Yahoo Answers and similar “question and answer” sites have been getting a lot of publicity lately.  While I’m not too keen on the current incarnation of these sites—they seem like nothing more than glorified database applications to me—it strikes me that this is the PERFECT application for social networks and a bastardized version of the semantic Web using microformats.  (Oh and by the way, anyone from the semantic Web crowd who reads this, I'd love to see if there's a way to make RDF work for this as well, I seem to have trouble wrapping my head around the practical applications of it so any help is appreciated!)

First a word on social networks:  iIf you read my weblog regularly you’re probably aware that I’m not a fan of the closed, proprietary social networks that are in use now.  Instead, I believe that social networks should be a loosely-coupled, organic collection of individual and business Web sites that form a social network.  The missing piece, I think is the semantic Web part--a data layer that can extract structured data from the member sites.

What does this have to do with answering questions?  Well, it occurs to me that what I really want to do when I ask a question of the Internet is pretty much along the lines of what Yahoo Answers is talking about doing, that is, see if anyone in my network has answered a question.  Essentially filter the Internet down to your trusted sources and look within that network of people for your answers.  Be your own filter.

The problem with the Yahoo solution is that it’s a closed system.  In order to ask or answer questions, you have to use Yahoo’s system, and you’re locked in.  And anyone who is providing an answer also has to use the system, which means that the probability you're going to find an answers dwindles because of the closed nature of the system.  There really isn’t a “network”, per se, except for the one that exists in Yahoo’s database.  The ideal solution to this problem, in my mind, is simply to create a network of trusted sites, a la Google Custom Search, and use a Microformat parser along with an extended set of Microformats to extract “answers” and “questions” from the sites.  All you’d really need is a set of microformats to delineate  “question”, “opinion”, “fact” and so on.

The only technology that would really be necessary to make this work is to embed microformats in site text itself.  I’m really not sure why this hasn’t taken off yet, it seems like a no-brainer to me.  What I’m talking about, and I’ve actually posted some working examples of this before, is to surround chunks of text from a weblog post or text published to a public site with microformat markup so that it can be extracted as meaningful data:

<span class="vopinion"><span class="question">What is the shape of the earth?</span><span class="answer">The earth is a flat octagon, I tell you, I've seen a corner!</span></span>

It really amounts to nothing more than coming up with some standardized CSS classes that can be used to mark up text with some more meaningful data.  I’ve really only seen microformats used as standalone information islands up to this point, such as embedding hidden contact information in an about page, which seems a crying shame and a waste of opportunity to me.  If people were to start marking up the text they publish with semantic data so that the underlying meaning could be extracted and searched, we would no longer be depedent on Google to extract that meaning for us.  Instead of asking Google for an answer, I could just ask the sites in my social network for an answer.

I guess I just don't see why we can't pick the low-hanging fruit and mark up the text we all write with richer data.  Entering things in database is all fine and dandy, but wouldn't it be better if the databases automatically extracted their data from the freeform text we all write on our sites and weblogs?