Thoughts, links, and essays from Peter Merholz
petermescellany   petermemes


Archives before June 13, 2001

RSS Feed

Adaptive Path (my company!)

About peterme

Most of the Time
Oakland, CA

American history around the time of the Revolution, figuring out how to marry top-down task-based information architecture processes with bottom-up document-based ones, finding a good dentist in San Francisco Oakland
Designing the user experience (interaction design, information architecture, user research, etc.), cognitive science, ice cream, films and film theory, girls, commuter bicycling, coffee, travel, theoretical physics for laypeople, single malt scotch, fresh salmon nigiri, hanging out, comics formalism, applied complexity theory, Krispy Kreme donuts.

Click to see where I wander.

Wish list
Show me you love me by
buying me things.

Track updates of this page with Spyonit. Clickee here.

[Editor's note: began as a site of self-published essays, a la Stating The Obvious. This evolved (or devolved) towards link lists and shorter thoughtpieces. These essays are getting a tad old, but have some good ideas.]
Reader Favorites
Interface Design Recommended Reading List
Whose "My" Is It Anyway?
Frames: Information Vs. Application

Interface Design
Web Development
Movie Reviews

Smart Mobs mini-blog: Context-Aware Computing: The Return of Ranganathan? Posted on 11/01/2002.

Among the more exciting developments with mobile technology is context-aware computing. Our devices will know where we are, and be able to augment our experience in that area in various ways--the ability to write and read notes that others have placed; read the history of the spot you're standing on; find out about any activities of interest occurring nearby, etc. etc.

The most common criticism of this future is how we're going to be bombarded by unwanted messages, how we'll be notified of specials, discounts, coupons, promotions for merchants near us, that kind of thing.

While a legitimate complaint, I see that as a minor one compared to what I fear will be the greatest problem--making sense of it all. Such locale-augmentation is meant to assist people, revealing useful information about where you are so that you can know better what to do. However, the amount of "augmentation" of any particular spot can be quite high, and I can't imagine anything more frustrating than fiddling with a device, trying to make sense of your surroundings, and being offered interfaces that are either useless (think of bad search engines) or poorly suited to the actual task (think of any email program).

And I can pretty much guarantee that frustration will be the norm if we develop context-aware computing in the same fashion as we have most of our information-rich internet technologies. This is because, historically, the people building such devices tend to be technology-focused (look at what we can do! the more we offer, the better!), and not user-focused (what on earth would people want to do? how would they go about doing that?).

Additionally, they've also not been meaningfully information focused. They'll understand that in order to bring "value" they'll need "content" that "enhances the experience." So they'll buy databases chock full of stuff, and shovel it in, and, in doing so, make every context-aware computing task an arduous information retrieval task.

This is what lead me to think of S.R. Ranganathan. Considered the father of Indian library science, he developed a system called Colon Classification, which categorized all of reality into five primary facets:

  • Personality—what the object is primarily “about.” This is considered the “main facet.”
  • Matter—the material of the object
  • Energy—the processes or activities that take place in relation to the object
  • Space—where the object happens or exists
  • Time—when the object occurs
    (definitions borrowed from Ranganathan for IAs)

    Perhaps as a testament to Ranganathan's visionary status, faceted classification is extremely difficult to practice in 'the real world,' but is remarkably well-suited to the digital world. In the last couple of years, web designers have increasingly turned to facets (knowingly or not) to organize unwieldy collections of information.

    Even so, his original PMEST classification has not really taken root (to the best of my knowledge). I suspect, however, that we increasingly take our computers with us, we'll find that we're finally ready to take full advantage of Ranganathan's original vision. I can only hope that the device developers understand the importance of smartly classifying the world around us, and incorporate wise information architecture into their products.

    4 comments so far. Add a comment.

    Previous entry: "Smart Mobs mini-blog: Free-Riding in the Commons"
    Next entry: "A must-read blog..."


