Comments on: Vander Wal and Categorizing http://csabaveres.net/blog8/2007/05/08/vander-wal-and-categorizing-2/ "Whenever people agree with me I always feel I must be wrong." Oscar Wilde Sun, 29 Jun 2008 13:00:13 +0000 http://wordpress.org/?v=2.8.4 hourly 1 By: Csaba http://csabaveres.net/blog8/2007/05/08/vander-wal-and-categorizing-2/comment-page-1/#comment-21751 Csaba Fri, 11 Jan 2008 23:15:47 +0000 http://csabaveres.net/blog8/?p=33#comment-21751 Yes, again I think we agree. Except I don't accept your claim that it is "not a consistent categorization scheme". I actually wrote a paper about this, which you can see on http://www.springerlink.com/content/m14xuq3593268143/. I am not sure if you can get this paper, but I give a brief overview in my post "Ad hoc categories". The main point is that a cognitive psychologist, Larry Barsalou, actually did try to investigate empirically the same claim, that everyday, ad hoc categories are not consistent. But in fact he found that categories like "ways to escape from the Mafia" and "things to save in a fire" were in fact remarkably stable across time and subjects, and displayed many properties of "normal" categories. (e.g. prototypicality effects). Interesting stuff. Yes, again I think we agree. Except I don’t accept your claim that it is “not a consistent categorization scheme”. I actually wrote a paper about this, which you can see on http://www.springerlink.com/content/m14xuq3593268143/.

I am not sure if you can get this paper, but I give a brief overview in my post “Ad hoc categories”.

The main point is that a cognitive psychologist, Larry Barsalou, actually did try to investigate empirically the same claim, that everyday, ad hoc categories are not consistent. But in fact he found that categories like “ways to escape from the Mafia” and “things to save in a fire” were in fact remarkably stable across time and subjects, and displayed many properties of “normal” categories. (e.g. prototypicality effects). Interesting stuff.

]]> By: phu http://csabaveres.net/blog8/2007/05/08/vander-wal-and-categorizing-2/comment-page-1/#comment-21739 phu Fri, 11 Jan 2008 08:52:12 +0000 http://csabaveres.net/blog8/?p=33#comment-21739 example 2 is the sort of categorization, we do most of the time! it's people's everyday life categorization. it's not a consistent categorization scheme, but most of the time it's sufficient (e.g. for a talk with your neighbour). example 2 is the sort of categorization, we do most of the time! it’s people’s everyday life categorization. it’s not a consistent categorization scheme, but most of the time it’s sufficient (e.g. for a talk with your neighbour).

]]> By: Csaba http://csabaveres.net/blog8/2007/05/08/vander-wal-and-categorizing-2/comment-page-1/#comment-21723 Csaba Thu, 10 Jan 2008 23:17:51 +0000 http://csabaveres.net/blog8/?p=33#comment-21723 Yes, I think I see what you mean, and I agree. If what you mean is this: The human cognitive architecture allows us to form "shared" aggregations of various forms, not just taxonomic categories. Here is an example of two different sorts of aggregations: 1. CAT is a taxonomic category that includes several specializations like SIAMESE, BURMESE, and so on. The category structure and the applicability of the terms to various exemplars is more or less constant between people. 2. HOUSEHOLD PET is a non taxonomic category, but an aggregation of exemplars that can vary between people. A cat is probably one, but is a snake, or a crocodile? Nevertheless, there is still some deep shared understanding that allows everyone to agree that a volcano is not a household pet. So only example 1 is the canonical case of categorization, and we should invent a different name for whatever is going on in 2. Yes, I think I see what you mean, and I agree. If what you mean is this:

The human cognitive architecture allows us to form “shared” aggregations of various forms, not just taxonomic categories. Here is an example of two different sorts of aggregations:
1. CAT is a taxonomic category that includes several specializations like SIAMESE, BURMESE, and so on. The category structure and the applicability of the terms to various exemplars is more or less constant between people.
2. HOUSEHOLD PET is a non taxonomic category, but an aggregation of exemplars that can vary between people. A cat is probably one, but is a snake, or a crocodile? Nevertheless, there is still some deep shared understanding that allows everyone to agree that a volcano is not a household pet.

So only example 1 is the canonical case of categorization, and we should invent a different name for whatever is going on in 2.

]]> By: phu http://csabaveres.net/blog8/2007/05/08/vander-wal-and-categorizing-2/comment-page-1/#comment-21708 phu Thu, 10 Jan 2008 14:15:49 +0000 http://csabaveres.net/blog8/?p=33#comment-21708 this is my understanding of vanderwal's definition: you're connecting two objects by tagging them with the same tag, and a tag is a hook for aggregating these objects. tagging works because of people using the same tag in the same way for the objects, this is not arbitrary but depends on "shared and emergent social structures and behaviors, as well as related conceptual and linguistic structures of the user community" (Marlow et al. 2006, "Position Paper, Tagging, Taxonomy, Flickr, Article, ToRead", p. 1) is there only one right answer to the question "what is tagging?" i think in some (most?) cases people do categorize, others just associate or do something else... this is the reason why one cannot make assertions about the nature of tags or a tag without taking the user(s) into account. this is my understanding of vanderwal’s definition: you’re connecting two objects by tagging them with the same tag, and a tag is a hook for aggregating these objects.
tagging works because of people using the same tag in the same way for the objects, this is not arbitrary but depends on “shared and emergent social structures and
behaviors, as well as related conceptual and linguistic structures
of the user community” (Marlow et al. 2006, “Position Paper, Tagging, Taxonomy, Flickr, Article,
ToRead”, p. 1)
is there only one right answer to the question “what is tagging?” i think in some (most?) cases people do categorize, others just associate or do something else… this is the reason why one cannot make assertions about the nature of tags or a tag without taking the user(s) into account.

]]>