Re: ACTION-406: Petname burden

So, is this something you can do? What are the next steps?





From:
Maritza Johnson <maritzaj@cs.columbia.edu>
To:
Mary Ellen Zurko <Mary_Ellen_Zurko@notesdev.ibm.com>
Cc:
Timothy Hahn <Timothy_Hahn%IBMUS%LOTUS@notesdev.ibm.com>, W3 Work Group 
<public-wsc-wg@w3.org>, "Tyler J. Close" <tyler.close@hp.com>
Date:
04/02/2008 10:24 AM
Subject:
Re: ACTION-406: Petname burden



here's one suggestion on a UT experiment. 

Should the UT be designed to quantify the cognitive burden of Petname or 
other aspects too?


Though I'm not sure this one can be reswizzled easily into something 
that's an easy lo fi prototype to run. Could it be something like: Explain 
the petname concept to people (learning about it is not the question). 
Show them some example sites, and ask them what petname they would choose. 
(Maybe get one site from them they regularly go to? Maybe use a large 
selection and select the subset they either regularly visit or at least 
know of?) Get back with them after some period of time. Show them some 
sites, some with pet names, some without. Ask them if it all seems right 
to them. I'm fudging this last bit a little.

Expanding on your ideas. We could have something like:


First session:

1) Ask the participant to list the sites they go to on a regular basis. We 
can offer a sample list to prod them in the right direction, something 
like: webmail, blogs, news sites, my bank, my credit card, my school/work 
website.

2) Explain the petname concept

3) Ask them to select the sites from their list they would assign a 
petname to. This will give us an idea of whether they understand the 
purpose of petname (of course taking into account that they were _just_ 
given instructions). 

Either on paper or with a quick functional prototype of petname (maybe 
something like what tlr suggested in another email thread with the ff3 
bookmark interface) ask them to assign petnames to each of the ones they 
chose. If they indicated they would use petname for a site that wouldn't 
need it we can ask more questions for iteration purposes.

If their initial list is representative of the sites they really interact 
with this step will give us a good idea of how many petnames a user will 
have to remember, then we might be able to talk more precisely about 
whether the cognitive load is appropriate or not.


Session 2: (I'm also less certain what to do in this part)

Based on their previously selected petnames we can show them sites with 
and without petnames, or their sites with the a petname similar to but 
different than the one they chose (which might not be fair since they 
haven't interacted with petnames in between sessions, but we could also 
have the time between sessions just be a couple days).

Have a post-study questionnaire asking how they felt about using petnames 
...


thoughts?

-- Maritza