The Four Webs: Web 2.0, Digital Identity, and the Future of Human Interaction

Published 1/05/06

Web 2.0 has been a buzzword shrouded in mystery. Although I've heard it used hundreds of times in the past year, I've never been able to find a good definition of what it actually means. This is in part because, as Paul Graham points out, Dougherty coined the term before defining it.

When brainstorming on Web 2.0, O'Reilly laid out a list of services that seemed qualitatively different than those that had come before and looked for patterns. What he eventually came up with was a list of seven principles and eight patterns. O'Reilly hits the nail on the head here with his inductive approach. However, I still come away from O'Reilly's manifesto wishing for definitive clarity.

So here I set out to put forth a working definition of what Web 2.0 really is, how it differs from Web 1.0, and how it will differ from what may come after. I am not proposing that these are absolute definitions, but rather I am sharing something that has been useful for me. So here we go:

Web 1.0 is about allowing individuals to create and share ideas.
Web 2.0 is about allowing groups to create and share ideas.
Web 3.0 is about allowing societies to create and share ideas.
Web 4.0 is the singularity.

The differences between each of the four have to do with something called digital identity, which I will explain using blogging as an example. One trend you will notice is that, paradoxically, the stronger our individual identities become, the more harmoniously we are able to coexist.¹

Web 1.0: allowing individuals to create and share ideas
Few people know that the first webpage ever created, Tim's home page, was actually a blog. Blogs are the epitome of web 1.0. They focus so much on the individual that even Ayn Rand would blush. At their best they can be truly uplifting and inspiring, but on average there are some serious problems with blogs as they exist today. Blogging will be around forever, but I think that personal blogs will be dead in another two or three years.² I have four reasons for this:

• The value of blogging comes mainly from two things: one, the feedback you get from others, and two, the ability to spread your ideas. In order to get enough people reading a personal blog to generate an interesting conversation I'd have to share something insightful every day for months or even years. And even then it's unlikely that I'd ever get more than twenty or so comments on average. However, I can go on a collaborative blog like Kuro5hin or Dailykos and write a diary about any random idea and get over 100 comments the first time if my ideas are any good. I'm not gonna lie, I usually don't even think of one insightful thing a day for myself, so trying to share something daily with others in order to gain a following is a non-starter for me and most other people. If I wanted feedback from just my friends and family I'd just ask them, and if I wanted feedback from lots of random people I'd just post somewhere where lots of random people will be reading. Personal blogs are a poor solution to both.
• Secondly, personal blogs aren't emergent. The value of interconnectedness is that you can propose a part of an idea and hundreds of other people can work to complete it. Or you can post a fully thought out idea and hundreds of other people can critique it so that the idea can evolve. When everyone posts their own separate opinions of the same events on their personal blogs there is no emergence and less new insight is gained.
• Thirdly, right now our public and private lives are translucent. As our public lives get more transparent, people may try to make their private lives more opaque to compensate, albeit our conceptions of public and private life will change.
• Lastly, most people don't have anything interesting to say most of the time. Eventually people will figure out that the world doesn't revolve around them and no one gives a shit about what Johnny said about Suzie at the pool last weekend. Certainly there is a place for this type of stuff, but because people feel obligated to post every day there is a lot of unnecessary signal pollution.

So I believe the way of the future is in collaborative blogging. Specifically, posting when you have something good to say and learning to keep quiet when you don't. That way writers get the most reward for the least effort, and readers get a better signal to noise ratio. As fate would have it, the theory is modeled after reality; the few collaborative blogs in existence today are doing remarkably well.

To give a concrete example of the benefits of collaborative blogging, let's take a look at one of my interests: amateur athletics. For many years I was a rower and I read all of the rowing magazines, newspapers, books, websites, and newsgroups as frequently as they were updated. However, with that volume of reading there is no way I could ever do the same for another sport, let alone several other sports. This sucks because there is so much I could learn from other disciplines that I could incorporate into my rowing, maybe even my life. But I just don't have time to read an additional fifty or so webpages, newsgroups, and bulletin boards each day. And I'm willing to bet that there are other amateur athletes who feel like me. Enter a new collaborative blog on amateur athletics. When I have some interesting insight to share I share it: when I do an interesting workout, having an eventful practice, a great race, a terrible race, etc. There are plenty of like-minded people to read my posts and be hearty in their approbation and lavish in their praise. And other people can do the same. And I can learn from them. And they can learn from me. And life would be good.

However, right now there are still some problems with the way collaborative blogs scale.

Web 2.0: allowing groups to create and share ideas
Your "use case" should be, there¹s a 22 year old college student living in the dorms. How will this software get him laid? —JWZ, via Many-to-Many

Trust is the currency of the participation age —Jonathan Schwartz

The greatest asset of collaborative blogging is that anyone can contribute. The greatest pitfall of collaborative blogging is that anyone can contribute. Once a collaborative blog gets a certain number of eyeballs there becomes an irresistible temptation to cause a little high-visibility mischief. Or a lot of high-visibility mischief. And so collaborative blogs seem to be capped at a fixed signal to noise ratio. To understand this one needs look no farther than K5. Once Slashdot got to a certain size it split in two, Slashdot and K5.³ And once K5 got to a certain size it split in two again, K5 and HuSi. The higher the quality, the more eyeballs; the more eyeballs, the higher the temptation for evil.

