Pre-web the sharing of information on databases on compuserve et al provided a useful service for people trying to get things done in the real world. The sharing of information that came out of that in (for want of a better term) Web 1.0 was driven by greater access,not by better services, which were frankly awful. My green screen online databases were a lot faster than the www!But the information was valuable......

Web 2.0 has been driven out of the back of Web 1.0 by commercial application - unless the services prove really valuable the investment just isn't there. I'm afraid that we geeks live on the periphery of a hard nosed reality - social networking per se does not create cashable value.

For semantic web to take off we need to identify how to create tangible value out of the sharing of data, and while I agree that this will only be possible because people's behaviours and expectations change BECAUSE of experiences in Web 2.0, someone, somewhere, has to make money to drive it.

GW - CEO at WorkDocx

Solution: If you want the general population to do something, you have to explain WHY in such a way that they perceive a loss of benefits to themselves if they don't.

The term Web 3.0 is a mystery to mainstream America and the somewhat unscrupulous Marketers are having a field day with this "new technology" used to develop "Web 3.0 Software" promising instant double optin leads by the tens of thousands.

Over 30,000 wanna be marketers couldn't wait to pay $370 to get 40,000 leads every month from the newest "system" called www-dot-StreamlineFunnelSystem-dot-com. That works out to 1,200 MILLION leads monthly - all double optin subscribers. Wow, are you impressed with their new Web 3.0 software yet? Okay, then check this out:

SFS Web 3.0 Automated Sales System (SFS Flagship Service): Web 3.0 Premium Auto-Pilot List Building & Broadcasting Software. We generate the leads, broadcast your link to these leads, you get paid! Demo here:
http://streamlinefunnelsystemofficialblog-dot-blogspot-dot-com/2009/11/please-help-us-with-system-upgrades.html

So there you have it, that's the Web 3.0 people are too lazy to find out about for themselves. This is not even true ignorance either. People are not stupid either. They are just LAZY, plain and simply put!

Don't they know lazy is why they stay broke? Lazy is synonymous with broke...and deviant behavior that carries many connotations: loud, dirty, delinquent, lazy, alcoholic and drug addicted.... Now I've gone over the top! Help!

I do hope you enjoyed this humorous yet sad tale of Web 3.0 in the internet marketing world.

There are two basic ways to navigate a distributed, linked database 1) request the set of results you want (i.e. query criteria) using a natural language description (e.g. the schools of the presidents of the U.S.), and 2) navigate to the results set by following the links between sets, filtering data as you go. The former is quite difficult. The latter is a lot like how you navigate to files on your operating system. What's needed in this case is a browser that exploits the semantics encoded in those links between things to make navigation easier.

One such example is razorbase.com, which allows you to browse all the data in the LOD cloud posted in the article (a reload of the DB is planned for today-early next week, so excuse the occasional weird characters in the results set). The approach here is to point the user to the most worn roads through the dataspace, and display sign posts that help the user along the trail to their destination. These slides show some walkthroughs of how to walk the LOD cloud using razorbase.

Yay for linked data!! Glad to see that it's beginning to catch fire with the big players.

take a look at this: http://www.adac.de/Auto_Motorrad/Autokatalog/Expertensuche/default.asp Thats all german cars with any extras. Just click around and look how easy the datasets can be selectet (reduced). Is that a kind of application for web 3.0?

Here's an article that I recently wrote explaining a new concept that I have the feeling it could definitely lead us to that "promised" Web 3.0:

Artificial Intelligence 2.0
http://technotations.wordpress.com/2009/06/02/artificial-intelligence-2-0/

Leave some comments if you like it (or not) :-)

In the field of research in biomedical literature, it is also creating great expectations as to having more accurate results and being able to connect them to more DB resources.

Sharing a whole amount of data & informations, in real time in one single place, for a determined user (and all his preferences) is a great challenge though. But it looks like this is where we are going.

I agree - the linked data flowers blooming concept of emergent properties

I think the issue of Web 1 vs Web 2 vs Web 3 is similar to the difference between Data vs Information vs Knowledge

Its the connections between the items - the richer the connections and the context the more we move from data to information to knowledge.

