[Sigia-l] mixing apples and oranges and tomatoes

[Katherine Lumb]

This is *exactly* why I cling to a tiny shred of resistance to my client's argument. I don't want to bully users with a forcibly rigid taxonomy... but I don't want things to spiral out of control, so the user becomes unable to build a conceptual model of the site over time.

One respondent claimed that few users will ever spend enough time at the site to be troubled by the redundancy. If I can bear that out empirically, then I think I can let go. But until then, I'm not sure which is the greater good.

Yes, you can use a faceted taxonomy to help a user find a wine according to how she classifies it... but I'm trying (I think) to sort out something a little more specific. On wine.com, if a user is specifically *browsing by type* of wine, she will find the varietal "White Merlot" under "Pinks" and never under "Whites." Sure, I want to set up my taxonomy so if my user was searching for tomatoes, she may be able to search by color, or by shape, or another classification and find it... But I'm speaking specifically about browsing a hierarchy of categories. Does that make a difference? Or am I stumbling over the same conceptual block?

K

On Thu, 11 Apr 2002 13:47:16 -0400 (EDT) Tanya Rabourn wrote:

> However, isn't it at odds with this bit of research:
> "Toward Usable Browse Hierarchies for the Web"
> http://www.microsoft.com/usability/UEPostings/HCI-kirstenrisden.doc
> 
> It would also seem to me that good navigation is based on learnable
> categories. If users encounter redundancy wouldn't that confuse them and
> hinder their attempts to figure out what features make something a member
> of one category and then where to look for the item that they want? (Keep
> in mind I'm not talking about some sort of forced learning of a taxonomy,
> but the sort that we naturally do everyday when meeting new concepts and
> sorting them into our own taxonomy of knowledge.)

Hanna, Holly wrote:
> True.  For example, in everyone's favorite example of a faceted thesaurus,
> evineyards.com (formerly wine.com), you've got Ravenswood Zinfandel under
> type/red wines/zinfandel, as well as under region/Californian/Sonoma Valley
> and under winery/R/.  To my mind, this doesn't confuse the user, since
> they're going to find what they're looking for regardless of how they go
> about doing it (and they'll do it without having to deal with
> cross-references).  It's the beauty of a faceted taxonomy.

But the way I understand faceted taxonomy, you still have mutually
exclusive categories within a single facet. The temptation seems to be to
place an item under more than one category within a single facet.

It would be like putting a single wine under more than one region. For
example, suppose there is a particular wine that's actually from New
Zealand, you have some reason to think that someone might mistakenly think
it's from Australia so you elect to put it under both.

-Tanya
___________________________________
Tanya Rabourn <rabourn at columbia.edu>
[User Services Consultant]
AcIS R & D <www.columbia.edu/acis/rad>
tel: 212.854.0295



_______________________________________________
Sigia-l mailing list
Sigia-l at asis.org
http://mail.asis.org/mailman/listinfo/sigia-l