2

China's Use of Open Sources

During the 1950s, when the PRC government was being consolidated, China’s modernization depended critically on three props: Soviet aid, overseas scholars, and foreign scientific literature. There is a tendency today in China to downplay the importance of these non-Chinese elements and to stress indigenous contributions. While no one denies the role of early Soviet support, its termination in the late 1950s is said to have forced China down an independent path to scientific advancement.1 Betrayed by Russia and denied free access to the world’s technology, China relied on its own legacy of creative genius.

The difficulty here lies in how one interprets “indigenous.” As shown in the previous chapter, the scientists who ran China’s laboratories and technology programs were precisely those people who had studied abroad. Although Chinese in one sense, the knowledge they brought “back” with them was entirely foreign. Also overlooked is the degree to which China counted, and still counts, on openly available foreign information to jumpstart innovation and short-cut the R&D cycle. Whereas all science begins with prior art, in China the systematic use of foreign sources to promote S&T development has been elevated literally to an “information/intelligence science”  .

This chapter traces the growth of this open-source exploitation system. In the first section we examine early PRC institutions and practices. In the second section we take a closer look at the inner workings of China’s “qingbao” ( , intelligence/information) network through an unusual book written by two architects of the system – a disclosure not likely to be repeated. In the third section we describe the system’s present structure and its leading organization: the Institute of Scientific and Technical Information of China. Other first-tier open-source organizations are scrutinized in the fourth section: Beijing Document Service, China Defense S&T Information Center, Patent Documentation Library, and National Library of Standards. In the chapter’s final section we examine some subsidiary organizations that contribute to the national open-source effort. 

Building a national S&T open-source system

Chinese who specialize in foreign S&T exploitation date the emergence of a national open-source “intelligence/information” system from 1956.2 In August of that year, the PRC State Council promulgated a “1956–1967 Long-term Plan for the Development of Science and Technology,” which included as article 57 the following statement:

The responsibility of S&T intelligence work is to report the most recent accomplishments and trends in domestic and foreign science in all types of important scientific and technological fields so that scientific, technological, economic and higher education departments get timely access to the information and materials needed to facilitate the absorption of modern scientific and technological accomplishments, reduce time and manpower, avoid duplication of work, and promote the development of science and technology in China.3

In keeping with the plan, S&T information offices   appeared that same year to support China’s armaments industry, along with a new Scientific Information Institute   under the Chinese Academy of Sciences.4 By late 1957, the Institute was researching how to match foreign science and technology information with China’s concrete needs. In 1958, it was reconstituted as the Institute of Scientific and Technical Information of

China   , ISTIC), the “first and central element of the national system”5 and China’s foremost facility for acquiring, processing, and distributing foreign S&T materials.6

Also in 1958, a Chinese University of Science and Technology Information   was established, the world’s only known example, with departments of S&T information, translation and library science. Like ISTIC, it was focused on foreign materials.7 The year also saw the first of a series of “National S&T Information Work Meetings.” Delegates to the meeting produced five documents defining the goals and methods of China’s S&T open-source system, including instructions on how to strengthen S&T information work, a list of principles and techniques, a proposed structure for a national S&T intelligence network, secrecy rules, and training requirements for “information workers.”8

These calculated efforts to put the exploitation of foreign information on a rational footing contrasted with the chaos overtaking the country during its “Great Leap Forward” (1958–1960). While peasants were melting down their farm tools in backyard steel furnaces, the State Science and Technology Commission, predecessor to the Ministry of Science and Technology, in 1959 created an S&T Information Office   to coordinate S&T “information work” nationally. In December the first specialized journal on S&T information appeared, along with a magazine devoted to methods of searching English-language periodicals.9 ISTIC established a Chongqing branch in the heart of China’s military R&D district to translate foreign technical documents, which by 1966 had reached several hundred thousand items.10

Meanwhile, a second National S&T Information Work Meeting in 1961 wrote guidelines for “national S&T document registration, search and retrieval” including three sets of regulations for foreign S&T documents, foreign S&T translations, and foreign S&T materials brought back by personnel going abroad. A third meeting in 1963 set out goals to 1972, defined responsibilities of S&T information organizations within the State Council’s technical ministries, and provided another set of rules for information units at both provincial and local level.

By the mid-1960s, most ministries, state committees, provinces, municipalities, large firms, research institutes, and universities had their own S&T information organizations linked by networks for exchange that crossed regional and even national boundaries – a fully elaborated S&T information system and service network to support China’s technical development. By 1966, a decade after it started, the system had garnered and was making available to end users 11,000 different foreign S&T periodicals; half a million foreign research reports, government publications, conference proceedings and academic theses; over five million foreign patents from 20-odd countries; more than 200,000 standards from 40 foreign countries; several hundred thousand foreign product samples; and had S&T document exchange links with more than 50 countries.11

The goal of China’s S&T information system during this early period was to “size up” developments abroad so as to “improve China’s ability to do scientific research, copy things   and make products,” according to Chen Zeqian and Bai Xianyang, who studied the system’s evolution. Its aim, in their words, was to “reflect comprehensively, accurately, and in a timely manner” the trends and developments in foreign S&T. “Intelligence/information work”   – China’s term for systematically tapping foreign and domestic information – was the “main process for thoroughly understanding international S&T developments.”12

This “information work” was limited at first to finding and translating foreign publications into Chinese but, as was noted above, this developed quickly into an end-to-end system with multiple layers of redundancy. As Miao Qihao, former director of ISTIC’s Shanghai branch, stated:

A unique feature of the Chinese qingbao system is that from its very outset, it has combined an intelligence function with conventional information activity.13

