I was asked about non-linear search and said I’d give it a go.
(The question comes from Simon Collister, who I am sure has a few ideas of his own up his sleeve. But since he wrote a fab post about tagging, which in turn fuelled my own effort on the subject, it’s definitely my turn to go first!)
Non-linear search is one of the bounties of the web 2.0 approach that has been relatively unheralded because of the noise surrounding the ongoing “digg/wikipedia/myspace/youtube is heaven/hell” wrangles.
The expression comes from a post on the BusinessWeek blog. Interviewing Joshua Schachter from del.icio.us, Heather Green notes:
Joshua made an interesting distinction. Instead of finding information a la Google, social search is about finding knowledge. The idea is how do you connect with the information you need in a context that’s knitted together by people and by human expertise, rather than the linear way we do it now, which is to type a search term into a box.
So what do we understand by linear and non-linear search in this context?
Linear search: You already know the answer in many senses. Or you know exactly what you’re looking for and there is a finite answer. What is the capital of Uzbekistan? (“capital city Uzbekistan”). Even quite tricky facts are accessible. Room rates at the Hilton Hotel in Paris? (hilton paris “Rates” -“Paris Hilton”)
A lot of other searches work less well, though. “Good hotel London” brings up a load of sites that want to sell me rooms. What I really wanted, though, was a recommendation of a good hotel in London that meets my criteria — pretty central, cheap, not horrible. How am I going to find that on Google? How do I find “romantic restaurant cheap rome”, “what are the best blogs about supermarket wine” and “inexpensive ways to make your wife feel special”?
So how can I do better?
Non-linear search: The same search terms about the cheap, central-London hotel on del.icio.us produced this:
A site with customer reviews, a Google site that might help me judge their worth, and a specific recommendation. So which was more useful? (rhetorical)
Back to the idea of knowledge. A lot of the questions we really want to ask are the questions that we’d ask another human being, if we knew a person with the right qualifications for the job. Finding the name ‘Tashkent’ is one thing, but what is it like there? Where should I stay? What things should I see? Is there anything or anywhere I should avoid?
I’ve got no idea how to find the answers to those questions using Google. Simply entering the country’s name into del.icio.us provides me with a lot of better ideas:
These sites provide me with some contacts. Some real people. If I were going there, I’d ask the authors of ProgressoR and Registan, or find a message board there. And I’d probably ask Craig Murray (the UK’s ambassador to the country, as it happens), since he’s a fellow brit. I somehow think they might be more use for my questions than the CIA or Jimmy Wales, the top links on a Google search.
The Yahoo Answers site provides another solution to this quandary, though obviously, you’d need to rely on a resident of Uzbekistan (or wherever, or whatever) being a willing and able answerer. I haven’t tried it, but I’m not sure I fancy my chances on Uzbekistan nightlife. Depends what you want to know — they would probably do better on London hotels.
So how do tags help? Tags are a way of finding people who speak the same language as you. “Good cheap hotel London” means something real to those people — the people who use my language know what I mean to a far greater extent than Google ever could. When I search on tags rather than keywords, there is a better chance of finding them.
It all depends what you want from the Internet. A really big library/encyclopedia — the way they used to describe the web in the 90s? Fine — I think we’re probably already there. But I also think we can possibly expect a bit more. Immediate contact, perhaps, with specialists in whatever field you’re researching. Access to knowledge as well as facts.
People, networks, tags, wisdom — these things, for me, are 2.0. Forget the buzz startups and believe the promise.
Share this post:
So prompt and so rewarding!
That’s a really great post, Ian. a) it answers my question completely (until then I only had half the answers to the whole question!) and b) it cuts through some of the web 2.0 hype and guff out there at the moment. And flattery will get you everwhere!
cheers
You have been very kind about this, but really, it’s just stabs in the dark.
But nobody else can do any more, so sod it. Roll on!
Great post!
I’d just like to add that del.icio.us and fellows can solve another problem which text search engines are really crappy at: Semantics. Try searching for an article on how to use the “this” keyword in JavaScript — I did once, and it was close to futile.
[…] Sometimes it can be really great to stumble across a blog and learn something new. I did just that this evening when I ended up reading Ian Delaney’s explanation of non-linear search. […]
[…] What is Non-Linear Search? at twopointouch: web 2.0, blogs and social media Back to the idea of knowledge. A lot of the questions we really want to ask are the questions that we’d ask another human being, if we knew a person with the right qualifications for the job. Finding the name ‘Tashkent’ is one thing, but what is i (tags: no_tag) […]