Here's a screenshot from Tim Berners-Lee's opening presentation:

Is the Semantic Web a Dream?

The first question is about the Semantic Web -- is it still just a dream? Nova Spivack of Twine says simply: yes! He says that the problem gets harder as more information comes on the web. You need to "disambiguate data". The AI approach puts burden on the software. The Semantic Web approach puts the burden on the data itself, so it's not about making smarter software - but smarter data.

He says there are technical and social challenges. Nova asks, rhetorically: is there an alternative to the Semantic Web, as the Web grows and scales? He says the HAL-9000 AI approach is an alternative, but it is not progressing much. Another solution is to "use the crowd", and he says while this approach shows promise, it doesn't scale to solve the problem. So both machine and human approaches won't scale -- hence the Semantic Web is the solution. He says "it's a huge cultural project" and is a long term goal.

Deborah McGuinness from Rensselaer introduces the debate, including a mention of ReadWriteWeb as part of the debate's "social media twist".

AI's Role in the Web's Future

Nigel Shadbolt is a Professor of Artificial Intelligence and Deputy Head of the School of Electronics and Computer Science at the University of Southampton, UK. He kind of defends AI, although he says we can expect "a very different form of AI" to what has gotten most of the press thus far. He says that we can expect "augmented collective intelligence". He says "fragments of micro-intelligence" will evolve into an ecosystem, so AI will contribute to the future of the Web in that way.

Multi-lingual Web

The next question is about a multi-lingual Web. Wendy Hall is a Professor of Computer Science in the School of Electronics and Computer Science at the University of Southampton, UK. She talks about the Chinese Web, which has 5 billion pages that Google doesn't index. She says that soon the dominant language on the Web will be Chinese, and much of that data will be via mobile phones. She says that a lot of "key sites" in the english language Web are not accessible in China, so the role of government will in a big way determine what will happen. She says that educating government all over the world will be important. The Web is already fragmented, but the solution isn't to teach everyone english. So the Web Science Lab has been established in China and progress is being made there.

Nigel Shadbolt points out that different cultures frame information quite differently. So he says there must be multiple ways to represent content on the Web. Nova says that each culture has to map their content to a conceptual representation, and then globally we need to map different culture representations together.

Tim Berners-Lee says that sometimes that will work and sometimes it won't. He says "the diversity of culture is really important" and he lists a kind of gradation of content that we can understand -- e.g. 10% of Web content everybody can understand, x% we can't understand, etc.

Democracy and the Web

A question from the audience about democracy. Wendy replies that the Web does have the potential to change the way we select politicians to represent us. So she says it's been interesting to watch the US presidential election - Obama has been using new technology and could bring young people into the process. So the Web brings a wider representation of people to be involved in democracy. She says it has the potential to "dramatically change" the way we select governments, mentioning the self-organizing of Wikipedia. "You can well imagine something which can shift what we do" in democracy.

Nigel chimes in that the Web can both mobilize democracy, but also there is danger of "cyber vigilantism". He mentions the dangers of a "rampant blogosphere".

Nova also mentions the "threat to democracy" of the Web - he says "freedom is actually at risk". He notes data logging and privacy issues. So he says it's too early to tell how the Web will affect democracy - he says the Web was built on trust, but that nowadays the Web is being mis-used in some ways. So we need to be worried about that - encryption, privacy, etc are things to look out for.

Web Science?

There is a question about ontologies, but (ironically) the answer got rather complicated and so I lost the thread :-)

Next an audience member asks: what's the difference between Web Science and Computer Science? Wendy replies that Web Science is inter-disciplinary and it's designed to get more people than just computer geeks into studying this domain.

Multi-modal Data; Can Semantic Web Capture Nuances?

The next, rather rambling, question from the audience is addressed to Nova and is about data. The question (when it eventually comes) is: what is the Semantic Web when it comes to different kinds of data (multi-modal, subtleties, nuances). Nova's first response to this involved question: whoa! He then says that we have to start with the simple cases. He talks about time and calendars, semantic representation of events, etc. But he says the Semantic Web won't capture the nuances of human interaction any time soon. He says "we're not trying to replace human intelligence", but "free up" human intelligence.