In other words, the system was targeted not at S&T information per se but at specific types of information useful primarily to the defense industry, graded by feedback loops and metrics. By the early 1960s, it was providing critical support to China’s nuclear weapons research, satellite launches, and the mainframe computers used for military development.14

Data on the workings of the qingbao system for 1966 to 1975 are scant; those Chinese who comment on it lament the lack of progress, here and elsewhere, during the so-called Cultural Revolution.15 National S&T Information Work Meetings did not reconvene until 1975 – a 12-year hiatus – and it was not until 1977 that the enterprise got back on track with the release by the State Science and Technology Commission of an S&T development plan that prescribed updated responsibilities for the S&T information system.16

This plan departed from past practice by mandating the use of advanced technology for the acquisition and distribution of S&T materials, thus closing the circle: China would use foreign technology to improve its ability to exploit foreign technology. The following year, 1978, was marked by three significant developments:

1. The China Society for Scientific and Technical Information  

 —a professional group of S&T information workers founded in 1964 but dormant during the Cultural Revolution – convened its first meeting. The group renewed its commitment to “broadly collect foreign scientific and technical information materials through all types of venues.” Units at all levels resumed their gathering of foreign S&T materials.17

2. ISTIC enrolled its first group of graduate students in “S&T information”  , followed by Beijing University, Wuhan

University, and other top colleges. Analogies outside China to this new academic discipline would be “library science” or, to stretch things, “scientometrics” albeit limited to S&T topics and focused on foreign materials. Essentially this was a graduate degree in exploiting foreign scientific literature.

3. The Beijing Document Service  , a key provider of foreign military S&T information, was founded by COSTIND’s S&T

Intelligence Bureau and the Beijing Science and Technology

Association  .

Another milestone was the Fifth National S&T Information Work Meeting in 1980, where a decision was made to tie intelligence work closely to economic construction and S&T development and to “broadly exploit”  novel information sources. This newer period is characterized by Chen and Bai as one of “integrated research and policy support” with greater “processing” of information, closer coordination with key national projects and their specific technologies, and an emphasis on solving “knotty” problems in research and production.18 At this juncture, China’s open source system had become, in all respects, a component of its all-source intelligence network targeting specific technologies for particular customers.

In 1982, the China Society for Scientific and Technical Information held its first policy seminar and issued recommendations, exercising its function as a “mass organization” that translates state doctrine to collectors and marshals their compliance. Two years later, a sixth S&T information work meeting created guidelines for computer search and retrieval, document indexing, format standardization, and professional training. The SSTC reestablished its defunct S&T Information Office for national oversight; ISTIC began conducting directed searches for customers in support of technical programs; and the first databases and automatic abstract compilers began operating.19

By 1985 there were 412 major S&T intelligence institutes nationwide, including 35 attached to the State Council’s technical ministries, 33 at the provincial and municipal levels, and 344 local institutes employing more than 25,000 people.20 Miao Qihao, the Shanghai ISTIC Director, adds to that figure some 3,000 “basic cells” in grassroots units such as companies and labs for a total of 60,000 workers engaged full time by 1985 in data processing (investigating, collecting, sifting, analyzing, synthesizing, and repackaging data “in response to specific requirements”), database mining, benchmarking, and “reverse engineering.”21

The proliferation of units engaged in S&T information work led to efforts by oversight bodies to reorganize the network. In a notice dated January 1989, the Science Commission stated:

Up to now China’s S&T information system has been in possession of a large volume of S&T documents that have had an important role in advancing the country’s economy and S&T development. However, it is still far from adequate and many problems remain. We have not been able to form an effective system of document resource support. It lacks unified organization and its distribution is irrational. Some areas [of China] have a surplus of documents with duplication and underutilization, while other areas do not have enough. We are not reporting adequately on the content of our holdings, indexing is incomplete, and we lack national and local registries of publications. Most of the document handling and storage is done by hand. There are a limited number of document databases and a rational division of labor between different systems across networks cannot be carried out. And we are not spending enough on document acquisition.22

The Commission laid out three general requirements for: a comprehensive solution in which the particular natures and responsibilities of individual components are clarified and made to supplement one another; horizontal integration of the main organizations; and different levels for the nationwide system, including state, specialized ministries and departments, and local levels.

It went on to specify the five “main national level organizations” as follows (Chinese names are those given in the 1989 document):

1. Institute of Scientific and Technical Information of China  

.23 “This is the national comprehensive S&T information center primarily responsible for collecting and storing documents on engineering technology, management science, and high technology.”

2. Chinese Academy of Sciences National Science Library

 . “This is the national natural sciences information

center primarily responsible for collecting and storing documents on mathematics, physics, chemistry, astronomy, geography, biology, hyphenated disciplines, and high technology.”

3. COSTIND S&T Intelligence Bureau  .

“This is the national military S&T information center primarily responsible for collecting and storing documents on military technology, engineering, weapons, and equipment.”24

4. China Patent Office Documentation Library  .25 “This is the national patents document center, responsible for collecting and storing documents such as patent manuals, patent announcements, and patent category indices.”

5. State Bureau of Technical Supervision Standards Information Center 

 .26 “This is the national standards document

center responsible for collecting and storing documents on international standards, regional standards, national standards, standards for particular industries, and corporate standards.”

