Conference ulysse::rdb_vms_competition

770.0. "Japanese database use predictions" by COOKIE::DEVINE (Bob Devine, CXN) Fri Oct 19 1990 21:18

  [Dr. David Kahaner is a numerical analyst visiting Japan for two-years
   under the auspices of the Office of Naval Research-Far East (ONRFE).  
   The following is the professional opinion of David Kahaner and in no 
   way has the blessing of the US Government or any agency of it.  All 
   information is dated and of limited life time.  This disclaimer should 
   be noted on ANY attribution.]

  [Copies of previous reports written by Kahaner can be obtained from
   host cs.arizona.edu using anonymous FTP.]


To: Distribution
From: David K. Kahaner ONRFE [[email protected]]
Re: Japanese Database activities.
Date: 11 Oct 1990

ABSTRACT.
We describe the activities of the Japan Database Promotion Center, and 
summarize the current state of affairs with respect to databases in Japan,
especially those of interest to scientists.

PREFACE. 
I am not a active database user nor have I ever done database research.  
Nevertheless it is apparent that databases are having a growing impact on 
science. This report is based on information from the Database Promotion 
Center, colored by my own personal impressions here in Japan.  

INTRODUCTION.
Japan is far behind the United States in the use and development of 
databases, particularly for science. The situation is well known and 
there is a strong effort being made to correct it. The Japan Database 
Promotion Center (DPC), established in 1984, has as its goals the 
promotion of database construction, research and development of the basic 
technologies related to databases, establishment of efficient clearing 
services, education propagation and training associated with databases, 
and international information exchange informing other countries about 
Japanese databases.  
   Database Promotion Center Japan
   7F World Trade Center Building
   2-4-1 Hamamatsu-chu
   Minato-ku
   Tokyo 105, Japan
   Tel: (03) 459-8581, Fax: (03) 432-7558.


Japanese online databases in science and technology were first developed 
by the Japan Information Center of Science and Technology, and the Japan 
Patent Office, in the mid 1970s. Business databases were established a 
few years earlier. This is about 10 years behind corresponding efforts in 
the U.S.  Partly this was technology driven, but also partly driven by 
early reluctance on the part of the Japanese to sell intangibles, in this 
case information, as a product. The Japan Database Industry Association 
(DINA), established in 1979, now has over 100 members and is quite 
active. Further, since the early 1980s both the information industry 
through its member organizations, and the Japanese government have been 
attempting to promote database development and use. Since 1983, MITI 
(Ministry of International Trade and Industry) has been publishing an 
annual Database Directory, which provides a comprehensive list of 
databases which can be accessed in Japan, both Japanese and foreign.  The 
DPC is the official database arm of MITI.  

There is no question of the growing importance of information in Japan's 
future economy. This is associated with interactions between (1) trends 
in Japanese industry, (2) the rise of the database industry, and (3) 
problems of database development.  

Traditional Japanese manufacturing (iron, shipbuilding, etc.) is not 
growing rapidly. There are problems of Yen inflation relative to other 
currencies, trade frictions, offshore production, and a decrease in 
interest among young Japanese in going into "dirty" industries. On the 
other hand the information industry is expected to account for more than 
20 percent of Japan's GNP by the year 2000, reaching 144.5 trillion yen 
(more than one trillion dollars, U.S.).  (It is also estimated that this 
industry will be almost one million skilled workers short of its real 
requirements, and it is hoped that some of the excess from manufacturing 
can find a happy home in the information industry.) 

MITI estimates the following tends in information services.

          Software   Info. Proc. Services   Info. Vending Services
   1990    16.3           12.2                  3.3
   1995    37.5           23.6                 10.6
   2000    78.7           45.7                 33.9

The figures are in hundreds of millions of yen. The last category, 
Information Vending Services, essentially means database services, 
including producers, distributors, value-added network vendors (VAN) and 
information brokers. In any units it is obvious that its growth is 
expected to be very great. The DPC notes that in the west these groups 
are clearly defined, but not so in Japan, perhaps because this is still 
an immature industry. Most data originates in public organizations, 
government labs or agencies, or as a by-product of commercial activities 
or research. VAN vendors take these data, process it further, add their 
own experience, and ultimately distribute it. Most accessible databases 
in Japan (76%) are of foreign origin, thus the specialist who is trained 
to search these can provide substantial value added service. Most 
database users (90% in a MITI survey) wanted lateral access with the same 
commands, that is the ability to search several different databases using 
a uniform search process, and this is clearly an area in which there will 
be many new services provided. The new technological developments such as 
laptop computers, data on CDs, portable telephones, etc., will make 
pressure for databases even greater.  
   