Digital identity is the name we give to the collection of technologies that will enable us to smash through this quality barrier. In the physical world we are largely accountable for our actions. In cyberspace we are not. In many cases this is a good thing; take anonymous political dissent for instance. In other cases, not so much. So if you get the feeling that the accountable web will change everything, it won't. Identity is just another tool for our toolbox; the more tools we have the more things we can do. Rather than approaching this in an academic way, I'll give an annotated example of one possible digital identity solution to the problems above.

Since we live in a capitalist society, these problems will most likely be solved by capitalism. As such, our digital identity solution is judged against three basic criteria.

• It solves a customer's needs or fulfills their desires.
• It provides more value than it costs the consumer.
• It can be sold for more than it costs to make.

In addition, there are seven laws of digital identity which must be followed. You can read Kim Cameron's full manifesto at your leisure, for now though just make note of the first three:

• Technical identity systems must only reveal information identifying a user with the user's consent.
• The solution that discloses the least amount of identifying information and best limits its use is the most stable long-term solution.
• Digital identity systems must be designed so the disclosure of identifying information is limited to parties having a necessary and justifiable place in a given identity relationship.

Product, Price, Promotion, Purple Cow
The solution is to create a system of anonymous accountability that enables webmasters to permanently ban problem users. That is, users can be banned without even knowing who they are. Currently, most scaling problems stem this inability to ban abusive users; there is no way to prevent a person from creating a new username. Nor can they ban by IP, because you can't ban all AOLers if one is causing trouble. Just ask any Wikipedia admin. No longer will a tiny minority be able to ruin it for everyone.

The system itself works like this:

1. A user has their iris scanned at one of several hundred nationwide locations. If their iris isn't already in the database then they receive an activation number, which enables them to create an account.
2. They enter their activation number at the identity company's website. If the activation number is valid, they will be permitted to create an account. The account does not store the user's iris or activation number, nor does the iris or activation number point to the account; that is, there is no way to determine someone's account based on their iris and there is no way to determine their iris based on their account.

To make this more clear, let me explain how the same technology has already been used by the United Nations in Afghanistan. When refugees were crossing the border into Pakistan, they would wait in line for their relief supplies from the UN. The problem was that once they got their supplies they would just go back to the end of the line and no one would be the wiser.

The solution was to scan each refugee's iris when they received their supplies. This way if they tried their scheme again, the computer would simply recognize that their iris was already in the database. The privacy implications were minimal because the UN workers didn't even know the names of the people they were dealing with.

In the same way, this would safeguard the privacy of Internet users better than our current identity patchwork. Currently, there are many sites where users must enter their full names and addresses to access content. In some cases this makes sense; for example, Amazon needs your address and credit card number to ship your order. But in most cases this information is collected only to limit abuse. Not only is this method ineffective, but it has grave privacy implications as well. Once digital identity evolves to the point where we can create systems of anonymous accountability, it will no longer be necessary or justifiable for sites to collect this information, unless of course a user opts to share it.

To create a nationwide network of a few hundred iris scanners would cost somewhere between six and fifteen million dollars, depending on whether the cameras were human run on existing computers or fully automated kiosks.⁴ Somehow this would have to be paid off, with enough profit to satisfy founders, employees, and investors. One way to do this is by using something similar to the proven profit model that cable TV and XM radio use. Charge an initial fee to get your iris in the database, say fifteen dollars. Then charge an additional five dollars per month. Of those five dollars, keep one dollar for the company and distribute the remaining four dollars to content providers like a pie. For example, if a user subscribes to four websites, then each of those websites gets a dollar per month. Or if that same user subscribes to eight websites, then each gets fifty cents per month. The genius of this is that it not only does it give content providers a good reason to implement the system, but it gives them an incentive to turn their customers into your customers. In short, the promotion takes care of itself.

The obvious question then is what happens to users who don't want to participate? Do you, for example, ban them from posting on Kuro5hin? That probably wouldn't make sense in most cases. The real money is in opening up new services that aren't even possible today. For example, it would be great if you could comment directly on New York Times articles, but this isn't really possible because of the potential for abuse and legal liabilities. Remember the LA Times failed Wikitorials experiment?

But even established websites could benefit from this system by giving plums to enrolled users. For example, if you make a premium account on my website then I'll disable your banner ads. Currently there are a few websites that allow banner ads to be disabled, but their profit models make no sense. Why should I pay four dollars per month to turn off banner ads on a website if that website is on average making only twenty-five cents per user per month?

I am using blogging for my case study here, but the same identity system could be used for for anything from reducing cheating in Party Poker to controlling griefers in World of Warcraft. And this is only one theoretical identity system. If digital identity ever develops to the extent that experts predict, there will be an identity meta-system that allows for many different types of digital identity to harmoniously interact. Use one identity system for purchasing alcohol online, another for interacting with the government, another for your online shopping, and yet another for blogging and online gaming. Some of these identities would be linked to your physical identity, others would provide anonymous accountability, and still others would only prove that you were a human and not a computer.

