January 31st, 2007

Husserl and tagging

A very nice paper on the “laissez-faire librarianship” often associated with tagging vs. more structured semantic web approaches. Most notable is that the discussion is put in the context of Husserl’s theory of reflections, intentionality and intersubjectivity.

A PHENOMENOLOGICAL FRAMEWORK FOR THE RELATIONSHIP BETWEEN THE
SEMANTIC WEB AND USER-CENTERED TAGGING SYSTEMS

D. Grant Campbell
Faculty of Information and Media Studies
University of Western Ontario
London, Ontario N6A 5B7, Canada

Abstract

This paper uses Husserl’s theory of phenomenology to provide a model for the relationship
between user-centered tagging systems, such as del.icio.us, and the more highly structured
systems of the Semantic Web. Using three aspects of phenomenological theory—the movement
of the mind out towards an entity and then back in an act of reflection, multiplicities within unity,
and the sharing of intentionalities within a community—the discussion suggests that both tagging
systems and the Semantic Web foster an intersubjective domain for the sharing and use of
information resources. The Semantic Web, however, resembles traditional library systems, in
that it relies for this intersubjective domain on the conscious implementation of domain-centered
standards which are then encoded for machine processing, while tagging systems work on
implied principles of emergence.

4 Responses to 'Husserl and tagging'

Subscribe to comments with RSS

  1. J-Wicz
    May 2nd, 2007 at 11:55 pm

    Nice paper. For my the technology potentials are missing. But essentially I just started to read Foucault recently and thought that his image of “suppressed knowledge” fits nicely with the struggle between tagging and SemWeb, and at the end gives this healthy struggle a meaning. I like the last paragraph from this research paper where the social tagging is taking care that the ivory tower ontologies are still connected to reality and might require adjustments from time to time. Still I too believe that it will be the SemWeb which will play a very significant role in one or two years time. Systems like delicious though will be the incubators for taxonomies (and maybe later ontologies) which have no academic or defence industry lobby.

  2. Mo
    May 3rd, 2007 at 10:59 am

    Hi, I think there is a very interesting discussion going on what the two approaches can learn from each other. It is a pity that the semantic web is often perceived as a very institutionalized, normative approach taking away the freedom of the individual — and you hit the nail on the head with attributing that to the academic or industry background. It just hasn’t “made it” to the normal persons. Two more papers I found interesting with reference to this point:

    http://www2002.org/CDROM/alternate/744/
    “This paper analyzes important uses of meta-data in the e-learning domain, from a pedagogical and philosophical point of view, and abstracts from them a set of fundamental architectural requirements for Semantic Web meta-data. It also describes some flexible generic techniques for working with meta-data, following these requirements.”

    http://www.iasummit.org/2006/files/164_Presentation_Desc.pdf
    “This paper uses theories of complexity, pace layering and resilience to address the question of tagging and folksonomies, and their influence on the practice of information architecture. “

  3. J-Wicz
    May 8th, 2007 at 9:15 pm

    I had no time yet to read the first paper, but just read through “From Pace Layering to Resilience Theory”. Some thoughts on that:

    Generally I like the statement, that information architects must seriously rethink their position. I like the slightly provocative quote: “The popularity and supposed benefits of tagging in systems like Flickr and del.icio.us suggests that these value-added features are unnecessary, and so, by extension, are information architects.”

    But there is a basic misconception in this paper. Beside of the public libraries and big governmental institutions most information architects are actually employed by corporations. You don’t need an information architect when you tag your photos at Flickr, because you are your own. But the second you try to convince your mates to use the same Flickr tags from your common holiday trip you are putting on the boring/administrative hat of the information architect. The corporation is just the extrapolation of that. It has so many tags (formerly known as keywords), that they can essentially employ someone just to think about “tagging patterns”, that will lead most likely to “lists”, “thesauri”, “taxonomies” and maybe on day to “ontologies”. The corporations sees itself as a legal persons (similar to Hobbes Leviathan projected onto a company, or http://www.thecorporation.com/ ). It wants to speak a common language and consistent messaging. Thats the main purpose of marketing departments and corporate identity bodies. Also internal communication suffers big times due to inconsistency in data and knowledge misinterpretation. So the freshly hired information architect will try to sort things out. He will most like start giving the whole thing hierarchical structures, which is a fair approach, because most companies are based on hierarchical thinking patterns anyway. But I agree that he will most likely miss the knowledge of the actual users of the applications which they COULD contribute in some way. Tagging would be a very valid approach.

    Short sidetrack: and no, search engines simply don’t work effectively on corporate intranets. Not Lucene and most likely not todays Google algorithms. But the enterprise produces a lot of content and needs to put somewhere, so either the c-drive (and it is lost after the next crash) or the intranet (and it is lost once it is orphaned and no intranet page points to it).

    So lets assume the approach outlined in above paper works and successful fast tagging feeds good reviewed enterprise taxonomies, we will run in a nasty technology problem: versioning of hierarchical structures. In current ECM packages it is so painful and you’ll end up with orphaned content, which you can’t risk. So the attitude of never “touch a running system” starts to creep into the information architecture. But it MUST evolve because market reality forces companies to do so. “This fits with paradigms of complexity theory, in which complex systems, if unable to evolve in response to external stimuli, merely freeze in place or disintegrate (Waldrop 1992).”

    OWL 1.0 has an version tag in its specification and still I personally don’t know a working and reliable approach in the semantic web world how to tackle this. The DOME project http://dome.sourceforge.net/news.shtml has little activity to what I can judge. The promising Jena stuff is alpha stage. Currently the FZI in Karlsruhe does have some active research on this and I’m curios on their results with SOBOLEO http://www.fzi.de/KCMS/kcms_file.php?action=link&id=678 they will present (actually this week? at WWW2007). And http://www.im-wissensnetz.de/ looks promising to. But (and maybe that is the reason why I visit blogs like yours) I don’t think that traditional simple text rendering will be able to visualise these concepts.

  4. Moritz Stefaner
    May 10th, 2007 at 2:41 am

    Wow — thanks a lot for your excellent comments.