There are still many problems slowing the growth of Japanese database 
activities. Some of the "we don't pay for information" philosophy is 
still at work although this is rapidly changing through urging by the 
Japanese government, via trade fairness pressure from other western 
countries, and associated changes in the legal system. As we noted most 
databases are of foreign origin and access overseas is limited and 
expensive, especially from outside Tokyo.  Japanese scientists provide 
data to JICST (Japanese Information Center of Science and Technology), 
and NACSIS (National Center for Science Information System). The latter 
is linked to NSF (National Science Foundation) in the U.S.  However, 
information from Japanese sources is simply not available to the extent 
that it is in the west, and it is not distributed throughout the country 
in nearly as uniform a manner.  Further, most Japanese universities are 
not as active as their U.S. counterparts in providing courses in database 
utilization and production.  


SCIENTIFIC DATABASES.
In 1987 there were almost 1800 databases accessible in Japan. Of these 
about 1400, or 76%. were of foreign origin, i.e., from outside Japan.  
Almost half of the accessible databases were business related, and 
slightly more than one quarter (27%) were associated with science or 
technology. The number of scientific databases have been increasing 
rapidly, but not as rapidly as other kinds of databases. Business 
databases associated with high technology such as Communications, Energy, 
Chemical Industry, etc., have been increasing particularly rapidly.  
Scientific databases were broken out as follows.  

   SUBJECT           FOREIGN  JAPANESE  TOTAL
 General sci/tech       32      10      42
 Patents                44      11      55
 Med/Pharm/Bio          88      16     104
 Chemistry              52       6      58
 Physics                 3       1       4
 Mathematics             4       0       4
 Electronics            60       4      64
 Machine                10       1      11
 Civil Engineering       8       5      13
 Space/Earth/Marine Sci 17       0      17
 Atomic Energy           2       0       2
 Environmental          28       0      28
 Energy                 22       1      23
 Agriculture            16       1      16
 Meteorology             5       7      12
 Metals                 11       0      11
 Food                    6       1       7
 Fiber/Timber            4       0       4
 Other                  18       2      20
 TOTAL                 429      66     495

The greatest growth in databases has occurred in "fact" (which contains 
primary data) compared to "reference" (which contains bibliographic data) 
databases. The primary reason for this is that the former now typically 
contain "full text," resulting from the lower cost of storage devices and 
communications. Essentially all databases are now "online" meaning that 
they can be queried directly.  

There is also rapid growth in the use of databases within Japan.  

GOVERNMENT'S ROLE.
For a brief road map to Japanese government agencies, see my report,  
"Japanese Government Science Structure and Computer Related Projects". 

MITI has formulated a database policy as follows.
(1) Study and research into production and organization of databases
(2) Taxation adjustments to encourage database producing corporations 
(3) Production of official databases as needed
(4) Creation of Database Directory 

Ministry of Education, Science and Culture is actively promoting the 
organization and production of databases in science and education. These 
include databases for national universities and various education 
research centers. Grants are also provided to academic researchers for 
appropriate scholarly research. The NACSIS system, mentioned above, is 
already accessible through NSF to disseminate Japanese scientific data.  

Agency of Science and Technology has accepted the "NIST" report (National 
Information System for Science and Technology). JICST will be the 
organization dealing with distribution of scientific and technical data.  
Funding, through Scientific and Technical Information Promotion Funds, 
are planned for advanced and basic research, encouragement of R&D between 
several organizations, strengthening cooperation between industry, 
government and academia, joint international research, and evaluation of 
research.  

UTILIZATION OF DATABASES.
Surveys show that about two thirds of Japanese corporations used 
databases, with numbers of users and costs increasing around ten percent 
per year, the iron/steel industry being the only one showing a decreased 
use. Naturally, the information processing industry is still the most 
heavy user, but financial industries also show rapid growth and actually 
spend the most money on databases. The most used databases in terms of 
cost are the Scientific and Technological Document File, Japan Patent 
Utility Model File, NIKKEI, MEDLINE, and CA-Search. Of course, users want 
easier-to-use database queries, more standardization, and more 
information. There is still a need to inform users about copyright 
issues. Those who do not use databases site no need, or nonexistence of 
needed databases.  

