You are here

Documentation or knowledge?

Originally appeared in 1973 as part of Toward a Concept Inventory. Also published under the title "Texts or Concepts: Documentation or Knowledge?" in International Associations, 1974, pp. 205-208 and in a revised version as "Knowledge-Representation in a Computer-Supported Environment" in in International Classification. 4, 1977, No.2, pp. 76 - 81.

1. The Documentation Problem
Knowledge Representation

[Parts: Last | All ] [Links: To-K | From-K | From-Kx ]

1. The Documentation Problem

It is the stated goal of the UNISIST World Science Information System to facilitate the "unimpeded exchange of published or publishable scientific information and data amongst scientists in all countries". Its concern is therefore with the extremely large number of documents and not with the relatively limited number of original conceptual entities formulated therein. Unfortunately, the UNISIST Study does not distinguish between documentation, information and knowledge (60).

Briefly, documents pose a physical handling, transfer and filing problem (which may be eased by reproduction at a distance). Information consists of signs which can be read, transferred, manipulated and filed electronically. They function as symbols of units of human knowledge, but only during the short-duration process of being read for meaning. Knowledge transfer depends on the ability of the momentary psychological system "sign and reader" to generate an unambiguous, coherent and consistent meaning in the mind of the reader, and conversely to convert a distinct meaning or concept into a suitable sign which can be interpreted with equal ease by another reader. Information, in the form of signs. can he read without resulting in the transfer of Knowledge and particularly of the knowledge intended (e.g. undecipherable hieroglyphic writing can be "read" without knowledge transf er) .

The Study does not recognize that the period covered b the proposed system is one in which increasingly, it is almost- impossible for the decision-maker or researcher to determine what information from which discipline is "relevant" .

" is a practitioner of any one discipline to know in a particular case if another discipline is better equipped to handle the problem than is he? It would be rare indeed if a representative of any one of these disciplines did not feel that his approach to a particular organizational problem would be very fruitful, if not the most fruitful..." (R.L. Ackoff, Systems, organizations, and interdisciplinary research.)

If he attempts to order all the relevant documents (or even subscribes to the appropriate abstracting service), the purchase or transport costs will be prohibitive (except to a small elite); if he waits for all the relevant information, it will be too late for him to make a useful decision (61); if he gets all the relevant information in the form it currently takes, he will have neither the time, the training, nor the inclina tion to read it all; and if he reads and comprehends it all, he will not have the time or the ability to convey his understanding to those whose support he must obtain to carry a vote on the matter or, ultimately, to the man in the street.

"Consider this dilemma: while our technological abilities to generate and disseminate potentially useful data have increased manyfold in the past few years, man's physical capacity to register and to process potentially informative data has probably increased very little, if indeed at all. The sheer volume of data that crosses the typical executive's desk today should serve to spotlight the inadequacies of the education and development of our acquisition strategies and practices. But no gain in ability could offset the widening gap between the exponentially-increasing quantity of data available for consumption and man's very limited capacity for acquiring and proces- sing useful informational) (62)

It is questionable, in view of present trends, whether knowledge transfer can continue to be effectively accomplished primarily via document transfer. The United Nations is potentially the most significant institution in existence and is at a vital nexus of multidisciplinary, international knowledge transfer - which it currently accomplishes via documents (63). And yet it has a documentation problem (which in a sense is equivalent to that of many, if not most, other large organizations and disciplines):

"This issue has been repeatedly recognized by the General Assembly, the Economic and Social Council, the Joint Inspection Unit and nearly a dozen of other UN bodies as one which directly affects the functioning of the UN. Suffice it here to note that in 1970, the UN, both in New York and Geneva, produced nearly a million page documentation in all languages. The massive volume of documentation produced by the UN prompted a former President of the General Assembly, Mr Lester B. Pearson of Canada, to remark that "the United Nations is drowning in its own words and suffocating in its own documentation." The Joint Inspection Unit stated recently in its report submitted to the present General Assembly session that" the inspectors do not hesitate to say that the point of saturation has now been reached and indeed overstepped." (64)

The last quote in fact continues with the significant phrase "and that the law of diminishing returns is taking over...Beyond strictly financial considerations, therefore ... the future usefulness of the Organization may well hinge on its ability and determination to set once and for all, and strictly enforce a reasonable but drastically reduced ceiling to the volume of documentation its various bodies call for and its services produce" (65).

One is not exposed to alternative hierarchies of conceptual nexuses linked directly or indirectly to more distant nexuses from which relevant knowledge may be obtained. (There are no "heights" in documentation systems - the general is filed with the particular, cf. the treatment of documents with an interdisciplinary emphasis.) The potential value of a knowledge-oriented information system as an active stimulus for creative social change and problem- solving may even be directly proportional to its ability to draw attention to the existence of established relationships of low probability (i.e. low entropy) between concept nexuses. This is not a criterion of document information systems where the emphasis is - for cost reasons - on facilitating access to those documents which are most probably relevant in terms of demand frequency.

Shuffling documents and signs might facilitate the transfer of meaning and knowledge between those who could identify the representative of the group whom a particular set of meanings could be consistently and unambiguously attached to the signs. But even within that group, advances in knowledge and reconceptualization have to be carefully related to the original set of meanings. However, making the documents and signs of that group available to other "outside" groups would only introduce "noise" and confusion. A knowledge-oriented information system would be needed to avoid such confusion and facilitate fruitful interaction between different schools of thought within the social sciences.

cilitate fruitful interaction between different schools of thought within the social sciences.

[Parts: Last | All ] [Links: To-K | From-K | From-Kx ]