The next twenty years in information retrieval; some goals and predictions

Abstract

Some of the history of retrieval machine development during the past 20 years is sketched as background. At present, there still are no completely satisfactory machines for retrieval from large collections. Solution of the language problem in retrieval is currently somewhat ahead of machine development.Future developments are sketched. Work is already underway on the use of machines for the assignment of descriptors to the text, which is a crude kind of mechanical translation. Retrieval machines will soon be called upon to assist the customer in using a retrieval system by helping him to formulate his search requests. This assistance amounts to education of the customer. The notion of education by machine is extended to machine assistance in helping the customer to read the documents uncovered. It is shown how this process, when combined with work on mechanical translation, will lead to machines which can provide essays on any given subject upon request. Machines will become archival devices to store facts, not texts. Human‐to‐machine and machine‐to‐human communication will become very important. Information machines can be expected to become as numerous as computers now are.

The History of Information Retrieval Research

“…A query (2) was also on film showing a negative image of the part of the catalog being searched for; in this case, the first and sixth entries on the roll. A light source (7) was shone through the catalog roll and query film, focused onto a photocell (6). If an exact match was found, all light was blocked to the cell causing a relay to move a counter forward (12) and for an image of the match to be shown via a half silvered mirror (3), reflecting the match onto a screen or photographic plate (4 and 5).…”

“…He would dial a number, press a button and after three seconds [a microfilmed copy of] the document would be projected. [ A number of other researchers produced devices based on similar principles: Mooers [6] described investigations by Davis and Draeger in 1935 on searching with microfilm. This work, according to Mooers, was taken up by Vannevar Bush in the late 1930s, who, with his students, built a filmbased searching prototype.…”

Mensch-Computer-Interface

“…Beide Entwicklungsmodelle überschneiden sich mit einer weiteren Richtung, die in der Steigerung des Intelligenzniveaus der Mensch-Computer-Kommunikation den zentralen Entwicklungsfortschritt sieht. Bereits in den 50er und frühen 60er Jahren führten erste Erfolge der KI zu Prognosen der Realisierung des intelligenten Mensch-Computer-Dialogs um 2000 (Turing 1950, Mooers 1959Ramo 1961) In den 80er Jahren erlebten diese Erwartungen durch Expertensysteme und Rechner der 5. Generation eine Renaissance (Shackel 1985), ebenso beflügelte die Entwicklung von Software-Agenten erneut die Erwartungen in intelligente Bedienschnittstellen Negroponte 1997;Maybury/Wahlster 1998).…”

A note on descriptors

“…The Uniterm approach, since it was introduced in 1951, seems however to be migrating both in concept and usage towards the descriptor methods, as is clear from many reports coming from projects where they claim to use Uniterms". (2) As a matter of fact, however, descriptors are Uniterms --"children" is a Uniterm as well as a descriptor-and Uniterms can be descriptors. They are not different, except when a descriptor is a rather long phrase with qualified terms-but Calvin Mooers himself pointed out in an early paper (3) that descriptors are expressly designed to avoid such phrasing.…”