Semantic Library

Several blogs have posted about Web 3.0 recently, most trying to come up with a central set of ideas about what it might be. For some, Web 3.0 = Semantic Web, for others, Semantic Web is just a part of it. My take on it that I wrote in October last year, if there is such a thing as Web 3.0, can be found on the About page of this blog:

• Semantic web: True write once, publish many: hamstrung until now by proprietary software, proliferation of XML schemas, and a lack of end-user tools.
• Metadata: Meaning and context within and between objects, new languages.
• Rich open data: Geotagging, eScience data for everyone
• Content anywhere, especially mobile
• Make your own software: bringing software and tool creation to the masses
• Two opposing ends: on demand anywhere (video, TV, radio, text), lightweight flexible architecture

Politics and governance issues will continue to evolve to bring;

• Ubiquitous Open Access
• Access to Knowledge (A2K) in the developing world

I think it is important to keep in mind the political and governance issues surrounding the web. The technical part of Web 3.0 is not possible without supportive research, funding, and policies. Additionally, if Web 3.0 impacts people’s lives by making communication and managing information easier, it has to include all kinds of knowledge (social, government, entertainment, scholarly) and be accessible by people all over the world regardless of language, socioeconomic and geographical barriers.

While I enjoyed all the good things the supposed Web 2.0 movement offered - community, interaction, etc - I am hoping that Web 3.0 will pay attention to the difficult issues around data, scholarly communication, and dissemination of research. Linked data is an area of research on this topic, but I hope to see more standards, policies and funding in this area.

A study of all the different ideas people have about Web 3.0 was posted by Jonas Bolinder, and fell into four categories -

• Semantic Web
• APIs and Web Services
• Mobile Web and other devices
• Implicit Web (personalisation and recommendation)

There’s a little bit of each of these in my view of Web 3.0

A post on Read/Write Web, Web 3.0 Through the Ages, sums up some of the current thinking around the term, valid or not and concludes -

“…the discussions we have about defining the next web help to solidify our vision of where we’re going — and you can’t get there until you decide where you want to go. “

I agree, and I’m interested to see where the discussions lead next.

Other recent posts on Web 3.0:

• It’s all still alpha in my eyes, Liako.biz
• After Web 2.0: WOW (Wide Open Web) - enough of version numbers for the web!!, Trends in the Living Networks

WikiProfessional

Via the RIN blog, Jan Velterop recently of Springer is heading to KnewCo, which is behind WikiProfessional. WikiProfessional is designed for researchers and scientists to interact using a database and wiki with a Semantic Web foundation -

The core functionality of this workspace is a newly invented “Semantic-Wiki”. The Semantic-Wiki connects the new relational MediaWiki software (Wiki editing capability in a relational database structure) to the computational text analysis technology, called the Knowlet™

They’re starting out with a large dataset from Medline, which is a great idea to avoid the problem of having no examples to work in new websites.

To get started, you create a desktop, where you can view Knowlet and Expertise workspaces. This is really getting to the trust aspect of the Semantic Web. One thing that is slightly disconcerting is being confronted with another self in the expertise space - my name is a fairly common one, and the system tries to find me in Medline.

If I was one of the authors listed, I could select articles and identify them with me. Other traditional databases have tried to do this, such as CSA’s Scholar’s Universe. Now if WikiProfessional makes this type of thing portable and linkable to other IDs, such as OpenID, then I think it would have a lot of value.

No doubt there will be many more features added to WikiProfessional as it makes the move out of Alpha, so it’s worth keeping an eye on.

Calais

From Reuters, Calais is all about making data richer by semantically tagging it. There’s a bit of questioning as to what’s in it for Reuters, but I can imagine that for journalists it would make sorting and identifying important news stories and other content easier.

If you want to try it for yourself and skip the API signup check out Calais Text Tagger. Feed it some text and it will give you tagged RDF in return. I fed it the last post on the blog, and got back:

Continent: Africa
IndustryTerm: so many services, web projects, social web, energy
Company: CNN, Google
Person: Ethan Zuckerman
Technology: mobile phones, PDF, SMS