These first-tier   units were collectively tasked with the “basic collating and distributing of the original foreign language documents in their holdings that the foreign magnetic tape services bring in.” A second tier   was made up of two “levels”: (1) “S&T information centers in the State Council’s [technical] departments and committees that collect and store documents closely related to their own specialties,” and (2) local S&T information organizations “at the provincial and municipal region level, and at the local city and county level. The scope of their document collection is determined by their [own] economic and S&T planning.”27

A second major reorganization occurred in June 2000 when the State

Council established the National Science and Technology Library 

  , “a virtual S&T document information service

organization”28 that subsumed ISTIC, the National Science Library, and three other technical libraries, completing the transition of China’s opensource intelligence system to one that is primarily library-based. Other proposals addressed modernization and professionalization of the system’s cadre:29

  To the traditional tasks of publishing and translation were added

"aggregation and analysis of intelligence and information" 

 , software research, timely service, and an emphasis on

end user consultation.

  China began looking at new methods for "collecting, transmitting, and managing foreign languages, text, voice, images, tables and data."   An earlier requirement for entry into the "intelligence corps"  to be "good in one area"  , such as a technical discipline or foreign language, was no longer considered adequate. The modern open-source worker had to bring to the job IT skills, management savvy, and an academic background.

  Open-source workers were also asked to play a larger role in the intelligence process, becoming a "think-tank"   for the party and government by providing top leaders with information and playing a direct role in the policy-making process  .

  "Security consciousness"   also came to be a factor.

Although some reductions were made in the number of institutes and personnel, this was a consequence of greater automation, particularly webbased services. The role of open source itself continued to grow. According to Science Ministry statistics, by 2005 there were 15,782 people working in 353 S&T information institutes, down from some 20,000 people in 433 institutes in 1995.30 Despite this absolute decline, the figures increased relative to the number of people and institutes engaged in real R&D: from 8.36 S&T information institutes per 100 R&D institutes in 1997 to 11.95 per 100 in 2005. The corresponding figures for personnel were 3.14 S&T information workers per 100 R&D personnel in 1997 and 5.62 per 100 in 2005.31

The same was true of budgets. According to ISTIC’s Chen Jiugeng, the annual funding increase from 1997 through 2005 for S&T information institutes exceeded the growth of state expenditure on R&D, absolutely and per capita.32 It is hard to see how investment like this supports China’s stated goal of becoming a scientifically “creative nation”  , except in the restricted sense of “creatively” adapting other countries’ creations.

Chen provides the statistics in Table 2.1 on the combined holdings of the 353 institutes in 2005, which gives a sense of the operation’s scale.

More statistics: in 2005 the number of networks used to host and distribute S&T information among the 353 institutes was 50,534, servicing 27 million “users”  . Chinese who accessed overseas networks through the institutes to obtain foreign S&T materials directly numbered slightly over one million.33

These are impressive numbers, which is not surprising since the qingbao system is largely library-based. Or is it library-based because that is the only way to manage a system this big? The key facts, however, that distinguish it from “libraries” as commonly understood are: (1) the Chinese system is run by intelligence experts working for the government, (2) in concert with end users, and (3) to short-cut R&D by leaning on foreign models. As Chen and Bai note,

 

Through more than 50 years of development, the scope and scale of China’s S&T information enterprise has continuously expanded. It has supported and advised the selection and research of comprehensive, forward-looking, and strategic key issues of the greatest concern to leaders and management at all levels, providing solutions for technology breakthroughs and management and policy decisions.34

S&T innovation in China is driven by foreign developments, tracked through open sources. Solutions to technical problems are culled from a knowledge base of prior art, freeing Chinese resources for commercialization and production. While this approach to innovation – the “early adapter” – is hardly unique, what differs with the Chinese system is its scale, and the degree to which it characterizes the entire R&D enterprise. In the following section we look more closely at how the system works in practice.

The view from China's "Spy Guide"

In 1991, two years after the State Science and Technology Commission reorganized China’s open-source intelligence system into its modern form, a remarkable book appeared describing the system’s goals, methodology, and targets. Entitled Sources and Methods of Obtaining National Defense Science and Technology Intelligence   , the book is a comprehensive account of China’s foreign military open-source collection, comprehensive enough to prompt one reviewer to call it “China’s Spy Guide.”35 Since the book deals with one intelligence venue only – open source – the nickname is overkill. But its characterization of open source as an element of PRC espionage is spot on.

Sources and Methods is based on materials compiled by authors Huo Zhongwen   and Wang Zongxiao   for a graduate course at the China Defense Science and Technology Information Center (CDSTIC), one of the five national open-source intelligence organizations that emerged from the 1989 regrouping. Although they do not acknowledge it, the authors likely played a key role in the overall system’s evolution. Many recommendations that appear in the 1991 book parallel the 1989 document on which the reorganization was based. If you allow for preparation and time-to-publication, it is likely that drafts of this 361-page volume predate the SSTC edict. As its authors state, the impetus for “a rational overall arrangement of national defense S&T document resources was begun by COSTIND’s S&T Intelligence Bureau in 1986” (i.e., Huo and Wang’s home unit), and spread to “each respective intelligence organization.”36

It is also worth speculating on how a book like this saw the light of day. One reason may be that it was not as big an anomaly in China as it seemed to be in the West. When Bruce Gilley drew attention to the book a decade after its release, materials on this topic were still available in Chinese. Huo and Wang admit their debt to “papers and books written by many of our domestic colleagues” and their bibliography contains a good sample.37 Thus the only real secret was the West’s failure to notice it. Another reason for the book’s release may be that the publishers took its thesis for granted. The use of foreign models to guide S&T development is so pervasive in China that no one imagined any comeback.

