Great piece on the Implemented blog this week about Web 3.0. Essentially, the piece tracks down lots of different definition of Web 3.0 that are floating around the web and sorts them into four main categories:

1. Semantic Web
2. APIs and Web Services
3. Mobile Web
4. Implicit Web.

The Semantic Web category seems to be the one getting the most buzz, and the one most people seem to think of when they talk about Web 3.0. But it is important to note that there are at least as many definitions of Web 3.0 as there are of Web 2.0 (in part because no-one really coined the phrase in the way  Tim O’Reilly did with Web 2.0). It was always an obvious step to start talking about Web 3.0 as soon as the Web 2.0 term got any traction, but inevitably it just became a buzzword to talk about all the next developments on the web.

I think the Semantic Web - the idea that computers will be able to go a step further in making connections between people and things on the web than they can at present based on various tags and other metadata - is the most compelling of these various ideas and so is deservedly at the top of the list, in that it involves the biggest change from the way the web works today and so provides the biggest step forward in terms of what people can do with the web.

The Web Services and Mobile Webs described in the post at Implemented feel like enablers or corollaries of the semantic web, and are in fact very much part of Web 2.0 as well. And the Implicit Web described at the end of the post is either an extension of the Semantic Web or the flip side of it - it’s still about computers deriving connections that are not explicit, but this time based on user behavior rather than tagging in web pages. As such it’s also an important part of Web 3.0 but probably secondary to the Semantic Web.

The post ends by discussing a few other views, including one that’s time-based. This is interesting but it really doesn’t make any sense because the various generations (to borrow an analogy from the mobile world) of the web co-exist at any particular period in time. Right now we arguably have Web 1.0 and Web 3.0 sites living alongside Web 2.0 sites and that’s likely to continue for a long time. The lines are very blurry indeed in the meantime. The term Web 3.0 is in some ways therefore even less useful than the term Web 2.0. But what we’re really talking about is how to make the web better, and what the components need to be, and that is a helpful discussion when it’s revealed by sweeping away the jargon.

I attended a couple of hours of the Money:Tech conference organised by O’Reilly Media in New York today. Tim O’Reilly himself - originator of the phrase Web 2.0 - was the keynote speaker, and was followed by a chat with Jim Cramer, host of Mad Money, founder of TheStreet.com, etc. etc. The conference was about Web 2.0 and financial services, and O’Reilly started out by talking about Web 2.0 and what it means to him. Ovum certainly has a definition of it, which revolves around four parts - social, business, content and technology models which define Web 2.0 services and sites. However, O’Reilly has a simpler definition, which stays away from specific technologies and services, and is simply this:

Web 2.0 is really about harnessing collective intelligence. It’s about creating a network-effects driven data lock-in with accelerating results to the winners. [I'm paraphrasing based on my notes but that was the gist]

In this way, O’Reilly says, it’s similar to Sun CEO Scott McNealy’s “red-shift” concept - that is, as you start to successfully differentiate yourself in something, your lead over the competition begins to grow ever more quickly. It’s all about creating business models which thrive off network effects - examples, according to O’Reilly, include Google (where the network effects come from the number of links people make), eBay (where the critical mass of buyers and sellers is the biggest barrier to competitive entry), Amazon (where he suggests the reviews are the key network effect) and so on. The value lies in accumulating data which leverages network effects in such a way that it is very hard for competitors to emulate what you have done.

Another major theme at the conference was open source software, and a debate during a panel session focused on whether open source adds or destroys value from a market. There were arguments on both sides, but it’s pretty clear to me that it destroys value for existing players, since it replaces proprietary products priced at a premium with free open source products. At the same time, it creates new opportunities for players which didn’t have the in-house resources to develop their own software, and it reduces the cost of doing business for everyone, which increases liquidity and therefore provides broader benefits.

So, how does all this apply to the OpenSocial program, the Social Graph API and efforts to create data portability? Do these effectively do to value in the Web 2.0 world what open source is doing in the software world? Does Facebook’s value proposition go away? Part of the answer may lie in something else O’Reilly talked about, which is Clayton Christensen’s “law of conservation of attractive profits,” which states:

When attractive profits disappear at one stage in the value chain because a product becomes commoditized, the opportunity to earn attractive profits with proprietary products usually emerges at an adjacent stage.

This would suggest that when open source enters a market, the value flees to the adjacent markets. And when data portability enters the Web 2.0 market, value will flee away from the Facebooks and MySpaces and to - where?

I would argue, as I’ve suggested in other entries, it flows to those best able to make use of the new technology - data portability - to create new services which thrive off it. I think this is the logical conclusion, and it’s another reason why Facebook, MySpace and others need to create value in something other than the information they hold about their users, because that will soon become commoditised and easily duplicated. They need to leverage that data in ways others can’t because of special sauce they themselves have concocted. It’s not clear to me that they have figured this out yet, hence (perhaps) their resistance to full data portability. But they’d better figure it out quick or that value really will go to someone else (and who would bet against Google here?).

I sign up for a lot of private betas when they are first announced and usually just forget about them until one day an email arrives saying the time has finally come for me to join the small in-crowd that gets to play with the latest online toy. That happened a couple of days ago with TypeRoom.

The premise of TypeRoom is that you can edit any website through a browser-based, WYSIWYG, HTML-free interface. To be able to play with any site, you don’t need to have any relationship to it whatsover. However, I’ve just tried editing the CNN and Patriots websites to say something amusing about the Giants’ win but the CNN website never actually came up for editing and on the Patriots site it was clear that you couldn’t put new text boxes in, only edit existing ones. So there are some limitations. But in my circumstances - where I have created several websites using Dreamweaver but not have a work laptop I can’t install my own software on - it’s perfect for the odd tweak here and there. You can set up FTP options or you can use the site to send an email with the HTML file attached to your webmaster (pity those poor webmasters who are now going to get emails from every Tom, Dick and Harry who figures out how to use the site with “helpful” suggestions for improvements).

At this point, it’s a great beta, but it will probably need quite a few improvements to really be useful to anything but the most basic users.