    COMMENT #1
    Hey, Peter. I agree with your point that sense-making is where devices can fail to make information usable. The point in offering a way of describing knowledge -- which is what Ranganathan's and almost any classification system offers -- is to represent upon retrieval the knowledge contained in a thing so that a person can hold or view it, grok it, and decide whether it satisfies some need. In the classic IR model that breaks down into:

    1) object is indexed (described) in terms of a some system of classification
    2) person comes to the system and asks it questions relating to some need
    3) person examines retrieved set of answers, if not satisfied, iterates or exits
    4) person uses information

    Ranganathan's colon classification excels in terms of presentation (step 3) because, unlike a pre-coordinately arranged list of subject headings like the antiquated Library of Congress Subject Headings, a description of an object based on PMEST and using the colon system offers the flexibility of rotation (post-coordinately rotating terms). A good example of this is offered in the Mystic Seaport example for AIR-TO-GROUND MISSILE. The advantage in presentation is that terms within each of these facets can be used to browse objects with similar descriptions within a facet. Additionally, when browsing an alpha list of descriptions the string can be found alphabetically by each term individually: AIR, GROUND, MISSILE.

    While this might be new in terms of presentation on the web in excellent examples like the Flamenco project, it should be noted that there are much older examples in paper and in electronic databases that have used rotated term displays of facet-based strings in indexes (like the Modern Language Association). These are excellent resources to look at when you consider the domain and scope of a hefty index like the MLA bibliography.

    So my point? Interface designers of devices have a rich literature that they can peruse to find tried and tested ways of displaying information. We shouldn't just muse about Ranganathan if we want to see how to really exploit what he's proposed. I think it also makes sense to see what has been done and take strength from that. I'm sure you'll find that the Information Retrieval literature is rich with analysis of methods influenced by Ranganathan.
    Posted by Michael @ 11/02/2002 04:52 PM PST [link to this comment]

    COMMENT #2
    Also agree with Peter and Michael. Context-aware computing is the next frontier in mobile computing. While "push" failed on the web, I think it has a definite place in the mobile world, for instance.

    And yes, faceted classification is the way to go. But, though librarians and info scientists know a great about this, as Michael points out, very little of was is known can be applied. (Typical gap between theory and practice). So, the real challenge is to break down what is already known about facets into a usable form for context-aware mobile computing. Is there an IA in the house?

    What scares me though (and I mean *scares* me) is WHO will be providing the information context for me? My guess is some commerical entity or entities. I have this image in head of my cell phone reminding me that a McDonalds is just around the corner and the in the next mall I can buy the latest Nike sneakers and that I can get the best news from CNN. Yuck.

    Posted by Jim @ 11/04/2002 09:05 AM PST [link to this comment]

    COMMENT #3
    Interesting point Jim. Actually Peter talked about Ranganathan's original intention. A part of that vision was a classification that corresponds to Reality. An implicit assumption is a "one classification systems fits all" approach. This also scares me a little bit. There was no "it depends" for that guy. I think the possibility exists that a MegaCorp will create a taxonomy offer that people can't refuse....

    Peter -- loved your comments and you are correct. The colon classification never caught on much outside of India. He had to jump through philosophical hoops to get shelf placement to work. As you say, this does not apply to hypertext.
    Posted by Mike Steckel @ 11/04/2002 01:23 PM PST [link to this comment]

    COMMENT #4
    Perhaps we'll be able to create our own categories and associate them with categories or feeds from other sources, a la MoveableType's TrackBack system. I can create a personal category called "Crazy Web Design Ideas" and subscribe to sources that variously call that "Information Architecture", "IA", "Interaction Design" or "McDonald's Web Organization Standards."
    Posted by Andrew @ 11/05/2002 09:55 AM PST [link to this comment]

    Add A New Comment:


    E-Mail (optional)

    Homepage (optional)

    Comments Now with a bigger box for text entry! Whee!

  • All contents of are © 1998 - 2002 Peter Merholz.