In this article,the author suggests that the "online revolution"will have a farreaching significance for the present day society,and that China should take into account the TRINITY of computers,communication and information in the development of information services and its related policy. He also discusses the current status of online networking and information services and their implications for information policy.

The importance of computerized library network is pointed out.Some external factors and internal reasons of deve loping the computer network system for the library application are discussed,and go further into the appli cation of OPACs terminals in the library.Based on the SIOCL system,a userinterface under the environment of network and the client server bibliographic system are been developin g.The expanded functions and features of the system are discribed.

The article introduces the des ign thoughts,implementation methods and technological features of CD-chinbooks according to special requirements of bibliographic,data CD-ROM and retrival system.

Teh article introduces several term-weighting approaches of automatic indexing,and their usage in automatic indexing experiment of Chinese document .It also provides data of the experiment,analysis of data and conclu sion,then discusses a few issues in automatic indexing of Chinese scientific and technological documents.

In this paper,the recent development on various aspects of the revised edition of the Chinese Thesanins and its advantages are disansed on the basis of compairing the rvised editior with the first editon,its impacf on the automatic indexing of Chinese Documents is expounded as well.The prspect of the Chinese Thesaurs is exploved in the view of avtomatic indexing for Chinese documents.

Some techniques developed by Japanese authors have allowed the application of natural language processing to the area of information retrieval.This paper briefly describes the research of automatic indexing and the usage analysis of English verbs based on the occurrence frequence of terms in a set of documents.An extracting method for Japanese nouns from texts which combining from both of Chinese and Japanese characteristics is discussed.

This paper discusses the advantages of applying hypertext techniques to information retrieval,introduces and analyses some typical models and corresponding retrieval techniques of navigation-based systems and query-based systems,and then draw the conclusion that an effective hypertext retrieval system must integrates traditional searching,hypertext and artificial intelligent.

It is important that pattern matching be used in fulltext searching systems. This paper introduces and disusses some fast pattern matching algorithms at home and abroad,improves part of function δ1 and puts forward the function δ3 for speeding up pattern matching.In add ition,the paper thoroughly describes pattern matching algorithm of function δ3.

Based on the analysis of the short comings of Fukushima method and its origin,a new method called Fukushima+ i s presented in this paper.More than 300 retrieval prescriptions used in the int ernational online retrieval have been processed by both Fukushima method and F ukushima+ on AST 386 microcomputer.The results indicate that the maximum number of working areas needed for all the qu ery formulations in Fukushima+ is only 5,but in original Fukushima method 5-7 working areas are required for only 63% of query formulations with 17% of overflow rate. 

This author discusses the research level and development status of multimedia technology in overseas,and also briefly presents the development of multimedia technology in China of last years.This paper approaches a critical technology of multimedia,system,e.g.,data compression technology,special microprocessor of multimedia,multimedia opreating systems, applicatoin software authoring tools for multimedia and multimedia systems network technology,and prospects for the developent of multimedia technology.

PRPRS(Patent-Reel munber Parellel Retrieval System)is a practical system.After you input a patent number of a country or an organization,it will timely output the number of the reel in which the original is located.The system is successfully developed with C language under DOS on PC.Its interface is very user-friendly,and its response time is fewer than 7 seconds.The database which supports the system has more than one million records,and it can expanded continuosly.Although the compress technique is used to set up the database,the record is logically put in order,therefore the retrieval speed is raised.