[Sigia-l] card sorting: dealing with multiple placements

[Jonathan Broad]

Derek R wrote:

>This course is about "LIS research" Michael ("questions in library and
>information science"), and has very little relevance to performing
>user-tests and/or conducting user-research (beyond glossing over some
>techniques).
>  
>
This seems to me a typical course in social science research methods.  
Of course it wouldn't include user-tests, as it's not a design class.  
I'd like to point out to you though, pointedly, that HCI/IS (as in the 
IS in LIS) was applying social science methods to information system 
design questions long before *ethnographers* realized their methods 
could be used to test *web sites*. 

Your comments valorize "field research" over the full panoply of 
user-research methods that derive more generically from qualitative 
research techniques.  LIS has drawn on these techniques since before 
personal computers were around.  Sure, OPACs and online database 
interfaces have been horrible since day one, but that doesn't mean that 
librarians haven't been trying to fix them, using user-centered design 
methods, since day two.  Does that story sound familiar?  My point isn't 
that librarians have some ancient wisdom here, but at the same time we 
know how to suck eggs, if you catch my colloquialism.

While I'm writing this, let me correct two other mistakes you seem to be 
making.  One, you should realize that all the classifaction "systems" 
that seem to really bother your postmodern sensibilities _were never 
intended for consumption by the end-user_.  They were originally 
designed by librarians, for librarians--for librarians who were to spend 
considerable time mastering them.  How and why these mechanisms were 
slowly exposed to the end-user is a historical question that's not 
pertinent to understanding their original design purpose.  The point is 
that librarians themselves were supposed to be the interfaces between 
information and users.  Classification systems are the paper-and-meat 
equivalents of content management metadata schemes, and should be 
compared on that basis.

Which leads me to point two: far from *imposing* any of that 
library-junk on users, even a semi-competent reference librarian is the 
very epitome of a user-centered interface.  We should completely isolate 
the user from the actual organization of the collection. 

We (and I was *only* semi-competent) are trained to listen very 
carefully, elicit clarifying information, and then *iteratively* and 
*interactively* search for satisfactory material using our knowledge of 
the aforementioned classification systems, our knowledge of available 
resources, etc.  This aspect of librarianship is the most closly allied 
with the more broad purpose of the 'user experience design' moniker.  
The equivalent of the reference interview is still sadly missing from 
many online user experiences.

>This MLIS degree course has, as its primary objective, to offer
>instruction toward the end of creating and understanding *the
>deliverable* -- systematizing data (statistics). This is consistent with
>where I previously indicated the LIS-IA skill level is at.
>
No, that statement is consistent only with your level of understanding 
of where LIS/IA is at.  As one of the aforementioned benighted LIS/IAs, 
let me weep publicly that you heap such scorn on us.  Did a librarian 
once break your heart, or steal your lunch money?  I for one have gladly 
added ethnomethodology to my toolset of "means" to "the end" of a good 
user experience.  (That, incidentally, is what a *conciliatory* gesture 
from a librarian looks like. :) )

Jonathan Broad