Huo and Wang begin their book by noting that the volume and variety of S&T information now available require changes in the way in which it is sought and managed. The old ways of operating could not deal with the explosion in foreign sources. These prior practices included: (1) gathering data and waiting for customers to use it; (2) judging quality from the amount of information collected; (3) focusing on written documents only; (4) considering one’s own collection in isolation; (5) hiring staff simply because they know a foreign language; and (6) spending without planning.38

The new paradigm, which the authors helped put into practice, involves: (1) targeted collection; (2) metrics based on customer feedback; (3) collecting all types of media and making it available in database form; (4) treating collection as a science governed by systems theory; (5) hiring staff with IT skills who can adapt to consumer demand; and (6) planning before spending.39

Intelligence to the authors is a “science” and much of the book is spent explicating its theoretical basis as a prelude to their practical recommendations.40 Collection, the basis for intelligence, is not simply a task but a complex social activity with continuous dialog among all of its elements. The emphasis must always be on consumer needs and helping consumers understand these needs, based in part on what is available for exploitation. In China, part of the collector’s role is to inform consumers about what is realistically possible based on what may be found.41

So what does China collect? Huo and Wang provide a summary:

S&T periodicals, conference records, S&T reports, government publications, academic degree treatises, S&T books, standards, product samples, patent documents, and others (such as newspapers, technical archives, and drawings).

Verbal information, an area of intelligence not normally associated with open source and usually gathered by a country’s clandestine services, is prized by the authors for its timeliness and accuracy. It is more current than written records. You get it instantaneously instead of two years after the lab experiments were conducted.42 It is also easier to collect:

The party presenting a lecture or the two parties in a discussion are lecturing or having a dialogue within the bounds of a determined topic, and obtaining intelligence from a colleague who is studying the same topic is clearly more suited to one’s needs and much more convenient than searching through the relevant sections scattered in hundreds or thousands of documents. Secondly, because the feedback in verbal exchanges is rapid, when there is something you don’t understand, you can ask about it and clear it up, and when you find some new intelligence leads, you can pursue them.43

The authors categorize foreign sources by how they are obtained: scanning type, tracking type, surveillance type, topical collection, and complete set collection, access to which is described in terms of an elaborate search theory. The efficiency of collection is assessed then by “numerical probability values,” such as:

the probability of collecting the needed information within the period of time stipulated by the consumer; the mathematical expectation of the amount of the needed information that will be collected within the period of time stipulated by the consumer; and the mathematical expectation of the amount of time required to collect the needed information, etc.44

They go on to propose a collection taxonomy based on media type, the level of processing, technical nature of the content, its field of application, transmission means, user demands, time constraints, level of expectation (the information’s predictability), whether the sources are “internal” (Chinese) or “external” (foreign), specialized or synthesized, organized or diffuse, the level of compression, accuracy of content, and the probability of its existing at all.45 Everything is tagged and binned. Nothing is overlooked. Source evaluation is based on an indexing scheme that takes into account reliability, suitability, timeliness, availability, cost, and ease of decoding. A number of formulas are proposed to quantify these evaluations.46

An entire chapter (6) is given to explaining transmission channels for information – what to do with it after you get it – for example, serial, centralized, ring, bilateral, and mutual, each having its own advantages or drawbacks for passing information from collectors to user. Other characteristics such as time, capacity, susceptibility to interference and, of course, security are also discussed.

The technical nature of S&T materials, the sophisticated requirements of consumers, and the rigors of “scientific” collection demand a professional cadre of open-source specialists working constantly “according to fixed collection policies.” By the authors’ count these workers populate some “4,000 intelligence organizations throughout all of China.”47 This figure roughly matches the ISTIC officials’ estimates (cited above) for the mid1980s of 412 major S&T intelligence institutes, another 3,000 basic cells, and an undisclosed number of S&T novelty search centers.

To clarify their topics, Huo and Wang give examples of real sources they have used. They acknowledge the Lockheed Corporation with its “Dialog database” online dissemination system.48 They also describe the AD reports sold by the US National Technical Information Service (NTIS). US military standards are highly valued:

The US Naval Printing and Publishing Center sells them in book form, while the American

National Standards Institute sells them both in book and film form. The US Information Processing Service Corp. sells them in cassette film form, while the Global Engineering Documents company in the United States can output a section or sections of the military standards data in book form, depending on the user’s specific requirements.49

PRC open-source collectors are enjoined to learn the “information output periods” of foreign sources and to scrutinize publication schedules, especially plans to declassify secret materials. One such case involved a Nuclear Weapons Data Handbook published in the United States:

The first volume was published in January 1984, and while it was originally determined that the next one published would be volume three, the publication plan was again adjusted so that they issued volume two in April 1987. There are some S&T personnel who are anxiously waiting to read this series.50

Under “Response time to user needs” we find:

Take for example when you want to buy a publication of the US Congress that has already been issued openly. Buying it through an information source like the China National Publications Import and Export Corporation takes about one year before you receive it. Buying it overseas through an information source such as an institution with an overseas office generally takes about two to three months before you get it. However, using the express collection method of a certain document company, in general you will get the material in two to three weeks.51

In describing “characteristics of national defense intelligence sources” the authors explain (with examples drawn from NASA, RAND, and NTIS) that the data are often classified. However:

A common saying has it that there are no walls that completely block the wind, nor is absolute secrecy achievable, and invariably there will be numerous open situations in which things are revealed, either in a tangible or intangible form. By picking here and there among the vast amount of public materials and accumulating information a drop at a time, often it is possible to basically reveal the outlines of some secret intelligence, and this is particularly true in the case of the Western countries.52

Sometimes collectors get lucky thanks to careless declassification decisions, one of which led to the release of a top-secret DOE report on thermonuclear weapons. In the authors’ words, “It was like finding a rare treasure.”53 Huo and Wang cite another example of US mistakes that bring cheer to their workday:

