There’s been a lot of recent posting about ‘Web 2.0′, ‘Web 3.0′, and beyond…

The biggest fallacy in all of this web versioning mania is that it treats the web as if it were some vast, singular gestalt. The reality is that there are multiple threads of evolution taking place concurrently in this space. Each of these threads is developing at its own independent pace. It’s the assembly of elements from various points along these threads that ultimately emerge as “user experiences” – not an easily versioned Web ‘singularity’.

I believe the evolution of the web overall is best served by discussing the various domains that help make it up. These domains are like the tectonic plates that make up our globe, appearing as a single whole, yet moving independently and sometimes in conflict with each other. I believe that there are five of these domains that are shaping what we see happening in the web space today. They are each at various levels of maturity, and each faces distinct challenges in their continued evolution:

1. The Interface Domain

Most people experience the internet through a browser. The browser model was designed to to let people easily navigate through the early web – little more than a globally distributed hypertext deck. It was a basic environment that was light on media and heavily textual. HTML – the web page format interpreted by the browser – became the basic descriptor for site implementation. It has since extended its capabilities with CSS and now AJAX to become a more dynamic user application environment. Adobe’s Flash – a web media technology – has also evolved into a significant component of the modern web experience and is used for almost all video on the web.

While today’s web experience is vastly different from the initial ‘Netscape 1.0′ experience of the mid-1990′s, the overall web access model hasn’t really changed substantially. It is still driven by search, bookmarks, and links. It still has a Home Page and Back/Forward navigation. It’s also something that you still need to ‘visit’ (using some browser type of tool). It isn’t yet embedded in software and devices in a way that makes it both ubiquitous and invisible. I believe current web interface models are still in their early days, and think they still have a long way to evolve.

2. The Social Domain

Social networking sites like Facebook, MySpace, and LinkedIn – the best known expressions of the social web – are based on the concept of personal identity. They depend on people ‘advertising’ who they are, and providing sufficient personal details (in a structured format) so that other people can discover them. The goal for sites like these is to make – and then leverage – connections between specific individuals.

But there’s much more to the social evolution of the web then just that.

Another side – made popular by del.icio.us – is social tagging. Social taging allows people to classify web sites they visit using text tags, and leverage what I call reputational identity. People have identity, but it’s only there to identify them as the creator of their tag set. People can look for tags created by other specific people, but they don’t get to see any details about them.

Yet another aspect of the social domain is the assignment of relevance in search. This area uses putative identity to define relevance. It isn’t dependent on having hard individual identity of the authors of specific web pages. Rather, it derives the authority of these authors through indirect means like the preponderance of links to their content from other ‘highly relevant’ sites. This network derived ‘authority’ becomes an authors identity in this domain. They have no direct role in defining it, yet it’s probably the most significant factor in the social discovery of what they publish.

There are also social technologies like Bittorrent that depend on technical identity – they use vast networks of individuals to create efficiencies around technical actions (like downloading a file), and only require sufficient identity to establish connections and manage resources. There is no personal identity involved in this.

I think the biggest issue that needs to be address for the social web to really move forward is the evolution of a model for verifiable personal identity. The vast potential this thread of development holds will remained shackled without a global framework for trust, reputation, and identity assertion and management. I believe the social domain has probably plateaued waiting for this to happen. Once it does, I expect to see this take off in many different directions.

3. The Connectivity Domain

Connectivity is all about the throughput and ubiquity of links to the internet. At a macro level, connectivity has been evolving at a pretty good clip over the last decade. The number of individual households connected to the net has grown steadily, and the bandwidth of those connections has continued to climb as well. That said, connectivity growth has been uneven globally, with certain parts of the world virtually invisible on the net (less than 4% of Africans are online!), while other regions (like parts of Asia) are aggressively expanding.

Unlike all of the other aspects of the evolving web, connectivity isn’t virtual – it’s physical. It involves hard costs and regional priorities. It runs up against land use rights. It needs to work around entrench interests looking to derail projects they feel threaten them. It crosses townships, jurisdictions, and borders – and that means it involves politics. And that means its messy and difficult to predict.

If you look at the battles that have taken place between individual cities looking to install municipal wifi, and the big telecom providers like Verizon and AT&T, it’s clear the battle lines are forming between the old and new guard here. The adoption of new technologies like WiMax scares the hell out of most cellular carriers because it can bring VoIP to the mobile market – and kill their current business models. (I’ll need to see how open Sprint will be with their WiMax implementation) You have China trying to filter out services like Skype to prop up their state run phone service monopolies. And the debate of network neutrality in this country is starting to shape up in a similar way. No one’s giving up without a fight.

Unlike other aspects of the evolving web, technology isn’t going to be the gating factor in the evolution of connectivity. It will be a political battle that, unfortunately, the old guard is better positioned to wage right now than the new guard. That said, Google is a wild card here, especially in the upcoming auction of spectrum. They have the resources to really shake up some of the entrenched interests, and the vision to move things forward. But any success they have would only impact the US, and it would do little to improve services to the most pervasive web platform globally – the mobile phone.

I believe connectivity will continue to mature in fits and starts. Openness and capacity will be distributed unevenly – not just globally but within this country as well. Progress in many areas will come reluctantly, and at the minimum level needed to keep the political interests -and hence, regulation – at bay.

4. The Content Domain

Content is exploding on the web. The evolution from passivity to participation is moving at a good clip and accelerating. This is being fueled by a combination of increasing bandwidth, easier to use, free publishing tools, and a shift in the cultural zeitgeist that now finds value and satisfaction in this form of self-expression. Media of all kinds is being produced and distributed via the net, bypassing the traditional gatekeepers who are powerless to stop it. This has allowed a viable micro-publishing model to emerge, where content is created and packaged by small groups to serve the interests of increasingly small market demographics.

