Team Murder No Brain No Headache.

29Sep/06Off

The Human Interpretation Of Search Sucks

I saved Danny Sullivan's post on the suckiness of search interface in an offline reader and promptly forgot about it for a week or so. This may sound like more of my milking a post from something I forgot to do methodology but there is actually a point buried in here somewhere. The point is about interface and by this I don't mean the more commonly and narrowly defined concept of how buttons are arranged in an application panes and in which order they appear. What I'm thinking of has more to do with the much maligned metaphor interaction. Neal Stephenson gave us a good deconstruction of why this is conceptually flawed in In The Beginning There Was The Command Line and compared the commercial operating system interfaces to the faux-thenticity that Disney employs in makes us suspend disbelief in order to participate in the illusion as an actual participant being emotionally drawn into the display instead of a simple observer apt to notice inaccuracy. The devil is really in the details when you need to just perceive the veneer as a zeitgeist entity that offers an illusion of choice in a world of constraints determined entirely by the environment. There is no way you're going to be truly convinced you're tromping around in the jungle if you simply circulate through the props on a cart that moves along a track. Ideally Disney might force you to travel the same route through the fake scenery but it will also maintain the illusion that you've either made a choice or reacted to another piece of the scenery. I guess that is the complicated part about interacting with metaphors as reality.

What I started thinking about was the metaphor of searching the web and now many different areas of enquiry are folder into that term. In many ways 'search' is a pretty simplified term that is close to the use of 'internet' to mean the large blue E on the desktop or just the web. Defining hugely expansive categories of both objects and sources with generic terms always steers like the Titanic into semantic icebergs that gains its massive size by alternating between the lexical and conceptual. This is where the interface to 'search' becomes really important. If I am running a query in a database of peer edited academic journals I really shouldn't be surprised by the returned results. I went into the text field with an expectation and the end result was within the constraints of my expectations. The academic database carries a whole lot of semantic weight with it and is usually more formally organized than say a Wikipedia article on the same general topic. It draws from a limited number of known quantities and makes for a whole lot less time spent thumbing through dusty indexes. You know what to expect.

The same is true of traditional search interfaces: you slap some words in a textfield and then press the appropriate go button and your results are returned in a hierarchical list, sorted by invisible algorithms that you're accustomed to. If you've come up using Yahoo as your search weapon of choice then you've come to expect the first set of results to be sponsored crap and know to skip those results unless you're in the buying state of mind. It works the same for Google or whatever as well. If you've come to depend on that format and accepted its limitations (advertising more or less) then you actually get what you're looking for. I think that is the reason why so many of the ultra-deluxe interface search engines haven't been successful. The initial presentation doesn't matter a whole lot as complexity or a whole lot of pulleys and levers that don't have an obvious use or impact on the submission of search terms. It's more crap to navigate through if you're unfamiliar with the interface and more presentation layer between user and results. Extended features are better presented as a part of the search and not another knob that will likely be ignored. CSS-fu is probably wasted on the search interface.

This all comes back to the reason that this article sat unread in my offline reader for so many days. I consciously went through the effort to archive this article with the intent of giving it a more thorough read while riding the bus to work and promptly forgot about it. I had a version open in an actual browser and accidentally closed the tab so in my head that article was gone and would have to be looked up again when I had a live network again. Why didn't I realize this? A good portion is probably attributable to it being before 7 am but again the interface that I'm used to didn't have the content so I just assumed that I wouldn't be able to access the content. Another item to add to the infinite to do list I guess.

Comments (0) Trackbacks (0)

Sorry, the comment form is closed at this time.

Trackbacks are disabled.