From 1971 to 1976, the US Department of Energy conducted declassification reviews of a large amount of classified material, covering a total of 2.8 million items, of which 1.5 million were declassified. At the Los Alamos National Laboratory, they reviewed a total of 388,000 documents in 33 days, so each reviewer had to review around 1,000 documents a day, about two a minute. The pace of the reviews was startling, and resulted in a large number of errors — around five percent – that is, some 19,400 documents were mistakenly declassified, and of these there were at least eight highly secret items regarding thermonuclear weapons, which ended up being open material that could be browsed freely by outside visitors.54

They conclude:

This incident tells us that, on the one hand, absolute secrecy is not attainable, while on the other hand, there is a random element involved in the discovery of secret intelligence sources, and to turn this randomness into inevitability, it is necessary that there be those who monitor some sectors and areas with regularity and vigilance.55

Reference works published by foreign countries on their open-source holdings are used by Chinese intelligence officers as a short cut to identify materials. Other such indices are compiled locally in China (e.g., Searching Foreign Science and Technology Documents and Materials by ISTIC,56 and Basic Knowledge About Chemical Literature by Yang Shanji and Yang Jingran, (1981):

This book focuses on introducing chemistry and chemical engineering materials that are commonly seen in foreign countries, including periodicals, conference proceedings, scientific and technical reports, patents, abstracts, summaries, book series and collections, dictionaries, and various large reference works. The book has 12 chapters in all, which are, in order: Overview of Books and Literature; Periodicals – An Important Information Source; Document Search Tools;

Scope and Application of the Index of the US Chemical Abstracts; Summaries, Collections, S&T Reports and Academic Degree Treatises; Patents and Searching Them; Dictionaries, Handbooks, Physics Tables and Spectral Data; Organic Chemistry Reference Books; Inorganic Analysis, Chemical Engineering, and Materials Reference Works; S&T Literature Retrieval Services; Development Trends in Chemical Information Retrieval; and Books and Materials.57

In a section entitled “Typical National Defense Intelligence Sources and Materials” Huo and Wang list several US sources. Congressional publications, particularly those by committees on national defense and appropriations, are analyzed in detail. The authors say that these sources yield data on:

The US military's view and estimate of the world situation.

The research and development plan for American weapons and equipment, as well as the objectives and rationale for the Americans' development of various kinds of strategic weapons, conventional weapons, and C3I.

  The status of American investment in the development of weapons and equipment.

  The status of scientific research, testing, and evaluation of American weapons and equipment.58

They also make use of the Congressional Information Service, through which “you can find on your own all the Congressional publications and statistical information you need,” especially “AD reports” (Armed Services Technical Information Agency Document), which are described accurately as “scientific research reports on research projects funded or assisted financially by the US Department of Defense.”59 Moreover:

The content of AD reports touches upon every area of national defense S&T, such as aviation, space technology, guided missile technology, nuclear technology, ordnance, military science, electricity and electronic engineering, communications research, etc. Therefore we can say that AD reports are a major source of intelligence on national defense S&T work.60

Huo and Wang show detailed knowledge of AD reports’ declassification procedures, their numbering schema, the organizational histories of US agencies producing and distributing the reports, and their distribution schedules. Besides CDSTIC itself, ISTIC’s main Shanghai and Sichuan branches, and the Beijing Document Service, all have complete sets.61

The book describes everything a collector needs to know about NTIS: what materials it makes available from whom, in what formats, its structure and customer base.62 NASA reports are described in eight pages of detail.

The American Institute of Aeronautics and Astronautics gets five pages. DOE reports get another five pages; their value is described as follows:

DOE reports include a large number which are concerned with research into nuclear energy, and which involve dual military–civilian uses. Examples are reactors of various types (including those used on ships); nuclear power systems used in space; research, development, testing, and production of nuclear weapons; laser nuclear fusion technology; isotope separation technology; production and control of nuclear material; nuclear material safety issues; personnel security issues; secret information security issues; export control issues; nuclear weapons control issues; nuclear power stations, etc. This portion of the reports continually gets a great deal of attention from those engaged in national defense S&T work in various countries, and it is a source of intelligence with great value.63

PRC collectors are given 13 pages of advice on how to exploit US military standards, which:

can be used for reference in formulating one’s own standards. Referring to them can yield an understanding and a grasp of the level of technology and developmental trends in US military industrial products. It can improve the level of one’s own research and design, promote the technical transformation of industrial enterprises, accelerate innovation and upgrade to new generations of products, improve operations and management, expand foreign trade and exports, and increase economic efficiency.64

The authors go through a list of open sources with military applications in the US and Britain, including publications of governments and private think-tanks, specialized books, periodicals, and conference papers. The last are described in detail according to type, format, organizations publishing them, their intelligence value, and problems collecting and exploiting them.65 On the subject of S&T periodicals we learn:

As an intelligence source, they are the first choice of rank-and-file S&T personnel as well as intelligence researchers. According to estimates, S&T personnel and intelligence workers obtain

60 percent and 80 percent, respectively, of their S&T intelligence from periodicals.66

According to the authors, in 1985 CDSTIC subscribed to 1,022 of these foreign-language periodicals. Some 45 of the top 56 foreign journals (measured by use) were from the US and Britain. Of the top 60 (measured by their value to China), 53 were US and British journals.67

Huo and Wang go on to describe the IBM-based “strategic intelligence database system” used by CDSTIC in the 1980s to process all of these data:

The strategic intelligence resource database system was designed to meet the needs of the China National Defense S&T Data Center’s intelligence researchers in their strategic research on weapons facilities development, and the needs of information collection operators in developing strategic information resources. Its purpose is to help intelligence researchers find the information they need for their own research subjects as quickly as possible; and to help information collectors to study, understand and master the circumstances, special features and publication rules of foreign strategic information resources.68

There are sub-databases with information on foreign “organization names, addresses, cable and telephone numbers, histories of changes and developments, nature of missions, distinguishing features, leadership organizations, financial circumstances, primary activities, publications and databases.” A sub-system on foreign persons contains:

given names and surnames, gender, biographical notes, work and home addresses, occupations, achievements, writings, range of primary activities, recent work circumstances, and whether they have visited China. It can display or print out all information according to name and whatever the database contains according to special subject category numbers, and the circumstances of all persons engaged in any special activity.69

So China’s open-source collectors not only provide information on foreign technology, they also serve up profiles of foreign organizations and persons with access to this information, for use by other PRC intelligence services.