The content thread is probably the most evolved of all the threads that make up the net, but it still faces some significant challenges. Censorship and disinformation are very real threats (just look at what happened in Burma – content still travels through physical wires and boxes, and these can all be controlled.) The lack of global consensus on a ‘Web Bill Of Rights’ leaves us in a situation where individual countries are attempting to apply their local laws to the global web, raising real issues around freedom in a trans-national space. On the business end, effective models for commercializing content have yet to emerge, which combined with piracy, is slowing the creation ‘professionally’ generated content. Cultural balkanization is also a concern. Non-English content is continuing to grow in volume and importance, but a technological foundation that can effectively handle discovery and translation doesn’t exist yet. And it may be a long, long way off.

Despite these challenges, I feel good about where content is at right now.

5. The Organizational Domain

The organizational domain is the one that seems to get everyone pumped up. It’s all about how information on the web gets classified and organized. Using current search tools, it can be very difficult to find many of the less common things you might look for on the web. The signal to noise ratio is very low, and finding specific details can require a great deal of effort.

That said, almost every recent discussion on the limitations of current web search technology also ends up talking about the “Semantic Web” and how it will help straighten this situation out. It has been touted as the ‘next evolution’ of the web.

The thought of a fully tagged web is a compelling one. Easily find just what you need. Make comparisons without visiting dozens of sites. Have the ability to create mash-ups out of virtually anything. It seems like the answer to all of the issues we struggle with today. Unfortunately, I think there are some very bright people that are simply glossing over the practical aspects of moving this academic concept into the real world. Implementing the promise of the semantic web will require reaching a global consensus on how things should be tagged, and then having everyone do that tagging themselves when they create new content. I see little chance of either of those things happening in a meaningful way.

People are generally lazy when it comes to things like tagging, and doing it correctly takes both time and effort. Most people will do the minimum they need to do here (which may end up being nothing). Without having a through job done on the tagging everywhere on the web, people that are looking for information will still need to use traditional search methodologies to find things. If they simply count on every site having precise and complete tags, they’ll risk missing out on lots of valuable content.

I also have no doubt that ‘semantic spam’ will emerge, distorting search results deliberately, or being included in results by taking liberty with how much they respect the intent of the query. Over-tagging content to raise its visibility has always been a problem in the professional content space, and I see it being even more problematic if implemented across the entire web.

On the commercial side, I don’t anticipate a rapid adoption of the ‘Semantic Web’ either. If I were a retail business, I would be reluctant to disclose detailed information about my inventory levels or prices, especially if my competitors could look at it just as easily as my potential customers. And that type of disclosure might not capture the key aspects of my value to the market. I may do specialized in-home installations, or provide unique types of training or consulting, or have pre-confgured bundles of goods that better serve my target markets. It’s possible that none of that could be expressed in a meaningful way using a fixed schema. And I wouldn’t want to find myself dismissed out of hand for not falling into the top three “best” stores based solely on a single unbundled price. If I thought there was a potentially meaningful downside, I’d simply avoid it.

Building the foundation beneath the Semantic Web is also a huge undertaking. Having worked on multiple industry standards bodies, I believe that reaching agreement on the broad set of schema needed to make the semantic web really work could prove elusive. Various parties in the marketplace will be advantaged or disadvantaged based on what’s in or out of a particular schema (a taxonomy + enumerations). In defining a schema, you essentially define the question you want people to ask to discover you. Everyone will want the schema to ask the question they know they will have the best answer for. As a practical matter, a schema that everyone in a particular discipline can agree on will either be way to complex or way too simple to be a meaningful tool for discoverability.

I also think that there is an ontological problem with the Semantic Web. Classification depends on definition, and how a definition is applied is often a matter of perspective. What makes someone a ‘discount supplier’ or a ‘full service dealer’? What makes an item ‘rare’? This conceptual definition of a space is called an ontology. It’s not about taxonomies or enumerations, but meanings.

Reaching agreements on the meaning of specific terms – even within a single language – can be a challenge. Having to deal with multiple languages and cultural references complicates it even more. Most legal contracts devote pages to defining a few significant terms, and base those definitions on the precedents established by courts in related litigations. Think of all of the unique terms (enumerations) that will need to exist within all of the different taxonomies that will end up being created for this effort. They will all need to be exactly defined. Even if it ends up being possible to get everyone to agree on the various taxonomies and their enumerations, I don’t believe it is practical to achieve concordance on rationalized ontologies.

I’m sure there are many people with a differing view on on this, but far as the evolution of the organizational domain goes, I would not look to the Semantic Web for a solution. I believe it will come down to how the search engine space evolves. I think we will continue to see a refinement in the way the main search engines index the web and integrate more social cues into defining result relevance. I also see the emergence of a ‘long tail’ in search – smaller, more vertically focused search tools that addressing specific market segments in a very deep way. In general, I expect all search engines will focus more on delivering ‘goal driven responses’ that return a range of potentially useful content related to predefined common activities. There is a lot that can still be achieved using this approach.

So where does this leave us?…

While it may be beneficial from a marketing or fund raising perspective to hang a web version number onto a particular type of technology or service, it really distorts what is happening in this space. I see many people developing for the web that are consumed by an almost sophomoric enthusiasm to rush ahead to the next sexy thing. Unfortunately, unless we find ways to solve some of the tough foundational challenges that exist right now in the various domains I outlined, the web will never reach its real potential. By flipping through “web versions”, we’re only creating the illusion of crossing major milestones. All these issues will still need to be addressed at some point.

And they won’t be any easier to solve by simply jumping to “Web 4.0″…