Search is more than Just Google
To analyze any service or product requires a step by step procedure from understanding the usage pattern to suggesting design directions. So what is the usage pattern of searching.
One thing we know is that we dun wake up one morning and start searching. That's fully faltoo and could form an edge case. For most people the need to search arises during inadequacy of present information. We need something all of a sudden and want to quench our thirst for the desired piece of info asap.
Let me jump to slightly different thread of thought. Technology, technology everywhere but not that all of us use it. Technology is still accessible to the geeks. Noone other than the geeks can just pick a technology and start using it. For the layman its still mumbo jumbo and they look for a nearby geek to break through it. Well, my point is search is something like that still.
Google search is excellent. It given me results instantly and is more relavent than other searches still. Google is become synonymous to searching. But still i find myself struggling trying to explain to the computer what is in my mind when i am searching for something. I change the words put then in a different order and use all my wits trying to understand what the computer can understand. And then luckily i get the results than i want.
The problem is that I need to talk the technology language for the search to work properly. To my relief one of my colleagues gave a shortcuts to search effectively in google. But then I am not a geek and do not have the patience to go through Help Files.
Search results in google is still a collectivist approach. It shows results based on popularity. It gives you results based on what most people mean when they type a certain word in the search box. However, this works when most people think alike. But most people do not think alike. Word associations differ with education, culture, context and past google experience. There are patterns of how people think.
Search is yet to be personalized to the way I think. Information should appear based on my past google experience. What is it that i ultimately select when i type certain keywords should somehow be taken into account. Based on the data accumulated for me, it should then predict what I mean when I type a certain keyword. Google should make use of the search data it accumulated to come out with persona types and searching patterns in order to predict search results effectively. It will know what I ultimately select when I stop searching. This is because I search only when information is inadequate. Once adequate I stop searching and continue with what I was doing.
This not only helps narrow down my search results but also reduced search results considerably. It saves time and effort. It makes search faster and more efficient. It also does not make me learn to think like a geek.
When i search for something, i am ultimately searching in a limited domain (like usability, hyderabad, graphic design, films, music, some more...) which it will recognize over a period of time. In a typical case (85% of all cases) I will search in one of these only. My word sematics is going to depend on my domain. When I mean User Interface it will be screens of appliances and not UML diagrams in most cases (95% of all cases). When I type potato recipe it will depend on my location and in a typical case will be an Indian recipe (85% of all cases) . When I type graphic design it means that i want to know more about it and not hire a graphic designer (95% of the cases).
So what I am saying is that in order to design a certain system you need to design it for a typical context of use who are trying to accomplish a typical goal. Untill then information foraging will be a geekish thing to do.