Huo and Wang make a few other points that bear on our present study. They support our own observation that open-source intelligence in China is library-based. In their words: “Information work is a spin-off of library science. It has benefited significantly from library science. Methods such as indexing and abstracting are very successful because they are the foundation of library science.”70

They also confirm in a roundabout way that Chinese researchers are expected to look first at foreign projects before starting their own projects. In a passage on “obstacles” to the use of foreign information, Huo and Wang note that there are people in China who do not believe copying, benchmarking, and reverse engineering are the right way to do science:

In principle, the advance and development of science and technology are the true desire in achieving China’s economic vitality. In reality, this has not been universally accepted by society. There are many important people in society who consider intelligence and information work to have a “supplemental” status and function in China’s economic construction.71

In other words, there are Chinese scientists who prefer to do their own research and we suspect many of them ended up emigrating abroad. This does not mean they do not benefit China nonetheless, as we demonstrate in later chapters.

Finally, after several tortuous pages discussing the meaning of qingbao and the differences between “information” and “intelligence,” the authors come clean with an innocuous admission that “There are similarities between what we refer to as ‘information’ and what the foreign intelligence community refers to as intelligence work.”72

We examine these nuances toward the end of this chapter. Meanwhile, thanks are due to Huo and Wang for a wonderfully detailed snapshot of one part of China’s open-source intelligence system as it looked three decades ago. And for all their complaints about its inadequacies, it is – on an international scale – a model then and now of how open-source intelligence should be done. In the remainder of this chapter we look at how the system is constituted today. 

ISTIC and the structure of open source in China

China’s apparatus for exploiting foreign science is complex and redundant, as one would expect of a system in use for half a century. The country’s size, the paramount role of mimicry in its S&T development, and bureaucratic inertia make this proliferation of organizations inevitable. That said, there is a logic to its “design” that invites admiration.

There are five distinct levels of “S&T information organizations” in China today, including a dozen or so national organizations (depending on how they are counted); another 89 attached to technical ministries and State Council committees; 303 S&T information bodies at the provincial and municipal levels; about 3,600 information offices in company units (“basic cells”); and some 400 websites nationwide for exchanging S&T information. These figures have remained roughly constant for two decades, although the number of websites continues to grow.

Another taxonomy divides these organizations by type, including:

(1) Professional S&T information management organizations. (2) Comprehensive and specialized S&T information enterprises. (3) Local and specialized S&T information networks. There are also (4) document centers engaged in important S&T information activities, S&T libraries, and universities and training centers that offer majors in information [management], along with state, collective and privately run S&T information organizations. Finally, there are (5) society-based resources engaged in information activities, such as personnel involved in information [gathering] as an adjunct to their regular jobs, retired S&T personnel, and translators. Types 1–4 are main components of the “national S&T information system” and type 5s are supplementary resources.73

No matter how you slice it, there is a huge number of organizations and people in China today that collect, analyze, and communicate foreign S&T information to state, military, and corporate consumers. Our focus for the rest of this chapter is on the national bodies, which are:

1. The National Science and Technology Library  , NSTL), a virtual organization formed in 2000 that encompasses four major “libraries” including ISTIC.

2. China Defense Science and Technology Information Center  

 , i.e., Huo and Wang’s organization, of which the Beijing Document Service  is a part.

3. Patent Documentation Library and its adjunct China Patent

Information Center .

4. National Library of Standards , part of the China National Institute of Standardization .

The NSTL has four divisions for basic science, engineering, agriculture, and health. They are:

1. The Chinese Academy of Sciences’ National Science Library 

 .

2. National Engineering and Technology Library  ,

NETL), which like NSTL is a virtual organization.

3. Library of the Chinese Academy of Agricultural Science  

 .

4. Library of the Chinese Academy of Medical Sciences  

 .

The China National Institute of Standardization and the National Institute of Metrology   are credited on NSTL’s website as joint sponsors but are functionally independent and treated separately here.

Drilling down further, the NETL itself comprises four major components:

1. The Institute of Scientific and Technical Information of China 

 .

2. China Machine Industry Information Institute  .

3. China Metallurgical Information and Standardization Research

Institute  .

4. China National Chemical Information Center  .

These institutes are depicted in Figure 2.1. We omit a few supporting organizations that are discussed toward the end of this chapter.

