1. Lee Bryant on July 20, 2005 12:10 PM writes...

Hi David,

One question: why 'war'? Strange language to use in this context surely?

Good to meet you again at Reboot - hope we catch up soon.

Lee

2. David Weinberger on July 20, 2005 1:41 PM writes...

Only because "culture war" is a cliche, and you know the rule: Make sure your blog title contains a play on a cliche.

That aside, I agree with you. It's a dumb title.

3. Emil Sotirov on July 20, 2005 2:08 PM writes...

I agree that, in fact, it is not so presumptuous to go as far as speaking of "cultural difference" or even "war" when considering the humble usage of tags.

"Folder" = "one-to-many" = "world as hierarchy" = "fixed point of view" = "authorship" = "exclusion" = "closed"...

"Keyword" = "many-to-many" = "world as network" = "changing points of view" = "con-versation" = "participation" = "open"...

4. Tom Coates on July 20, 2005 2:24 PM writes...

I'd push that one a little though - the stuff that Joshua has tried to do with del.icio.us is basically post-folder, but not quite keyword - it's a view of tags as a way in which something can belong to multiple folders without there being any cognitive dissonance. I think my point was that people were pulling that more towards a keyword style use though. One consequence of this might be that his interfaces aren't built for lots of tags, unlike the Flickr interface which conceals the less popular ones and lets recurrence form emergent patterns.

5. Michal Migurski on July 20, 2005 6:08 PM writes...

Emil... a little heavy on the loaded imagery, no? "(many/one)-to-many" I agree with. Tom's analysis is spot-on, though I could do without the implied competition. I think that keywords & tags need a context (folder) to make sense, e.g. Del.icio.us tags live in the "/us/icio/del/" universe, while Flickr tags live in the "/com/flickr/" universe. Mixing them rarely makes sense.

6. Emil Sotirov on July 20, 2005 9:15 PM writes...

Michal... you're right. I realized quickly (after posting) that my statement came out a bit heavier than what I was trying to convey...

Back to the topic... As I understand it, Tom is using "folder" and "keyword" as "legacy" metaphors - to say that people oscillate between the two as modes of understanding and using tags... with "keyword" becoming predominant with the time. Seemingly, given the freedom of folksonomy, people tend to move from hierarchical "folder" modes of tag interpretation (one-to-many) towards more open "keyword" modes (many-to-many).

I think that these notions - of "folder", "keyword", and "tag" - will all eventually become "legacy" in a new culture of simple linking of terms... simple "pointing to" - without instrumentalizing one of the terms as "defining" the other. Remember the favorite "mode" of tagging in teenage land... "like"... "Like" is simple linking... nothing more... no meta-data... a thing is just stuff... like other stuff. "Blog" is a blog... not a tag. See here... a few items... they are "like"... uuh.. blogs. Blogs are "like"... uuh... web sites.

7. Paul B Hartzog on July 20, 2005 10:33 PM writes...

sounds "like" my blog entry on information tyranny and embedded ontology:

"Being a reluctant revolutionary, I am always wary of tyranny and the many insidious forms that it takes. Here's one that really gets me.

The "folder" system. Otherwise termed "hierarchical" information systems.

Filesystems are hierarchical. Operating systems are hierarchical. Scientism is hierarchical. Modernism is hierarchical.

The information systems of the world are infected with an embedded ontology.

And I intend to do something about it..."

http://blog4zog.blogspot.com/2005/02/information-tyranny.html

;-)

8. Paul B Hartzog on July 20, 2005 10:36 PM writes...

In fact, if you are interested, I have saved some links on "Relational Programming" on del.icio.us about this topic:

http://del.icio.us/PaulBHartzog/RelationalProgramming

9. David Gilbert on July 21, 2005 2:55 AM writes...

If I see a picture on flickr of a redheaded woman with the tag "red," this may be a "description," but we can go further--we might notice that it's a specific sort of description known as a "synecdoche," in which a part (the red color of her hair) stands in for the whole (her hair, the woman herself, the photograph itself?). Or what about a picture of a girl tagged with "angel"? This might be understood as a "metaphor," the comparison of one thing to another. Further, imagine a picture of a girl with her finger up her nose tagged with "angel"--this is likely to be an example of "irony."