As the connections between "things" on the web develop the more we move from web 1 to web 2 to web 3.

I guess we could say that 2009 is where we think we see Web 3 - the rich connections resulting from Open Data initiatives.

Search 1.0 was AltaVista. They only looked at the text in the document. The more frequently the keywords appeared, the more relevant the document. This was an easy target for spam.

Search 2.0 was Google. They began also looking at the linking between documents to determine "authority" and thus relevance. Link farms thus developed as a strategy for spam, but the results were significantly superior.

Search 3.0 is personalization. Google and others use query and click history to build long-term profiles of user preferences in an attempt to further disambiguate user intent. The pros and cons are discussed at http://blog.surfcanyon.com/2007/09/19/hold-the-pickles-hold-the-lettuce/. In general, however, long-term signals are weak and thus the ability re-rank results is modest.

The quantity of information on the internet is growing exponentially, yet, while the numbers might be creeping up slightly, people generally still search with two- and three-word queries. Reformulation is a growing issue. Something else is therefore need to help users dig through the ever-expanding lists of results.

Call it Search 4.0 or 3.0+, but we believe that solution is real-time personalization, or the immediate re-ranking of results based on implicit user feedback. The new information being exploited to improve relevancy is the "at the moment" actions of the user.

You may try it for yourself at http://search.surfcanyon.com.

Home Care, Eldercare San Jose

At AltSearchEngines.com, we think in terms of Search 3.0.

Search 1.0 is one dimensional or linear. It is the straight line of Google search results.

Search 2.0 is two dimensional, as in a Quintura tag cloud or KartOO image map or a KoolTorch page. It allows the user to better see their options, usually in one glance.

Imagine Search 1.0 being a simple list of cities, and Search 2.0 being a road map, which is more useful?

Search 3.0 is three dimensional, or a 3D effect. Think of SearchMe or CoolIris or best of all TagGalaxy.de or those new revolving tag clouds that you may have seen in some WP blog sidebars (WP Cumulus). The search results are Left, Right, and Forward and Back.

This is the difference between a road map and a globe. A globe shows you the true relationship between Russia and New York, a road map does not.

Finally, we are moving into Search 4.0 which is realtime search. Think OneRiot or Collecta or the Omgili stream.

The fourth dimension is time, and while these projects are not 3D engines adding the element of realtime updates, that will be the next thing that you see. In fact the best example that I can think of is the WP Cumulus tag cloud since it beautifully revolves in a simulated 3D barrel shape *and* updates automatically each time a new post is published.

There is no Search 5.0.

http://betterdot.wordpress.com/2009/05/07/evangelists-the-semantic-web-needs-you/

It's at http://www.semanticsincorporated.com/2009/05/tying-web-30-the-semantic-web-and-linked-data-together-part-33-structuring-chaos-.html

"Dr. M", I am really glad that you enlarge the concept of Web 3.0 to those other technologies, because I agree that a lot of them tend to be more mature and simpler to implement than the semantic web technologies, semantic technologies, or linked data technologies, whichever one want to call it.

As much as my comments on the semantic part of Web 3.0 have received good traction so far, my inclusive attempts at extending the concept to other technologies, such as the mobile web, cloud computing, multimedia data conversion, and the "embedded" web (RFID etc), have not been the subject of any comment. See my visualization of Web 3.0 here in that regard: http://gregboutin.typepad.com/.a/6a00e54f8bc15888330115707ec092970b-popup

Part of it may be that the business case and practicality of those technologies is probably easier to understand, and the semantic technologies look a little less ready and messier at this point. But still, I'd personally like to see increased crossover between those communities and how the semantic web could benefit them (and vice-versa). I have seen some posts talking about the applicability of semantic-based browsing to the mobile web given the small screen real estate of those devices, for example, and found them fascinating. It seems this is relatively unchartered territory, and I'd love to see more of it.