 

FIGURE 2.1 China’s Foreign S&T OSINT Organizations

The National Science and Technology Library, one of the four top-tier elements, was created by China’s State Council in June 2000 as “the authoritative Chinese repository and service center for S&T document information and resources.”74 A virtual organization, it has no holdings of its own. Its purpose is to direct and coordinate the activities of member libraries. The NSTL’s main responsibilities are:

to exercise overall planning and coordination, collect in a more integrated way domestic and foreign document and information resources, draft data processing standards and norms, build S&T document databases, utilize modern networking technology, provide multi-level services, promote the joint creation and sharing of S&T document and information resources, organize in depth development of S&T document and information resources and their digitized application, and expand cooperation and exchange between China and the world.75

Member libraries manage their own operations according to guidelines laid down by the NSTL board, which is made up of scientists, open-source specialists, and library representatives. Actual leadership is vested in the Ministry of Science and Technology.76

NSTL also runs a comprehensive distribution network called the

“National Internet-based Science and Technology Information Service

System” (NISS) that makes the holdings of member libraries available to the public round the clock. As of 2004, its online retrieval services offered access to some 26 million items, including 4.9 million foreign journals; 1.8 million foreign S&T conference papers; some 900,000 foreign dissertations, reports and standards; and about 11 million foreign patents. Subscribers get navigational support for “important science and technology websites from all over the world.”77

Beyond its own (vertical) hierarchy, NSTL has resource-sharing agreements with the horizontal components of the system – CDSTIC and the patents and standards libraries – along with major PRC libraries outside the defined system, including the National Library of China, the China Academic Library and Information System (CALIS),78 and the Shanghai Library – a merger of the former Shanghai Library and Shanghai S&T Information Institute  , one of many such composite S&T information organizations in China that were consolidated over the past two decades.79

As illustrated in Figure 2.1, NSTL’s four member libraries are the

National Science Library, the National Engineering and Technology Library, and libraries for agriculture and medicine. The latter two are not within the scope of the present study, while the former two are central to it.

The National Science Library, also known as the Library of the Chinese

Academy of Sciences and the CAS Document Information Center 

  ,80 is managed by a board of trustees appointed by the CAS. It has a main library in Beijing, with branches in Shanghai, Wuhan, Lanzhou, and Chengdu;81 120 document information offices at R&D institutes throughout China; smaller collections attached to factories, schools, and publishing houses; and some two dozen specialized information networks.82

The NSL regards itself as NSTL’s “key member” and the “national reserve library for natural sciences and high tech literature.”83 Some 70 percent of its holdings are in the fields of mathematics, physics, chemistry, astronomy, geology, biology, electronics, computers, semiconductors, and nuclear energy.84 In his 2006 study Chen Jiugeng called NSL and its branches China’s “basic science information resources system” and credited it with 6.3 million books, 21.8 million issues of periodicals, and 13.4 million “other types of documents” in natural science, basic science, and new and high technology.85 NSL itself claims:

a collection of about 11.5 million items. In recent years it acquired or developed more than 30 databases, covering over 5,000 foreign science, technology and medicine full text journals, 11,000 Chinese full text journals, 80,000 foreign theses and dissertations, 180,000 e-books, and an increasing number of full text proceedings and reference books, all accessible from 89 CAS institutes over 24 cities across China. NSL provides an interlibrary loan system connecting every CAS institute, and connecting to NSTL and major academic libraries, delivering documents within two working days from a pool of more than 20,000 foreign journals.86

NSL branch libraries serve regional needs. The library in Lanzhou, a backwater notable for its proximity to China’s nuclear weapons industry, collects “special document information” and conducts “deep-level studies” for local research institutes. It has some 50,000 visitors annually and fields more than 100,000 queries a year from its 1.7 million document holdings. Its specialties include chemistry, chemical engineering, nuclear science, applied mathematics, and computer science. Unspecified “strategic information research”   is also done by the branch on demand.87

The engineering counterpart within NSTL to the National Science Library is the National Engineering and Technology Library (NETL), an umbrella organization for the China Machine Industry Information Institute,88 China Metallurgical Information and Standardization Research Institute,89 China National Chemical Information Center,90 and those offices within the Institute of Scientific and Technical Information of China concerned with engineering. Chen credited the NETL in 2006 with more than 7,000 types of foreign-language periodicals and with providing online access to some two million foreign and domestic documents.91

NETL’s aggregated holdings are in electronics, automation, computers and networks, materials science, environmental science, aerospace, biotech, energy, transportation, architecture, hydraulics, and other engineering fields in general. Beyond its five million books, the library has been adding some

4,200 foreign-language S&T periodicals annually, covering all the world’s key engineering technology journals. It also holds over 100,000 foreign conference proceedings and 114,000 foreign academic theses. Its collection of nearly two million US government S&T reports is augmented annually by 20,000.92

This brings us to ISTIC. As we noted earlier in the chapter, the Institute of Scientific and Technical Information of China was established in 1958 and grew to become China’s premier organization for gathering and disseminating foreign technical information in fields that have a direct impact on PRC S&T development. While part of the NETL framework, and a component within the national S&T information infrastructure, ISTIC is NETL’s largest member, predates it by decades, and is responsible for developing it. The reality of the relationship is depicted on ISTIC’s website, where NETL’s name appears under ISTIC’s logo.93

ISTIC in fact views itself as “the center for managing and providing services in the field of S&T information for the entire country” and as “exercising leadership over and serving as a model for the entire country’s S&T information system.”94 ISTIC at present has 850 employees in seven functional divisions and six “public-good” departments. It owns three corporations including the digital S&T document provider Wanfangdata. Its business areas are data research and analysis in support of government decision-making; S&T information services; research, development and propagation of new technologies and service platforms; fostering talent in the field of S&T information; and media publication services.”95 More specifically, ISTIC:

processes and reports on domestic and foreign S&T publications including documents, translations, reference works, reports, and research; builds domestic and foreign document databases that conform to China’s needs and circumstances; does research and analysis on domestic and foreign S&T sources that pertain to China’s national economic and S&T issues; reports on domestic and foreign S&T achievements and trends; performs strategic information services for [government] policy making departments; provides in a planned fashion specialized information services for the nation’s many priority science research programs; does research on information science, policy, management, service, methods and research; and develops international cooperation and exchange in S&T information work.96

