[Sigia-l] Are web users more sophisticated

[Karl Fast]

> >Is that so wrong?
> 
> The point is not whether it's wrong or not. The point that it's an
> indication that web users are confused. 

That is one interpretation. And I think it accounts for *some* of
this behaviour.

But that's a lot of people making a lot of mistakes. I doubt the
overall "error rate" is that high.

As noted in my other post, analyses of large query logs have found
this pattern repeatedly. But this pattern hasn't shown up in
observational studies (not that I'm aware of; I'd love to hear about
any studies where this has been observed). I suspect this is
probably because even well-designed studies are always slightly
artificial.

As I noted in my other post, Andrei Broder of IBM research has
suggested that people are using Web search engines in ways that our
classic models of IR systems don't really account for.

  Broder, A. (2002). A taxonomy of Web search. SIGIR Forum, 36(2). 
     http://www.acm.org/sigir/forum/F2002/broder.pdf


Broder points out that a basic assumption of IR systems is that
people use them to fulfill an "information need." Not so he argues.
He suggests that on the Web, people use search engines for three
types of tasks:

1. Navigational. The goal is to reach a particular site/page.
2. Informational. The goal is to fulfill an information need.
3. Transactional. The goal is to perform a "transactional activity"
   like buying a book or ordering movie tickets.

That's his "taxonomy of Web search." The classic model accounts only
for #2. But types #1 and #3 help to explain (in part) why people
would type URLs into search engines.


OTOH, I should note that the only time I type a URL into a search
engine it's a mistake on my part. I always type my URLs into the
location bar (I am constantly typing CTRL-L). If Broder is right
then I'm one of the anomolies. That's fine by me. Life is pretty
great out here on the lunatic fringe.


--karl
http://www.livingskies.com/