CHI workshop: faceted metadata for IA and search

flexibility, previews, meaningful organization, support for expansion and refinement
none of these speak to the conception of location
talks about the difficulty of classifying in more than one location, and offers the faceted approach as a solution to this problem.
rather than finding the place for items (the right category), facets allow us to attach descriptors to the items where they are.
cooking method: stirfry, ingredient:chicken, course:main dish…
this is pretty much just like what i was calling genres, eh?
facets make refinement simpler from a UI perspective – what is relevant is determined on demand.
issues?
handling facet similarity/synonymity
adding facets is done separately from assignment to items?
makes me think about how my idea of merging annotation and search parallels the wiki movement’s build as you go philosophy. there are strengths and weaknesses to this

claims that it could make automatic categorization easier, so i’m not the only one wondering about that…
“themes” as a name for groups of facets, ways of capturing larger ideas like “japanese internment during wwii”
facets are shown independently in the interface, attributes still exist, but are associated with individual items.
labels are the names used within the facets, not the facets themselves.
this is a lot of names for things…
the ebay example she gave for attributes is shipping rate, tax, etc. couldn’t this be grouped into a facet? right, but then all of those things wouldn’t be shown on the main page. but, in my view, they Would be accessible in a detailed view of the facet itself, not just on the screens of the detail results that contain the attributes.

the lists of facets in the nobel prize example still feel overwhelming to me.
there is a lot of learning required to understand to affiliate the colors of each faceted section with the meaning behind it.
so what?
well, it also takes up a lot of screen space.
and for the halo to be powerful in a wide range of contexts, i am pretty sure it’s going to have to fade into the background pretty well.
how should i test that?

a few years since the last usability study?!!
fairly or unfairly, this triggers my wariness of the academy. i am more and more convinced that i need somewhere that encourages me to make shit fast and put it in front of people immediately. i already naturally resist this out of shyness, or perfectionism, depending how you look at it.

she talks a lot about the goal of helping people not feel lost, but none about the sense of location. the main wayfinding aid seems to be breadcrumbs. and the general sense of relevance that comes from seeing the multiple facets. and yeah, that’s huge, but is something lost from not making the new location metaphor a bit more explicit? if not, maybe i should give that up. but i should think about it.

all shared facets of a selected facet are already “or”s – allowing people to start with the most obvious first step by making the process of choosing subsequent steps less overwhelming. this is a good point.

key points i see to make/clarify in incorporating these thoughts into my capstone:
help finding “nearby” items
location as constellation
merging annotation, search, and exploration
facets and genres
spatial consistency for key facets? iconic representation to speed absorption at a glance?
and/or/not filtering easier if we think about the filter as a place?
facets/genres support hierarchy And multiple categorization. this is huge.
ebay could provide interesting examples.

Leave a Reply