We see all these things in flickr, et al., simile, irony, metaphor, metonyme, synecdoche, and these relations are not properly heirarchical or generically descriptive but RHETORICAL. Some tags may even imply arguments. If I see a picture of a well known politician tagged with "evil" is this fundamentally an act of categorization or description? I'd argue it's better understood as a "claim," perhaps even containing an implied "argument," a persuasive appeal.

I would argue that, more or less, what we are now calling "folksonomies" are what the Greeks called "rhetoric." And if we begin to try to categorize or systematize all these usages, these manifold relations, tropes, figures, then we run into the same problem as Aristotle and later Cicero and the Romans, who never exhausted all of the categories needed to explain the various uses of language. Which leads me to conclude with a bad paraphrase of Wittgenstein: "The meaning of a tag is in its use."

10. Phil on July 21, 2005 6:46 AM writes...

I don't really understand the keyword/folder distinction, and I suspect that the emergence of tag clouds may
turn it into a difference that makes no difference. (There was a longer version of this comment, but it got far too long for this box - and my blog was looking undernourished - so I put it up here.)

David G. - very interesting. I think you're a bit dismissive of logic, which I think does have a place within the folksonomy/ethnoclassification debate: it would be good to be able to associate tags explicitly - "A equates to B", "X is a type of Y" - as well as through propinquity. But it's certainly worth stressing that rhetoric will never be entirely absent, and that rhetoric's truth-claims are of a different order from (and cannot be reduced to) those made by logic.

11. jim wilde on July 21, 2005 5:43 PM writes...

aboutnees tagging; the who, what, where and how to solve complex problems.

tried to post url, but it got rejeted

on my blog, complexity_theory_meets_ i named a bug

12. David Gilbert on July 21, 2005 6:53 PM writes...

Phil:

You say, "rhetoric will never entirely be absent." I think you've got it the wrong way around. I'd say rather that the act of tagging is fundamentally rhetorical, and that "logic will never be entirely absent." Let's begin with a non-contraversial function of tagging: It's a mnemonic device; I tag something in such a way so that I can recall it later, so that I can find it again. It may be that logical relations are useful in this regard, but mnemonics was historically a canon of rhetoric.

Now let's consider the history of folksonomies. The "topoi" used by early Greek rhetors were reference works that contained all figures of language one might want to use arranged in such a way that they could be searched handily. Through the 19th century people kept "commonplace books" into which they would copy and paste snippets of information, quotes and ideas they collected on various topics. Topoi and commonplace books were useful primarily as resources for thinking and for writing--their organization schemes were subservient to their purpose as aids for the production of new material, not to any ontological project per se.

As to the social aspect of tagging (especially prominent in flickr), the moment we begin to talk about how someone tags something for someone else, we are in the middle of issues of "audience." In other words, right in the middle of what has historically been the province of rhetoric. Logic, on the other hand, has always been indifferent to issues of audience, if not downright hostile towards them.

Now the formal "taxonomy" is the logician's dream. A group of experts, who knows the TRUE ONTOLOGY OF THINGS, gets to work out the relations of things and categories, without having to consider what the "crowd" believes, which is, after all, mere "opinion." If you want an historical perspective on the logician's estimation of the folksonomy, take a look at Plato's Phaedrus or Republic. The moment we begin to take the crowd seriously, we are in the middle of rhetoric, right along with Gorgias, Protagoras, and Aristotle.

In his work titled The Rhetoric, Aristotle explains that what makes a rhetorical argument different from a logical one (in his terms, what makes an enthymeme different from a syllogism), is that the premises used in rhetorical arguments are drawn from phronesis--that is, from practical wisdom or common knowledge about the world that has been accumulated, not by the logician on his mountaintop or in his cave, but by the rabble in the assembly and in the agora; if you like, in the bazaar.

13. stowe boyd on July 22, 2005 11:33 AM writes...

Somewhere in the Coate's thread, the term annotation was used. Do we really use tags as annotation? For example, I have never encountered the use of tags for ratings: "4 out of 5", "A+", or the like.