Archive of UserLand's first discussion group, started October 5, 1998.
Re: Too many editors
Author: David Carter-Tod Posted: 10/29/1999; 10:16:11 AM Topic: SalonHerringWiredFool.Com Msg #: 12515 (In response to 12505) Prev/Next: 12514 / 12516
Instead, he tried to get into the head of someone he doesn't know, the mythological random reader of this site.
I read this as an attempt at informal scenario-based design which is a technique with a solid pedigree in HCI (http://www.acm.org/pubs/citations/journals/tois/1992-10-2/p181-carroll/), and which I have some background in (http://ijhcs.open.ac.uk/erskine/erskine.html).
Our research suggests that the "mythological random reader" isn't a useful place to start, rather you need to start with identified (potential) users and preferably ask them to construct scenarios of how they would ideally use a site to accomplish goals. Alternatively developers themselves can take on an identified persona and write scenarios that way.
The next step is to make claims about things identified in the scenario. This should usually be a formal process (i.e. write scenario, make claims - not intermix the two), but people do it informally all the time, and most web designers have some kind of internal narrative going on in their head like this a lot of the time. The technique formalizes that process.
Actually, I'm considering creating a mainresponder app that does this.
There's a Netscape developer article that talks about scenarios, but I can't find it right now -- wasn't that good anyway since it still insisted on task and content analysis, which is unnecessary if you do your claims analysis correctly.
Here's an IBM article that has a lot in common with our article (so much so, we wondered where they got some of it from).
This page was archived on 6/13/2001; 4:53:16 PM.
© Copyright 1998-2001 UserLand Software, Inc.