If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

TechCrunch released information today that Yahoo is about to join in the semantic web goodness game. I thought I would quickly write up something from a developers point of view what would come out the other end for users.

I have not got the full list of micro-formats being supported but lets go with the easy ones anyway.

hCard

This can represent people, companies, organizations and places. So for instance if you are a company with an about page (everyone has them) having an hCard for Yahoo to pickup on means they can clearly return that data in a search result and in fact allow for a vast yellow pages style search that only looks at hCard information.

hReview

Names says it all, the format covers any webpage that is reviewing something. It contains most of the simple information that would be useful to a search spider such as the name of the review (stored in hCard format) the rating (1-5) and the description. Again this can easily be seen as allowing Yahoo to leverage this structured data to produce clearer results. At the moment they rely (like everyone) on either relationships with companies via datafeeds, or unstructured data to return any kind of review data (film, restaurant, book) - with hReview they can build up a much more accurate picture.

hCalendar

Yahoo already have upcoming.yahoo.com which is an events site. But if everyone started storing hCalendar data they could aggregate it all into one central location without having everyone having to go along to upcoming (or any other rival site) to re-enter info about any events they are running because the company website would have the hCalendar data embedded and the next time the spider came along Yahoo would add it into its events database.

XFN

It is probably the widest adopted format of the bunch. Already supported by anyone using the wordpress blogging platform (so 10+ million users?) twitter I believe uses it and lots I am probably not aware of. XFN describes any friend relationships described in website links. So a normal HREf link contains a little bit of extra information to say if that link is saying that is a friend, college, etc. Google has already launched Social Graph API that already aggregates this information and allows developers to leverage the data, but I have not as yet seen anything useful developed from it as yet.

What will change

At the moment the search engine writers do a lot very clever stuff to try and understand all the rubbish we put in our unstructured web pages. By adopting micro-formats Yahoo is hoping that website owners will see a benefit in adding more structured data that Yahoo can then feed back to end users to then make the search experience more accurate (and hopefully I have given examples above to back this up.)

The real question is that does Yahoo with is much smaller search market share still have enough clout to convince a large enough percentage of sites with the argument that adding semantic data will then drive them more traffic from Yahoo? Or will they just ignore it because taking the eye off the Google SEO ball would be far more dangerous?

For me as a techie structure data is a very exciting prospect for the future as it removes many barriers that are currently in place for competing with the likes of Google for search. It means building specific spiders/aggregators that work with all this new structured data could mean a revolution in certain sectors (e.g. yellow page sites, review sites)

All in all this is a great move forward for Yahoo and for the web at large. I await the Google reaction.