Nigel, coming from AI angle, says that behavior is a key part of the puzzle. He says understanding behavior, on a Web scale, is happening. He talks about sensors - where sensors report on data in our environment. 'Ambiently intelligent environments' is his term for this.

Tim's response: "an ontology does not represent the same thing as a haiku does." The audience laughs appreciatively. He interprets that as mening there are different languages for different things.

Will there be an Innovation in Logic?

The next question talks about "an innovation in logic". For example "visual logics" will need innovation if the Web is to understand multi-modal information. Tim's response is that logic has given us formalized reasoning, but it doesn't describe how people think. So he doesn't believe we need to formalize some kind of data as logic (e.g. the way people dance).

Nova says that we don't have an equivalent of a functional MRI for the Web, in other words there aren't enough ways to measure things on the Web. He says you need sensors and math to do measurements, rather than logic.

Conclusion

The final question is about data: how do you validate it, and if it's to be shared, how to you manage privacy etc. Nigel points to Wikipedia as a way to validate information, self-correcting by people. Nova says reasoning is where the Semantic Web can help validate data, as well as the human self-correcting (e.g. wikipedia).

And that ends the debate. Feel free to contribute comments below!

Richard MacManus, Editor, ReadWriteWeb:

1. Semantic Apps will become popular in 2008, due to their ability to get better content results and make better data connections. Think search engines like Hakia and Powerset, wikipedia-like efforts like Twine and Freebase, and apps that use semantic technologies under the hood (such as AdaptiveBlue and Snap).

2. In tandem with #1, Google will experiment more with Semantic Apps in '08. The Knols project, although not overly semantic, is a hint of this direction.

3. Web Services platforms will be a fierce battleground in '08, with Amazon, Microsoft, Google, Mozilla and others competing to provide 'Web OS' and online storage to consumers. Unfortunately this may spell the end of a number of startups in this space.

4. Zoho and/or ThinkFree will be acquired by big companies wanting to leapfrog into the Web Office space.

5. The online advertising market will consolidate, after the spate of acquisitions in 2007. CPM will continue to dominate for media brands and CPC for niche sites, although there will be experimentation in VRM and other forms of highly specific targeting of ads. Privacy issues will prevent the latter from becoming mainstream though. The much-hyped CPA (Cost per Action) will continue to be a pipe dream, because publishers simply don't want it.

6. The big Internet companies will surprise us all by embracing open standards, and attempting to compete with each other with features instead of data lock-in (OK, this could just be wishful thinking!).

7. The most interesting innovations on the Web in 2008 won't happen in Silicon Valley, but in Asia (China, Japan, Korea). At least one startup from China will break through in the US market with Twitter-like success in 2008 - and it will almost certainly be a Mobile Web app.

Marshall Kirkpatrick, Lead Writer, ReadWriteWeb:

1. Twitter will be acquired.

2. Most ad networks will start producing their own content to advertise against; and some content companies today will get acquired by ad networks.

3. Online video will become so ubiquitous, including live and mobile, that everyone will wonder how the internet existed without it. It won't feel like a big deal, though.

4. A handful of big companies will let you start logging in with an OpenID associated with your account.

5. The value of recommendation engines will become all the more clear; the era of data will be celebrated.

6. People will rebel against Google, at least a little bit. Maybe.

7. People engaged in the new web will do some really awesome stuff that we'll all be in awe of.

Josh Catone, Lead Writer, ReadWriteWeb:

1. Tumblr will be acquired.

2. Privacy will be a growing concern in the mainstream, but ultimately people won't really take any action and for the most part, things won't change. Some companies and groups (think Mozilla) will push for better privacy controls for users, while others (think Facebook) will continue to push the envelope and continue down a slippery slope. Users will eventually push back, but I am hesitant to say that proverbial "straw that breaks the camel's back" will come in 2008.

3. OpenID will be adopted by more startups and larger web companies, but most people (mainstream users) still won't use it - that's a couple of years off.

