1. Scott Rubin on January 26, 2005 12:49 AM writes...

You know, all this talk about ontologies and taxonomies is really deep stuff. But in a more practical line of thought it doesn't really matter which way you do it. Whichever way produces the most useful categorization for the most people is the one that you want to use, regardless of the source. What we really need is a metric for determining the quality of any particular ontology. That way we can more easily compare their quality and select the best ones for our purposes.

2. Lizzie on January 26, 2005 2:47 AM writes...

Speaking of taxonomies, "Borges" is a Spanish name, not a French name, so its orthographic representation can't possiblly include the grave accent that you added above the 'e'.

Additionally, because that name's pronunciation conforms to rules governing stress in Spanish, it is pronounced BOR-ges, which means no acute accent should be present, either.

So,just plain ol' "Borges" will be fine, thanks.

3. Ulises on January 26, 2005 1:56 PM writes...

So not only did BORG-es predict the importance of folksonomies, he also pretty much laid out the structure of a library that would contain all books (sounds familiar?):

http://rehue.csociales.uchile.cl/rehuehome/facultad/publicaciones/autores/borges/borges_n1.htm

Te extrañamos, Jorge Luis!

4. Jay Fienberg on January 26, 2005 5:30 PM writes...

"hierarchic ontologies vs tag sets"

I'm having a lot of trouble with the posts here on Many-2-Many in that technical terms are being used to tag different point views, but those terms aren't always being used consistently. And otherwise, some terms are maybe new and need to be defined more singularly if we're going to be able to understand the differences in these views.

(And, Kevin, I think your phrase that I quoted above is fair on your part--I'm just quoting it because I think that it reflects the term-hinkiness in previous posts.)

So, for example, I didn't see ontologies as having much to do with the debate about tags prior to Clay's self-tagged tirade. Unil then, it seemed sufficient to talk about the relationship between user-tags vocabulary and system-controlled vocabulary without assuming the idea that the vocabulary itself might also be described by some ontology. Oh no, it's not a meta problem, it's now a meta meta problem! (or is that, a meta meta meta problem?)

I think it's tasty to state that tons of controlled vocabularies are not defined by ontologies (i.e., they do not have "formal, explicit specification[s] of a shared conceptualization"). Borges "animals" is actually a good illustration of categories that can be used perfectly well as a (controlled) classification system that nevertheless does not depend on an ontology.

And, of course, controlled vocabularies are very often not hierarchic either, though I think a lot of the concerns have been about hierarchy vs flat (though flat has become a hinky term too, so I don't know any more).

5. Jodi on January 26, 2005 6:40 PM writes...

Conversational metatags or how to talk without any content. I overheard the following conversation (or something like it) yesterday at the gym.

--"class was great, we talked about, like, different definitions of terrorism; and, like, I still think mine was the best"
--"Oh?"
--"yeah, like the defense department has like two definitions and then this other group has like three, but, you know I think mine is still the best."
--"yeah." In my class we talked about nationalism and how there is like more than one way to talk about it.

No one ever asked for, mentioned, clicked on, opened up any content.

6. Kevin Marks on January 28, 2005 7:12 PM writes...

Sorry Lizzie,
I had read a Borges citation in French, where they do accent his name that way, and as my french is far better than my spanish, it stuck in my head that way.