Three years ago an international conference on the use of Japanese 
information on science, technology, and commerce was held in Warwick UK, 
sponsored by the British Library, NTIS (National Technical Information 
Service--U.S.) and JICST (Japan). The proceedings of this conference are 
a good place to read about problems of language, information quality, 
utilization and access issues of Japanese databases. A survey conducted 
at that conference indicated five issues concerning Japanese databases.  

(1) Many Japanese databases cannot be accessed from outside of Japan.
(2) No Japanese counterpart to American Center or British Council for 
    public information services.
(3) Overseas users have to rely too much on document delivery services.
(4) No comprehensive business related database in Japan.
(5) No service allowing access to a number of databases.

As of July 1988 it was determined that only 83 databases from within 
Japan are offered for overseas use including only 20 related to science 
and technology and 51 to business, although about another two dozen were 
in the planning stages.  Overseas users noted serious cost issues in the 
use of Japanese databases, associated with translation and communication.  

PROBLEMS IN ACQUISITION OF JAPANESE INFORMATION OVERSEAS.
The DPC notes that "the provision of Japanese information to the 
international market via online services is a national priority which 
must be dealt with jointly by government, industry, and academia." 
They have identified four major problems.

(1) International information frictions.
    Domestic and foreign efforts are separate with little cooperation. 
    International standards are needed so that overseas users are 
    considered even during the initial design of Japanese databases. 
    Efforts to improve the distribution and publicity of academic 
    Japanese journals in science and technology are very weak.  
    Improvements need to be made during the writing, editing, and 
    publication stages.  Technical reports and government publications 
    are difficult to access (these are sometimes called "grey 
    literature"). 
(2) Production, distribution, and utilization of Japanese information 
    needs to be internationalized. Standards for information on science 
    and technology need to be established and adhered to. Relationships 
    must be forged with overseas database service organizations.  Basic 
    information, such as tables of contents, titles, etc., need to be 
    translated into English.  As a case in point I would like to mention 
    the difficulties that I have had with papers from the Japanese 
    Information Processing Society.  At each of this year's semiannual 
    meetings more than one thousand papers are presented. The papers are 
    mostly in Japanese but over 98% have English titles.  Nevertheless, 
    the only place these English titles appear is at the bottom of each 
    article in the proceedings, currently six volumes per meeting.  The 
    Society can only extract the titles by manual keyboard entry. If the 
    organization representing Japan's most advanced computer technology 
    has not come to grips with this problem it is not surprising that 
    other fields are even further behind.  
(3) Online databases must be made easier to use by providing, access via 
    networks, 24 hour support services, appropriate communication 
    software, a supply system for document information to libraries and 
    others, and a mechanism to reduce the telecommunications costs.  
(4) Increased publicity, and intermediaries between the end user outside 
    Japan and the Japanese database producer. Active use of new media 
    such as CD-ROMs, video disks, and electronic publishing.

NEW TRENDS RELATED TO DATABASES.
(1) Integrated services to allow access to data and then action, such as 
    product ordering.
(2) Globalization of databases, mostly through new trade agreements 
    allowing vendors of VAN to offer dedicated international lines.
(3) Multimedia databases with figures, video, and sound. One important 
    aspect of this in Japan is the rapid push to ISDN and associated 
    communication technology. At the moment Japan seems to be several 
    years behind the U.S. in the practical use of networks, but this is 
    seen as a key technology by both government and industry and the 
    pressure is really on to provide these services quickly.  
(4) An urgent need--standardization. There is an international 
    organization (ISO/TC46) working on this but the Japan Database 
    Promotion Center feels that the Japanese response to this has been 
    very negative.  
(5) Personalized databases because of links between mainframes and the 
    increasing number and power of workstations, as well as interactive 
    CDs, and more portable laptop computers.  
(6) Growing connection to AI and expert systems to support ease of use.
(7) Growth of electronic publishing.
(8) Growth of networks. 

----------------END OF REPORT-----------------------------------------