4. Facebook will continue to grow and their platform will be adopted by other large social networks. Google will sweat.

5. Mobile web usage will be a big story in 2008. It's already big in many parts of the world; and Westerners are about to get hooked. With new mobile devices that makes web surfing less painful, people will be more and more connected away from their computers.

6. Mainstream media coverage will be a catalyst for the adoption of Web Office apps by consumers; and Microsoft will eventually be forced to change their Web Office strategy and offer a fully online office suite (but that latter won't happen in 2008). Offline mode (Gears, AIR, Silverlight, etc.) will be what really tips the scales and causes mainstream users to to embrace the as-of-yet unfamiliar world of Web Office applications.

Alex Iskold, Feature Writer, ReadWriteWeb:

1. 2008 will be slow and cautious, with the first half dominated by recession or fear of recession.

2. Facebook is going to see the same kind of decline in popularity in 2008 that MySpace saw in 2007.

3. Digg is going to be acquired by one of the mainstream media conglomerates.

4. Implicit applications, which monitor our habits and automatically infer our likes, will rise.

Emre Sokullu, Feature Writer, ReadWriteWeb

1. Facebook will acquire companies that do the following, in order to strengthen their advertising unit: personalization, behavior tracking, image recognition (Riya?)

2. Facebook will release a browser.

3. However, despite all that... Facebook will decline.

4. Google OpenSocial will be a failure; Google will try to create its own social networking empire by making acquisitions in this space.

5. Microsoft will become more aggresive and buy many popular companies at once (remember Ballmer's quote). Candidates include SixApart, Technorati.

Sean Ammirati, Editor, ReadWriteTalk (our podcast show):

1. Google will really start looking vulnerable in 2008. While the 'one trick pony' comment by Steve Ballmer drew sarcastic responses, this will begin to look prophetic. While they'll maintain market share in the search industry, the lack of traction in any other of their other initiatives will start to cause frustration. Plus, they will increasingly be perceived as the 'evil' company in many of these new initiatives.

2. Closely related, Yahoo's Hack strategy (see ReadWriteTalk's podcast with Bradley Horowitz) will start to bear fruit and things will look much more optimistic in Sunnyvale this year.

3. Facebook will start to feel pressure from two trends that will emerge on the web: distributed social networks and distributed commerce systems. For distributed commerce systems, look to see a first proof of concept from the VRM project. Chris Messina's diso project with Wordpress will be a great proof of concept for distributed social neworks.

4. Non-search advertising on the web will increase in value significantly. This will be done through a lot of innovation in the ad targeting systems (both behavioral and contextual) and new metrics being adopted by Madison Ave beyond CPC and CPM.

5. There will be a lot of innovation in the hyper-local space, putting the final nail in the newspaper industry's coffin. This will include companies like Outside.in and Yelp moving toward widespread use and new web properties (from both startups and big Internet Cos) emerging.

6. Finally, a 3G iPhone! OK, I don't know if this is a prediction, but I really really want it to be true :)

Charles Knight, Editor, AltSearchEngines (RWW network blog)

1. In the 1st Q 2008, the true "Google Killer" in search will be in Stealth Mode. In 2nd Q 2008 the first prototype will begin in closed Alpha mode. In 3rd Q 2008 it will be ready for the final closed Beta testing. In 4th Q 2008 it will launch and "Rock and Shock" the world!

2. The classic Vertical Search Engines (Job Search, Health, Consumer Electronics, Shopping, Video, People, more...) will continue their dominance over all other Search Engines in their various niches.

3. The Alternative Search Engines will pick up the pace of partnerships and cooperation, for their solid mutual benefit.

4. Mainstream Media interest in the Alts will increase until it begins to rival coverage of the five major search engines.

5. The trend towards 'widgetization' of the Alts will continue. Approximately 2 in 10 Alternative Search Engines (20%) have widgets now, and that number will double in 2008 to 4 in 10 or 40%.

Conclusion

Now it's time for you to tell us your Web predictions for 2008. Please leave a comment or trackback below!

Crystal Ball image by Blue Cubic Electron Syncrony, via Flickr