Really interesting to see how ideas develop. Tony Hirst posted to my blog post about librarians as publishers, linking back to something that he wrote last year: "Could librarians be influential friends?" This in turn is something that I talked about at a CILIP Sussex AGM last week. I started to read Tony's post, then decided that I'd stop and write my own, then go back and read his to see where we chimed. So I'd suggest that you read both posts if you have the time to do so.
Search engines have one holy grail, and that is to get the right information to the right person at the right time. The algorythms that they use are closely guarded secrets because that's the one thing that really makes them different from the rest. (That and what they index of course, but that's an aside as far as this post is concerned.) All of the major search engines are trying to emphasis the social and personal nature of the results that you receive, and social results have become an integral part of the SERPS (search engine results page). One only has to look at Google for this: a search for librarians brings back a wikipedia result, videos from YouTube, images for librarians, MySpace, Twitter, blogs, and 'Results from people in your social circle'. Then we have the adverts. Google attempts to work out who you are, what you're interested in, and what adverts you click on in order to give you ones that you're more likely to click on. Then we also have the Realtime option, Top Links, Blogs; both related about librarians and blog posts themselves. Google is now toying with the +1 option, and I've previously ranted (errr, written) about that, along with their concept of a social network.
With Bing social we have something similar. 'Shared links for librarian', public updates for librarian from Facebook and Twitter, Trending topics; all of this information comes from social data, and not from websites. Then we've got Social Mention which in some respects goes even further, as you'd expect a dedicated social media engine to do; sentiment, top keywords, top users, hastags and sources. My final example is Blekko, which allows users to link to their Facebook account, so that results can include the material that friends haved 'liked'.
Then we've got Facebook itself, with people, groups, communities, the like option, search function and so on. Slowly but surely it's moving itself into a position where it can act as a fully fledged search engine - companies are already finding that they get more hits on their Facebook pages than on their websites and a question now starting to be asked is 'Is Facebook killing off the company website?'
So I think it's fair to go down the assumption route that social in search is becoming more important, day by day. It's not going to become less important, that's for sure. So I think it's important that we start to see how the information professional can leverage their knowledge and experience in this area. Librarians are generally well regarded and trusted - I made this point in my post on librarians as publishers and although I'm preaching to the choir at this point, no-one has disagreed with me. So let's jump ahead a few years and see how search could be affected by individual reputation.
I use 'a search engine' - it might be Google, Facebook or something entirely new in order to do my search. The engine checks out Twitter to see who is talking about that - looks at who talks most about it, how many followers they have, how often they are retweeted, what their reach is and so on. (The Wefollow directory already has a listing of 'influential' people in each category, so this is going to be a given.) So we then end up with 100 people for example. What do they link to? If a lot of them link to a few resources that half a dozen of that group have referenced, those results get a high ranking. The engine can then look at something like Flickr to see if they post photographs, and if so, how often they're viewed and added to others favourites. (Working on the idea that an expert is going to take photographs of the thing they like, and yes there are limitations to that concept.) The search engine can then bounce off to Delicious and Diigo and other bookmarking services. Two algorythms can be used here - how often is the expert's content bookmarked, and the size of their network. Next, let's bounce off to Slideshare because if they're an expert and thought leader, they'll have posted material for conferences etc. The same for Youtube, various Podcast resources, radio stations, reviews of/by them in Amazon.
Our search engine will be doing two things, just to be crystal clear - it'll be finding the experts and thought leaders in an area in order to see what they are recommending, and the extent to which their own information is being cited by other people. Between those two criteria, and based on identifying say 100 experts I think you're going to get a pretty good set of results. Of course, this can then be monitored further once the results have been displayed. Which resources does the searcher actually click on, and then cite or recommend to their own networks? The balance can easily be twisted around - if someone wants a search for librarian images, Flickr and similar resources become more important, and if another searcher needs academic information, the engine can utilise a whole bunch of different resources and pay particular emphasis to .ac.uk or .edu sites. Oh yes, and we must not forget the actual websites either - how important are they, traditional links, updates and so on.
It's worth stressing that *none of this* is difficult for a search engine to do. What does make it particularly interesting though is that the balance of power really starts to shift. Twitter (or some other microblogging service) holds a huge amount of content - they can easily allow access or block access to different search engines however and whenever they want. Facebook has a huge amount of information on its users and what interests them. We are then moving away from the concept of results based on global popularity (which is broadly the way that Google works) down to personal popularity. Any search engine that is able to tie up these elements together is going to be extraordinarily powerful. Google should be worried - very worried, because so far they have totally failed to understand social as the failures of Wave, Buzz and so on demonstrate. Social is Google's Achilles heel.
OK, let's swing this back to librarians. There's no reason why our new social search engine cannot use as an algorythm occupation. After all, if it can identify that someone is an expert in physics and pay really close attention to what they're doing if should be able to identify a librarian without problems. The more that librarians do - NOW - with social media, the more that we're going to already be embedded into the social medium. The more contacts, friends, links, tweets, photographs, likes, +1's that we have, the more influential we can become. The more influential we are, the more people will link to what we're doing, the more we'll be working in networks of influence and the more useful we can be to people.
In many respects though, nothing has changed. People come to the library because they want to talk to the librarian. If a librarian tells them what resources they need, they'll use them. So all that is happening is that our sphere of influence moves from face to face or organisational level to a much wider, indeed global level. However, and this is a big however, if we don't embrace these resources now, we're going to find it much more difficult to leverage ourselves into that influential role in the future. The more that we can do now, the easier it's going to be in the future. So involvement in things such as Facebook, Twitter and the rest of them isn't a desirable, it's not a fun thing, it's not a trivial thing, it's an absolute necessity.
OK, off to see what Tony wrote now. :)