On a similar note, new monetization models for web 3.0 and the semantic web also is an important topic to tackle further. On a past semantic web gang podcast (February I believe), I raised the idea of a monetization model in which one could associate a small cost to every small entity of information, e.g. blog post paragraphs, nodes of RDF triples etc (just like you pay for articles today, except more granular), to compensate the creative process further. I don't know if that's realistic, but I think it is possible :)

I saw there will be a panel on that, called "Monetizing Connected Data, Not Pages" at the web 3.0 conference next week. Such discussions on monetization models, and even better, practical applications, would be useful, because the ad-supported monetization mechanism only goes so far and does not reward a number of activities that could generate revenue in this new model; and, on the other hand, the formal subscription model is often too pricey and makes us pay for a lot of data we don't need just to get the one we need. So ironically, such a model may have the effect to reduce the cost of information. And you'd still get a lot for free, like my blog posts, since reputational benefits is enough for a lot of us whose monetization model is to offer consulting services outside the web :)

My guess/hope is that there will be ventures trying to solve that problem, and that one of them will make it very big.

Semantic web is an emerging trend that will accelerate pretty quickly especially now as it is fully embraced by the big guys.

It is also resonating with people as I personally experienced with Raveable.com - semantic analysis for hotel reviews.

Really, Web 3.0 is a convenient bucket to represent the process of building deeper human meaning into the web, and to tag those services that arise to take advantage of such structured datasets in ways that deliver much more substance to end-users.

I don't know what the answers are, but it seems to me that scrolling through a list of search results defeats the purpose/promise of "Web 3.0."

We all know the power of Web 2.0, even for the entrepreneurs themselves in terms of the possible virality of their marketing campaigns and the power to spread quickly.

There is also a great deal of power to be had from moving forward with the idea of Open Linked Data, and things like Microformats, Data Portability, OpenID and OAuth come to mind as ways the new and current entrepreneurs can develop their applications and link their publicly available data on their service.

I will certainly be doing my bit with my upcoming ventures.

The main thing here, is that like people have said, terms like OWL, RDF, SPARQL, etc are very academic and the typical web user will not be able to get on board if we use those terms. Usability is key here: phrasing and presenting these concepts in a way which gives value to the user.

Imagine a Firefox that enables us to "speak" what we want to do: not just goin' to google and type keywords, but saying "hey firefox, search mac news on google". Or stuff like this. Ubiquity is just a starting point, but I think we should evolve Web together with browsers. Not just RDF.

Anyway, I've collected all of my thoughts here, it's a pretty long post :)

http://www.macstories.net/2009/04/29/firefoxnext-an-itunes-like-browser/

In the discussion of standards vs. methods, market needs and user adoption usually win over standards, especially if standards require a big jump from where we are. What we are talking about here is various "knowledge infrastructure tools", and the reality is there will be several paths to get there.

RDFa holds a promise, once properly adopted. But even juxtaposed with existing ontologies, it's not enough, because you have to also provide the proper context for that knowledge, and it takes some additional work to accomplish that. On a sad note, given that webmasters can't even properly publish RSS, 10 years after its birth, RDF/RDFa is a big jump from RSS and we will require a lot of education and tools to make it the assistive technology it aspires to be.

Various players (beyond Google and Delicious) are contributing to systems that enable us to manage knowledge more effectively. Delicious is great for some things, but it does not guarantee data consistency, nor does it structure its data in a way that easily facilitates deep mashups or logical inference as systems such as Freebase do.

The standards and tools under the web 3.0 umbrella help people solve a large class of problems. Of course these tools and standards are not panaceas, but they are very good (widely agreed upon standards), which will continue to evolve over time with global consensus and oversight.

- Organic classification
- Statistical relevance
- Social rankings
…

I have the feeling that RDF, OWL, and SPARQL are just too “academic” and the history of Internet has shown us that the “community” (developers and users) prefers simple non-academic solutions. For example… http://www.delicious.com… offer a clear benefit for the user (save your bookmarks online) and out of that simple egoist benefit TONS of emergent knowledge can be extracted:

- Classification of sites (actually a machine can now understand what a site is about based on the statistic probability of its tags)
- Popular links
- Popular tags
- Treds
- etc etc

I still remember how OpenGIS was the “academic standard” that was going to be the future of online mapping applications… and the BOOM! Google Maps came out and transformed forever the world of online mapping.

