Epistemographer

Still more panel notes:

• Jeffrey Veen (Adaptive Path)


• Tantek Celik


• Don Turnbull (Asst Prof, UT Austin)


• Thomas Vander Wal

– Jeffrey Veen [presentation] –

Solipsism = “There’s nobody but me”

• Metaphysical assumption that the self is the only definite thing in the universe


• Look at filesystem interface: structured around the self’s computer


• del.icio.us: tagging is about replicating filesystem conventions w/r/t online bookmarks


• friendster: fundamentally not about socialization, but about amassing a self-oriented list of “People I know” (like a high-school yearbook)

Handful of examples of “leveraged solipsism” (Connection of personal information with other people’s personal information)

• del.icio.us: other people’s tags


• Amazon recommendations


• flickr: open standards, extensible

– Thomas VanderWal [presentation] –

• Personal View: “We each have windows out onto the world, and they shape the way we see it.”


• “We don’t control what happens outside, but we shape the room behind it; we each have our own organizational system.”


• “On the web, we have a great view through the window, but really poor personal organization systems in our rooms.”


• Information shapes us


• One problem with information ecologies: we get lost early (no persistent trails, no convenient trail markers, difficult “refindability”)


• “That syncing feeling” – have to keep various applications/devices/information storage systems synchronized


• The “personal InfoCloud”: “the rough cloud of information that follows us as we go from place to place, this cloud keeps all the information the person wants to be kept nearby.” [cite from his site]


• Building a personal infoCloud:


• Portibility/ubiquity; access; personally organized


• External storage: flickr, del.icio.us, WebDav, Personal Portals, e-mail, attention.xml


• Personal storage: PDA, laptop, iPod, desktop, keydrive


• Standards: open APIs, standard connectivity => interoperability => personal control

– Don Turnbull [presentation] –

Uses of Tagging:

• Collaborative filtering systems


• Recommender systems


• Information Filtering


• Search system augmentation


• Focus on the user’s perspective rather than the system

Folksonomy issues:

• How do you get people to cooperate?


• How good can tags be? (Find things you’d never find, categories you’d never think of)


• Volume of recommendations vs. number of recommendable items


• How accurate can the recommendations be


• What about changing interests?


• Web is a shared information space without much sharing

Tagging issues:

• Tag spamming and gaming


• Tags are explicit


• Tags are text and can be analyzed (feature extraction)

Tag Properties:

• Power Law distribution


• Popular Tag terms


• Prolific taggers (expertise)

Social issues:

• Who controls the sharing?


• Who controls the controls (ontology)?


• “Give to get” systems


• Anonymity vs. community (community of “friends” vs. people as “data points”)


• Free ridership


• Personalization vs. community


• It may be more interesting to find a like mind than a resource recommendation

Use(ful) Metadata

• Implicit tagging vs. explicit tagging (Amazon purchase vs. flickr tag, for example)


• Read wear, clicks, dwell time, chatter (analyze unintentional personal traces)

Don’t fence me in

• Tag mobility


• Common tag API


• Desktop vs. server

– Tantek Çelik [presentation] –

Technorati tags

• rel=“tag”


• ultra-fast development turnaround for plugins/etc.


• step toward interoperable tagging systems

– Q&A –

Here’s the question I asked Veen via Rendezvous:

I’m wondering: it seems like we’re talking about two things – truly solipsistic data (amazon purchase records, traces of web viewing) and data generated with an understanding of the broader social landscape (flickr, del.icio.us tags, etc). In the case of the latter, you’re not so much leveraging solipsism as using people’s ostensibly solipsistic actions which are really intended to create a public identity. For example, a blogroll might look very different from a record of all blogs actually read; I’d love to hear the panel’s thoughts on whether the conflation of these two kinds of data is a problem, and how (if at all) we might solve it…

Seems like a crucial point, which didn’t come through when he raised the subject; a lesson from Goffman is that we’re perpetually constructing our identities through perceptible actions, which raises big flags when you’re talking about a system that might leverage our truly solipsistic actions (i.e. the one we do without thinking of anyone else watching). It’s at core a question of public and private, and one which is totally muddied in the (albeit really rich and valuable) discussion here.

This entry was posted on Sunday, March 13th, 2005 at 3:53 pm and is filed under SXSW. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.