Want to hear some crazy uses for Web 2.0 style digital identity? How about a protocol that allows you to post on your blog saying that you are selling your bike, and having it automatically appear on eBay and Craigslist in a way that is secure and spam-proof. Impossible you say? It's called structured blogging and it's already being developed. Or even better, how about the reverse? What if you could tag the content you posted to various online spaces with your identity? Then at the end of the day all your content shows up in your queue, and you could check the box next to each post to have it appended to your blog. Or, for the lazy, set your personal blog to automatically display your highest rated posts from around the web. Show off your best, shit-can the rest.

So, how does all this identity speculation relate back to the current state of Web 2.0 technologies? I started off talking about how Web 2.0 was about allowing groups to create and share ideas. While it's true that there are some truly great Web 2.0 platforms out there, only when our current identity patchwork is replaced by formal standards will the space truly explode. Currently, most of the websites that can get away with uncensored comments don't have enough money to be worth suing. This is bad. The reason Web 2.0 technologies are often described as empowering is because of the people. And the more people we help over the new digital divide, the more power we get. As Cory put it, it's a sheep-shit-grass situation.

Web 3.0: allowing societies to create and share ideas
If Web 2.0 is about groups, then the next logical step would be groups of groups — societies. There is one major problem with this though:

Web 3.0 is when the web stops being about sex.

When I think about allowing individuals to express themselves, I think sex. And when I think about individuals forming groups, I think sex. But when I think about societies expressing themselves... What were we talking about again?

This is a problem, because the adoption of most communications technologies has been driven by sex. While there are already many Web 3.0 systems being developed, without sex the technology may develop much more slowly than it would otherwise. Sure we may have eVoting already, but without sex to drive identity and accountability to the next level, will it ever be secure?

It's hard to imagine what new identity technology will be needed to fuel Web 3.0, because frankly it's hard to imagine even a small fraction of the possible Web 3.0 applications. The best place to look for inspiration at this point may actually be speculative science fiction. For example, Cory Doctorow's book Down and Out in the Magic Kingdom incorporates a social identity system called Whuffie. Set in a post scarcity economy, one's social standing, or Whuffie, is what drives ordinary schmucks toward greatness. Perhaps Cory will be to the future what Leonardo da Vinci is to today.

Or perhaps we can draw some inspiration from a simple pen and paper game. Nomic and it's superset, internomic, provide an excellent model of what a Web 3.0 enabled society might be like. Or alternatively, what if our constitution resides on a wiki, but the difficulty of making edits approximates the difficulty of passing a constitutional amendment? In truth, we can speculate all we want but it's just too soon to know what Web 3.0 will be like for sure.

Web 4.0: The Singularity
So you want to know how you can make money off the singularity? Write a book about it. —Ray Kurzweil

When thinking about singularity, Neil Stephenson's Snow Crash is a good place to start. Not because it takes place in a post-singularity world, but for its raw unselfing power. Was it me who killed the orc, or was it my character? Am I successful beyond my wildest dreams, or trapped in a masturbatory fantasy? Many of the trickiest questions posed by the singularity have parallels in the MMORPG type games that Snow Crash depicts. And when you pick up a copy, notice the review on the back by Timothy Leary. Indeed, Web 4.0 is truly the land of the absintheurs and temporal synesthetes.

But fortunately you don't have to resort to hallucinogens for singular lucidity. There is a wealth of writing on the singularity and what post-singularity life might be like. My personal favorite, The Age of Spiritual Machines, is like vision quest in a box.

On the subject of accountability, Kurzweil recently said, "We hold people accountable because we hope they will realize they are accountable and do something about it. Once actions are taken by non-humans we need to rethink this, and come up with a way of motivating intelligent non-human entities into acting responsibly." Vision quest indeed.

What the Bleep
The "killer apps" of tomorrow's mobile infocom industry won't be hardware devices or software programs but social practices. —Howard Rheingold

In his recent essay, Paul Graham pans Web 2.0 because it can't be used to make predictions. Paul is right; the reason is that we have been classing Web 2.0 by its technology instead of its social implications.

Because, really, who gives a shit about technology? I don't care about technology, I care about me. I don't want to know how Web 2.0 will get me AJAX, I want to know how Web 2.0 well get me laid.

When caught in the throes of our meme 2.0 ideations, it should be the social over the technological that inspires. When we do this, not only can we make falsifiable predictions, but we can make actionable business plans and compelling emotional appeals as well.

So if you think it's too late to start a billion dollar AJAX business... You're right. But don't worry; the revolution isn't over, it's barely begun.

Footnotes
[1] Reader exercise: Is this true in real life? Why or why not?
[2] Not dead in the sense that they won't actually exist, but in the sense that they won't have nearly as much influence as they have today.
[3] Rusty may have created K5 for other reasons, but it's a fact that many of K5's first users migrated from Slashdot because of the signal-to-noise problem.
[4] This is according to my own back-of-the-envelope calculations. It assumes the identity company would partner with a retail chain with an established computer network, e.g. FedEx Kinkos.