So… let’s not forget that all these semantic web standards come from academic origins and not from the natural evolution of the Web. The future will be much simpler… I assure you

An interesting link:
http://lexandera.com/2009/04/crowdsourcing-the-semantic-web/

I think 3.0 will only happen when broadband is capable of sending high quality film without waiting for the content to buffer, and TV / Internet will combine. Will we see less content driven websites and more websites offering just a specialised RSS / XML feed?

Agreeing on a single term and promoting a more or less similar definition makes it easier for people outside of this Web 2.0/3.0 world to understand.

Honestly, I think it is more a "sales" term and a term everyone can understand and talk about. Most importantly, a term business owners can understand and then challenge their team on: "Is our business Web 2.0/3.0"?, even if they don't necessarily understand everything Web 2.0/3.0 implies.

Then, this is where as consultants/advisers we become important: "How can you make my business Web 3.0?"

Final note, even though I don't like the terms, I understand the benefits of gathering around the Web 2.0/3.0 terms. I use the term Web 2.0 everyday, it's a term most people "understand".

Ben.

I am aware there is at least a perceived difference of perspective and culture between the technology and business communities on this, and in fact trashing the terms Web x.0 has become a unifying force in some corners of the tech community (or perhaps more than "corners", I don't really know). But again, I think that blanket terms like web 3.0 are good at taking us beyond the technologies and help us reflect on what it all means for the user. It lets us gather at the web 3.0 conference which also helps break silos and incorporate new elements into the fabric of the semantic web.

Asking us to drop the term in favor of more "precise" definitions, with "precise" often defined as "narrower or more technical" feels to me a little aristocratic. So, while not being in love with the web 3.0 expression - I agree we could try and improve on it for the next version - I do see value in it and understand why RWW embraced it. It's out there and we talk about it, so as tech commenters it's normal that they tackle it.

I understand the human need for formal definitions and everything, but I think there's just one version of the Web, evolving every minute to become smarter and more useful.

Great article though on what's coming and the challenges ahead. I don't have much time at the moment to go through every reference in the article, but I will.

Benoit Tremblay.

A long time ago links became page turning mechanism, rather than a tool for defining associations. Given that the Semantic Web is built to make machines smarter, it will live under the hood a long way from the end users of the web. AI people dreamed this up, so people were a distant thought.

Articulate looks at what is coming, looming, and leaning in to technology is what makes it interesting to be a part of it all!

Like Kingsley Idehen, the Jesus, the Messiah of Linked Data, the lonely sage in the desert (dessert or desert, semantify that with a dereferenceable URI) who campaigns that data sources should be able to auto negotiate their exchange and transformations, according to the best interests of the applications that consume said data.

The stall in uptake can be traced to the opaque specs that the W3C entombed in such a manner as to halt productive development by the rough and ready corps of real, working journeymen engineers who would never be caught dead writing such hilarious claptrap, but yet have to deal with the fallout of the RDF and OWL specs.

Now that the concepts of RDF are not wise, rather it is the the opacity of the transmission of concepts has been the barrier.

Thanks a lot for taking the time to read my posts and write this review, this is the best I've read so far :)

Your comments extend and enrich mine. They also provide a good quick overview of my long posts. I'll write a quick post to refer to this page so my blog visitors can benefit from it.

I especially appreciate your addition on RDFa, which indeed has been gaining widespread acceptance. RDFa does add semantic depth to the data in a simple way. Now that the Big 3 (Google, Yahoo and MS) have somehow blessed the technology, I am hoping to see business models emerge from it soon from innovative start-ups in the space. I sure will be thinking hard about those myself.

Greg

It's amazing how Web 2.0 seemed to just be making it mainstream and attempt to have just reach scalability and yet Web 3.0 appears to be getting close too!

I'll certainly be watching this space.

Thanks again.
Twitter: @ArmandAguillon

I wish it was possible, but I've already come to grips with the idea that even the word "database" is passe. Try to explain "flat-file" to the average user today, and see how far you get.

I'm not saying people are stupid. They are exactly as smart as they've always been, that is self-evident. The Web is now TV and telephones. It isn't computer science. Anything beyond clicking a remote and typing text is not going to gain acceptance. Sad, but true.