Re: [unrev-II] DKR-IR : Towards Topological Directories ?

From: N. C a r r o l l (
Date: Mon Feb 12 2001 - 21:03:06 PST

  • Next message: Gil Regev: "RE: [unrev-II] A very timely paper"

    From: Bernard Vatant <>
    To: <>; Nicholas Carroll

    > I had a look at your overview at
    > I've some comments about what concerns directories, hierachies, etc.
    > As a matter of fact, I've drifted during last year from activist Open
    > Directory editor to Topic Map true believer :))
    > with a due passage through a non-hierachical crisis last summer. See
    > Now I've come to a more balanced viewpoint, see

    I'm pleased you have recovered from the mal au foi.


    > 1. Most topics are multidimensional and are in fact the intersection of
    > several hierarchical threads.


    > 2. The hierarchical relationships - or associations, to use the Topic Map
    > terminology - are but one type of associations any Topic may be involved
    > him.
    > We have a strong basis to consider Topological Directories, with no "top
    > level", but a "semantic core" of highly connected topics, where browsing
    > would use both accurate hierarchical links, and all sort of directions
    > transversal to this hierarchy.
    > It looks like the general structure of such a "topological directory" will
    > be, if not a formal Topic Map, at least something very Topic Map like.

    I am not sure the eyes are able to fully understand complex topic maps.
    The body is much better at understanding three dimensional structures.

    > My hunch is the next generation web directories will be built on such a
    > structure. I agree that the management and updating of such directories
    > will need a wise blend of google-like algorithmic and ODP-like distributed
    > human expertise. There is a great waste of human energy in ODP now due to
    > the very incapacity of the mono-hierarchical structure to manage
    > complexity, both at content and social model levels, leading to permanent
    > power games and ontological unstability

    The blend of algo/human I use is not particularly sophisticated; if there
    is any explanation for it working fairly well, it is that I had substantial
    experience in massaging databases years ago, and I hate sys admin work,
    which leads me to put smart people in control of data as soon as
    possible. But right now there is still a certain amount of batch work
    being done. Yuck. For the *next* generation, I hope the human
    expertise is being used to play the database as a conductor leads
    an orchestra.


    Nicholas Carroll

    ------------------------ Yahoo! Groups Sponsor ---------------------~-~> eGroups is now Yahoo! Groups Click here for more details ---------------------------------------------------------------------_->

    Community email addresses: Post message: Subscribe: Unsubscribe: List owner:

    Shortcut URL to this page:

    This archive was generated by hypermail 2b29 : Mon Feb 12 2001 - 21:28:42 PST