Beyond Boolean Logic and Hypertext:
Alternative Dimensions of "Information Space"

Presentation to "Organization of Knowledge" Class
Texas Woman's University
School of Library and Information Studies
December 12, 1998
Dr. April Bohanan, Assistant Professor

by joanne twining, Doctoral Candidate twining@intertwining.org

View Presentation Handout (pdf format)
Hyperlinked References

Introduction
(delivered via email to class participants prior to meeting):

Returning to the discussion in our first class meeting about the d->i->k (data, information, knowledge) transfer process, I found this very interesting interpretation in the lecture notes for an Engineering class at Dartmouth:
http://thayer.dartmouth.edu/~gvc/lectures/lect-12.html

>snip
1. Data are messages or signals with uninterpreted content. A newborn infant sees the same light waves, hears the same sounds and tastes the same foods as an adult but the signals are not yet interpreted very much. Things like loud sounds might be interpreted as danger, a mother's voice is familiar from the womb and so on however. This is analogous to listening in on random packets over the Internet. The meaning of the data is not clear except to the sender and receiver of the packets.

2. Information is communicated when when signals or data have some semantic or content. In order to communicate information, the sender and receiver must have previously agreed to the representations of the symbols and the meaning of message content. Thus for example, a child eventually learns that different sounds mean different things and so communication using natural language can begin.

3. Knowledge is communicated when the alphabets and codings have enough representational power to describe other codings within a class of possible coding schemes;

4.Wisdom is the ability to generate new coding schemes, decide when they are needed and appropriate as well as the ability to communicate the process by which such new coding paradigms can be generated.
<unsnip

Relative to our projects, per these definitions, if we have created symbolic representations, or graphical guides, which can stand alone without accompanying interpretation, or are intuitive (need no instruction or pre-negotiation of meaning), then we have created information, and mostly we have been approaching it from the data end of the d->i->k equation.

We have, of course, been paying particular attention to the underlying ARCHITECTURE of information: the structures (and particularly the memes, metaphors, graphical representations and other cognitive tools humans use to interpret or ascribe meaning to data), relying on our ability to maximize "how" we believe users most generally think in order to create information products which are of general use.

Such products are based on existing data systems: the structure of the library collection and the (to appropriate Amy's metaphors) finding and locating tools that give primary access to that structure. This is of practical importance.

Since I do not have the luxury (or the impediment) of a practice tied to a discrete, legacy system, my project looks at an alternative understructure, and from an alternative direction...for tools we might appropriate in order to construct an alternative environment, or information space, on which to base our thinking.

As you recall, my architecture problem is rendering the Nitecki Model of Metalibrarianship. http://venus.twu.edu/library/Nitecki. My goal is to free Nitecki's Model from its literary and two-dimensional information space in order to exploit its potential for experimenting with the intellectual structure of librarianship.

As you know, the complicated intertwining of the many aspects of our existing library and information systems make it technically inpracticable to actualize change beyond a small tweek here and there without risking potentially devasting consequences for the system whole. We have no way of visualizing beforehand the systematic consequences of various manipulations of the many aspects of the system. It is my belief that rendering the Nitecki Model will give us the tool we need to experiment, and from there, to improve our prescriptive and diagnostic ability...without causing the existing system to crash.

Unfortunately, the graphic rendering capabilities of even state-of-the-art computer systems do not provide the ability to represent multiple three-dimensional objects in simultaneous, interactive space. We cannot SHOW what might happen if we took the the Nitecki sphere (the information environment comprising mind, material, and meaning) AND the Nitecki cube (comprising any combination of 27 procedural, contextual, and conceptual aspects of LIS), and realigned or ascribed new attributes or algorithms to their inter-coordinated aspects. In fact, we cannot even ascribe manipulable attributes to the various aspects of even ONE dimension of the model.

My argument to this point has been made in my article, "New Dimensions in Information Architecture," http://www.intertwining.org/collaboratory/papers/twining/003.html, which will be published in the Spring 1999 issue of Library Philosophy and Practice http://www.uidaho.edu/~mbolin/lp&p.htm

Failing there, then, I looked at the underlying structure of our state-of-the art information delivery systems and to the experimental ISR research for a clue. I found myself travelling beyond Boolean logic as the basis for retrieval, and beyond hypertext as the basis for delivery.

My presentation looks at the intersection of the Vector Space Model (VSM) of Information Storage and Retrieval, developed by Gerald Salton in the late 1980's (and particularly at its Lexical Semantic Indexing capability,) and the newly-emergent eXtensible Markup Language (XML) protocols. I will explore the potential at the intersection of these two environments for rendering and manipulating the Nitecki Model of Metalibrarianship.

My presentation, "Beyond Boolean Logic and Hypertext: Alternative Dimensions of 'Information Space'" will describe the geometric, multidimensional intersecton of document and query spaces possible via the Salton's VMS model and his SMART system, and the object-oriented, element/attribute aspects of inner-documentary space made possible by XML.

I will attempt to show that rendering the literary representation of Nitecki's Model as a graphical XML document served in a VMS system MIGHT allow us to "see" what will happen if, for instance, we conceptually tweak the library's economic subsystem by reallocating a fixed dollar amount from one function to another.

View Presentation Handout (pdf format)

___________________________________

Hyperlinked references

twining, joanne. 1998. "New Dimensions in Information Architecture." www.intertwining.org/collaboratory/papers/twining/003.html In Press, Library Philosophy and Practice, Spring 1999 issue.

The Nitecki Trilogy http://venus.twu.edu/library/Nitecki

 

VECTOR SPACE MODELS

An Overview of Information Science
www.cs.umbc.edu/~nicholas/courses/491/html/IR.html

Vogt, Christopher C. 1997. Adaptive Combination of Evidence (ACE) for Information Retrieval: A Thesis Proposal. University of California, San Diego. http://www-cse.ucsd.edu/users/vogt/papers/proposal/proposal.html

A Vector Space Model for Automatic Indexing
cs-tr.cs.cornell.edu/Dienst/UI/1.0/ Summarize/ncstrl.cornell/TR74-218

Personalized Information Delivery: An Analysis of Information Filtering Methods. Peter W. Foltz and Susan T. Dumais, for Bellcore. Preprint http://superbook.bellcore.com/~std/papers/CACM92.ascii

Salton, G. and McGill, M.J. 1983. The SMART and SIRE Experimental Retrieval systems. NY: McGraw-Hill. pp. 118-155.

EXTENSIBLE MARKUP LANGUAGE

XML FAQ http://www.ucc.ie/xml/ - FAQ-ACRO

THE SGML/XML Webpage http://www.oasis-open.org/cover/

Why XLM Mattters, from the October 6, 1998 issue of PC Magazine
http://www.zdnet.com/pcmag/features/xml98/index.html

What is the Document Object Model? www.w3.org/TR/WD-DOM/introduction.html

The Legion of DOM http://webreview.com/97/11/14/index.html

The DOM Tutorial http://www.webcoder.com/howto/15/index.html

CommerceNet's XML Exchange http://www.xmlx.com/

Beyond HTML www.beyondhtml.com/markup/xml/index.shtml

PROPOSALS TO THE W3 CONSORTIUM:

Vector Markup Language (for markup of graphic objects)
http://www.w3.org/TR/NOTE-VML#h3:introduction.example

Precision Graphics Markup Language (PGML)
http://www.w3.org/TR/1998/NOTE-PGML

-30-