with UGC have been enshittified as companies turn what used to be a product that you’d buy and upgrade when you felt there was enough to the upgrade deal to a monthly bill that you must pay.

If you look back, a Usenet feed and a directory on the web server for your own UGC was part of what you paid for. I do not think the Usenet model of community was necessary something that you could have scaled to meet the size of the modern Internet even though I think that most UGC companies would benefit from someone in the room who can say “Well, on the Usenet, we…” during every design meeting. Hosting on your ISP’s website meant that you could never change providers, so that’s also really problematic. To be clear, I’m not suggesting that the answer is to just resurrect Usenet or create some sort of distributed Internet filesystem that you pay for in your Internet bill.

Part of why everybody was comfortable letting Usenet and the tilde directory was because we were still believing in the promise that our UGC-specific hosting sites sustained by non-creepy forms of monetization was actually a possible thing. It felt like progress.

Folks admitted defeat early with the micro-penny-meter idea. It wasn’t until some ex-Stripe/ex-Square people started writing some blog articles about a bunch of details about why doing payments is so darn hard that it really sunk that it is fundamentally impossible and has always been so to actually make some sort of micro-penny-meter actually work in the real world.

But I’ve spent a lot of time explaining how all of these little silly communication things we do that has been wrapped up in this User Generated Content buzzword are both important but also how literally everything we’ve done starting with a bunch of Web 1.0 boom books and turning into the whole Web 2.0 dream and continuing onwards to now has completely not worked out well in the real world either. The way that Reddit and other UGC sites are completely prepared to alienate their core userbase just to reach some abstract fiscal goal, forgetting that their core userbase is the only thing that makes their site work, is just a symptom of the larger disease.

Maybe we need to figure out how to move forward with the understanding that the UGC boom that fueled all of the big social sites was just another of those Making Money On The Internet The Easy Way myths and was always a mirage.