Liveblogging the UNC SSS: David Weinberger and Thomas Vander Wal

David Weinberger, author of Cluetrain Manifesto / “Everything is Miscellaneous” (Times 2007): Folksonomy as symbol.

Folksonomy as political: what right do other people have to tell us how we should think about information? (Obviously expert-based folksonomies have value –> I think he said this b/c there’s a lot of (potentially) indignant mils in the room)

Not just “emergent”, but “ours” b/c emerges from us

Emergence is a fascinating phenomenon because it explains complexity through intrinsic simplicity. E.g., termites build complex towers by following rules so simple that they fit in a termite’s brain. But there is also a political side to our interest in emergence, beyond its explanatory power. Emergence is hope. It says (or we take it as saying) that left to ourselves, without extrinsic structuring or regulation or governance, we will be magnificent. This is beyond the hope implicit in democracy, that says a group will be able to live together if all are given equal power. We won’t just live together, but something far beyond the capabilities of any of us will emerge. Simply by being together, cathedrals will emerge.

I also like his point that folksonomies work best w/ excess: lots and lots of tags, lots and lots of clusters. Author of “ambient findability” (-> a book still on my shelf, unread) worried that there are too many tags, and this will end up in non-findability (bogged down, poor signal to noise ratio), but this hasn’t been held up by actual experience. Folksonomy says the “more the better”.

Rejection of essentialism: if we use a doll as a doorstop, we know we’re not making the “real” use of the doorstop – we know there is a “real” sense. We want to preserve this sense of the real and the absolute. There’s something intuitive about this, but when you take this to essentialism — this is the way it really is — requires “an impossible metaphysics” (not sure what he means by this exactly, as essentialism remains extremely prevalent and the idea of fixity of meaning is very strong– how can it simultaneously be strong and be unenforceable?).

Thomas Vander Wal

Overview of tagging. (Starts out simple and gets much more complex)

What’s a tag:
Simple data/metadata externally applied to an object
“Taps into the existing cognitive process without adding much cognitive cost”. – Rashmi Sinha (rashmisinha.com)
Tagging for one’s own recall – this is how we think ourselves so it adds our explicit understanding which meshes with everyone’s implicit understanding.

What’s folksonomy:
(funny anecdote about people citing Wikipedia but not citing the version of the Wikipedia page)
Personal free tagging of pages and objects for one’s own retrieval
Usually done in a social environment shared and open to others
* Gmail: people can pull your tags from email? I didn’t know that!
Active tagging is done by people consuming the information, not the content author

Value
Comes from people using their /own/ vocabulary that adds some explicit meaning, which may come from inferred understanding of the information/object
Not categorizing as much as providing a means of connecting items (placing hooks) and providing their meaning in their own undertanding
Not so much findability as refindability (awesome!)
Putting hands on your own information (“I think I saw it on a blog somewhere.. or maybe someone told me over email..”). Whether or not the information originated from them means nothing- it’s all their information.

Folksonomy Triad

Object / Identity / Metadata (he drew a complex diagram here that I can’t reproduce in ASCII easily)

the link between “object” and “identity” is interest
link between “identity” and “metadata” is vocabulary
link between “community” and “metadata” is terminology (and so forth)

Folksonomy vs. taxonomy

Both have terms
Taxonomy is business-driven (efficiency, absolute)

Folksonomy is “customer” (I’d prefer a less commodified term) driven

But: business has to pay attention to customer terminology
And customers need to understand business terminology

// This reminds me of when my BF and I were trying to hang a picture last week and neither of us knew how to work with the sawtooth hangers. The first thing we had to do was to figure out what they were called before we could google how to use them correctly. So our first few google searches were just trying to zero in on taxonomy.

Not a taxonomy VS a folksonomy but BOTH – product exists in the middle
(Again, enterprise focus — we could equally say that an object, article, etc. exists in the middle.)
Businesses don’t have to spend anything on the folksonomy : if they view an emergent term from the users, how do they integrate it in to their official, “paid” taxonomy.

Business tensions:
Naming control < --- > people’s vocabulary
(Reminds me of problems using Kleenex, Xerox)
Sample group <-----> everybody (potentially)
In-house tagging <----> outside tagging services
(What are people calling our products vs. what do we call our products)
$$$ w/ value <----> Unknown value
Consistency <----> Emergent

Business gains
* Improved understanding of customers
* Current terminology – all of it
(Note: what about data that could never be used for business terms? Ever? Like totally un-monetizable?)
(GREAT example of the tags on the Kevin Federline album – they say something far beyond anything about his music. I especially like “aka vogon poetry.”)

Who is tagging?
1/2 of 1% of people on the web, coming from Nielsen (3-5 million people tagging, including flickr – 30-40% of photos on Flickr are tagged, I note: these are probably the same people who upload a ton of is he conflating taggers with users?)

Reasons People Tag
- People have to have their own intrinsic value for tags – del.icio.us lesson (** biggest reason **)
- Add perspective/context
(baby socks used for ipod nano covers- “ipod” tag would make little sense to most people)
- – Missing metadata
- – Emergent vocabulary
- – Personal descriptors

Refindability (** big one ** solving the technical pain of finding information)
- – Aggregation of informatoin
- – Task-based aggregation

“Every tag is sacred” – every person is an expert in their own vocabulary or tag, and if you remove a tag you begin breaking a person’s ability to find the information again
(but what if people are tagging things in order to be political? Like tagging a feminist author as “man-hating bitch”? I guess we can’t assume people will see information just in nice, good ways; maybe that person is writing an article on man-hating bitches and needs to find that author again)

A few sites I hadn’t heard of before
Geo-location: platial
Mobile: socialite
Intranet – IBM’s Dogear, Scuttle

“whereever there is a digital object or marker you can add a tag to it”
(In an ideal future world?)

He really likes RawSugar – I’ve never used it so I know nothing about it

He advocates that we should try to get people to understand the value of tagging for non-taggers
(show them a video (dubious that this would work), express the technology pain problem)

Understanding the “volatility of the tagged object”
- If you’re interested in an object, knowing when that object has been changed is useful
- When it’s been updated, removed, etc.