BBC Sport - Sport The BBC have been at the forefront of the real application of Linked Data techniques and technologies for some time.  It has been great to see them evolve from early experiments by BBC Backstage working with Talis to publish music and programmes data as RDF –  to see what would happen.

Their Wildlife Finder that drives the stunning BBC Nature site has been at the centre of many of my presentations promoting Linked Data over the last couple of years.  It not only looks great, but it also demonstrates wonderfully the follow-your-nose navigation around a site that naturally occurs if you let the underlying data model show you the way.

The BBC team have been evolving their approach to delivering agile, effective, websites in an efficient way by building on Linked Data foundations sector by sector – wildlife, news, music, World Cup 2010, and now in readiness for London 2012 – the whole sport experience.  Since the launch a few days ago, the main comment seems to be that it is ‘very yellow’, which it is.  Not much reference to the innovative approach under the hood – as it should be.  If you can see the technology, you have got it wrong.

In an interesting post on the launch Ben Gallop shares some history about the site and background on the new version. With a site which gets around 15 million unique visitors a week they have a huge online audience to serve. Cait O’Riodan in a more technical post talks about the efficiency gains of taking the semantic web technologies approach:

Doing more with less
One of the reasons why we are able to cover such a wide range of sports is that we have invested in technology which allows our journalists to spend more time creating great content and less time managing that content.

In the past when a journalist wrote a story they would have to place that story on every relevant section of the website.

A story about Arsenal playing Manchester United, for example, would have to be placed manually on the home page, the Football page, the premier league page, the Arsenal page and the Manchester United page – a very time consuming and labour intensive process.

Now the journalists tell the system what the story is about and that story is automatically placed on all the relevant parts of the site.

We are using semantic web technologies to do this, an exciting evolution of a project begun with the Vancouver Winter Games and extended with the BBC’s 2010 World Cup website. It will really come into its own during the Olympics this summer.

It is that automatic placement, and linking, of stories that leads to the natural follow-your-nose navigation around the site.  If previous incarnations of the BBC using this approach are anything to go by, there will also be SEO benefits as well – as I have discussed previously.

BBC Sport Data Model The data model used under the hood of the Sports site is based upon the Sport Ontology openly published by them.  Check out the vocabulary diagram to see how they have mapped out and modelled the elements of a sporting event, competition, and associated broadcast elements.  A great piece of work from the BBC teams.

In addition to the visual, navigation and efficiency benefits this launch highlights, it also settles the concerns that Linked Data / Semantic Web technologies can not perform.  This site is supporting 15 million unique visitors a week and will probably be supporting a heck of a lot more during the Olympics.  That is real web scale!

Comment   or   Contact us

RDF MagnifyIt is well known, the business of SEO is all about influencing SERPs, or is it?  Let me open up those acronyms:

Those engaged in the business of Search Engine Optimisation (SEO) focus much of their efforts on influencing Search Engine Result Pages (SERP), or more specifically the relevance and representation of their targeted items upon those pages.  As many a guide to SEO will tell you, some of this is simple – understanding the basics of how search engines operate, or even just purchasing the right advertising links on the SERP.  Quite simple in objective, but in reality an art form that attracts high rewards for those that are successful at it.

So if you want to promote links on search engine pages to your products, why would you be interested in Linked Data?  Well there are a couple of impacts that Linked Data, and RDF its data format, can have that are well worth looking into.

Delivering the Links – the BBC Wildlife Finder site is an excellent example of the delivering the links effect.

The BBC started with the data describing their video and audio clips and relating them to the animals they portray.  What was innovative in their approach was that they then linked to other information resources on the web, as against creating a catalogue of all that information in a database of their own.  This they encoded using Linked Data techniques, using RDF and a basic Wildlife Ontology that Talis consultants helped them develop and publish.   The stunningly visual website was then built on top of that RDF data, providing an intuitive navigational experience for users, delivering the follow-your-nose capability [that characterise Linked Data backed websites] to naturally move your focus between, animals, species, habitats, behaviours and xthe animals that relate to them.  Each of these pages having its own permanent web address (URI).  In a second innovative step they provided links to those external resources (eg. Wikipeadia – via dbpeadia, Animal Diversity Web, ARKive) on their pages to enable you to explore further.  duck pagecurlIn yet another innovation, they make that RDF data openly and easily available for each of the main pages.  (Checkout the source of the page you get when you add .rdf to the end of the URL for an animal page – not pretty, but machines love it)

So a stunning Linked Data backed site, with intuitive follow-your-nose internal navigation and links to external sites, but how is this good for SEO?  Because it behaves like a good website should.  The logical internal interlinks between pages, with a good URI structure that are not hidden in the depths of an obscure hierarchy, coupled with links out to to relevant, well respected [in SEO terms] pages is just what search engines look for.  The results are self evident – search for Lions, Badgers, Mallard Duck and many other animals on your favourite search engine and you will find BBC Nature appearing high in the results set.

Featured Entries – Getting your entry on the first SERP a user sees is obviously the prime objective of SEO, however making it stand out from the other entries on that page is an obvious secondary one.  The fact that ebay charges more for listing enhancements indicates there is value in listing promotion.

RDF, in the form of RDFa, and Linked Data become important in the field of Search Engine Results Promotion (another use of SERP) courtesy of something called Rich Snippets supported by Google, Microsoft, and Yahoo.  From Google:

Google tries to present users with the most useful and informative search results. The more information a search result snippet can provide, the easier it is for users to decide whether that page is relevant to their search. With rich snippets, webmasters with sites containing structured content—such as review sites or business listings—can label their content to make it clear that each labeled piece of text represents a certain type of data: for example, a restaurant name, an address, or a rating.

Encoding structured information about your product, review or business in [the html embeddable version of RDF] RDFa gives the search engine more information to display, that it otherwise would not be able to reliably infer by analysing the text on the page.   Take a look at these results for an item of furniture – see how the result with the reviews, from, stands out:


Elements such as pricing, availability, are also presented if you encode them in to your page.  I would be leading you astray if I gave you the impression that RDFa was the only way of encoding such information within your html.  Microformats, and Microdata now being boosted by the initiative, are other ways of encoding structured information on to your pages that the engines will recognise.

One of the major benefits of using RDFa is that it can encode the links to other sources, that is the heart of Linked Data principles and thus describe the relationships between things.  It is early days with these technologies & initiatives.  The search engine providers are still exploring the best way to exploit structured information embedded in and/or linked to from a page.   The question is do you just take RDFa as a new way of embedding information in to a page for the search engines to pick up, or do you delve further in to the technology and see it as public visibility of an even more beneficial infrastructure for your data.

At Talis we know the power of Linked Data and it’s ability to both liberate and draw in value to your data.  We have experience with it [in SEO terms] delivering the links and have an understanding of its potential for link featuring.

I would love to explore this further with folks from the world of SEO & SERP.  I also work alongside a team eager to investigate the possibilities with innovative organisations wanting to learn from the experience of the BBC, Best Buy, Sears and other first movers, and take things further.  If you fit either of those profiles, or just want to talk through what I have described, I encourage you to drop me an email or comment on this post.  There is much more to this than is currently being exploited and to answer the question in the title of this post – yes, those interested in SEO should be focusing in on Linked Data.

This post was also published on the Talis Consulting Blog
Comment   or   Contact us