ISTIC’s “service center” has two sections for “searches” and “special services.” The first is an online facility for searching full text, abstracts, TOCs, and web resources within five media categories including print, AV, web-based (e.g., ISI Derwent Innovations Index, ProQuest Science Journals, LexisNexis), “other” web media (e.g., Lawrence Erlbaum Online Journals), and CDs. Print media offerings include “western language conference proceedings, western language periodicals, foreign language S&T reports, Chinese language conference proceedings, and a Chinese language academic theses database.” The last three categories – web through CDs – provide access to 110 sources, only three of which are Chinese; the remaining 107 are foreign.97 Under “special services” ISTIC:

obtains and delivers for registered clients original documents from its own and outside  holdings; verifies the originality of S&T content for customers setting up research programs; appraises S&T achievements, does evaluations, and applies for patents; offers document research services and advice to central Party, government, and military leadership organizations, to state key lab production units, and to clients at large; verifies the recorded content and citations of papers and works publicly available through Chinese and foreign search engines; arranges proxy searches and loans of resources in domestic and foreign libraries and information organizations; and performs other personalized services.98

The Institute does more than collect and provide science and technology materials to domestic customers. It actively supports state R&D projects through “comprehensive, policy-driven strategic research” on the latest worldwide S&T achievements and trends for leading government departments.99 Here is an example from ISTIC Director He Defang, in a celebratory article he wrote for the industry trade journal Zhongguo Xinxi Daobao:

According to He, ISTIC over the course of three years provided a Beijing design bureau working on a maglev train documents in English, French, Japanese, German and Russian from which they “found answers that solved many key problems in their R&D” and reduced the time needed for experimentation. Decisions were made by “seeking truth from facts” (foreign facts). By using foreign resources, China’s researchers reduced their costs by 40–50% and their time by 60–70%. Thanks to ISTIC, China can “stand on the shoulders of giants” to reduce the risks of innovation. By studying trends in worldwide S&T, he continued, China can distinguish hot areas of research from “empty” pursuits, and get a grip on competitors’ technologies and patent strategies.100

ISTIC’s Chongqing branch was established in 1960, in the heartland of China’s high-tech military industrial region. Its responsibilities, according to a Chinese wiki source, are to collect materials on new technology and practical technology of benefit to China’s southwest; edit, translate and publish periodicals with S&T information; create databases for documents on energy, computer and information technology, and for facts on S&T developments in the former Soviet Union and Eastern European countries; compile an index of foreign S&T periodicals in the southwest region; provide search services; perform research assigned by the state; provide information consulting and fixed topic services; and organize training in information work.101

The branch holds 310,000 books; 15,000 foreign language S&T journals; some 7 million patent records; 190,000 documents on standards; 64,000 product user manuals; and 28,000 technical manuals.102 Following ISTIC’s launch of its commercial venture Wanfangdata, the Chongqing branch in 1993 transformed its former Database Research Center into the for-profit digital S&T service provider CQVIP  .103 There are ISTIC knock-

offs in all PRC provinces and major cities that use the suffix of ISTIC’s old   or new   names. This arrangement is unique to

China; we know of nothing similar elsewhere.

Other top-tier S&T information providers

Three major organizations outside the NSTL system support China’s access to foreign S&T information. They are the China Defense Science and Technology Information Center  , CDSTIC), the Patent Documentation Library  , and the National Library of Standards 

 . Each is important in its own way.

CDSTIC, previously the Commission on Science and Technology

Industry for National Defense 104 S&T

Intelligence Bureau  , is focused on foreign defense-related science and technology information. Being a military technical intelligence unit, openly available information is sparse, although we were acquainted with its workings by Huo and Wang. The Center’s functions and relationship to other S&T units were outlined in a 1984 document entitled “Regulations on National Defense Science and Technology Information Work”:

The national defense S&T intelligence work system is made up of the following components: the

Commission on Science and Technology Industry for National Defense (abbreviated below as COSTIND); each defense industry ministry (including the Ministry of Electronics Industry, the China State Shipbuilding Corporation, etc.); the PLA’s General Staff Department and General Logistics Department; related departments from each branch of the military services; organizations responsible for S&T intelligence work within the national defense S&T industry offices of each province, autonomous region, and municipality directly under the central government; national defense S&T intelligence professional working units at each level; local national defense S&T intelligence service centers; and the defense S&T intelligence network.105

The proliferation of military S&T intelligence units at each level is notable. The document goes on to name CDSTIC’s predecessor as the main unit within this network.

COSTIND’s S&T Intelligence Bureau is the integrated center for national defense S&T intelligence. Each of the national defense industry departments’ S&T intelligence bureaus is an S&T intelligence center within the system. The S&T intelligence offices directly subordinate to related departments and bureaus in the General Staff Dept., General Logistics Dept., and each branch of military service should gradually develop into S&T intelligence centers within the system.106

Article 9 outlines the military S&T centers’ duties.

Each of the national defense S&T intelligence centers have the following responsibilities: collect, organize, and make available domestic and foreign S&T materials needed by the system; take responsibility for managing the systems’ series of S&T intelligence reports and organizing S&T intelligence exchanges; carry out analytic work and provide intelligence for policymaking and scientific research and production; write reports; research S&T intelligence theory, methods, and the application of modernized techniques; and provide guidance to the system’s units.107

Another section emphasizes the importance of S&T intelligence work to defense.

Collection of S&T materials is the foundation for the defense S&T intelligence enterprise. Each department must actively, systematically and on a priority basis collect and utilize relevant domestic and foreign national defense S&T intelligence materials, including AV material and materials from professional and academic conferences. Scientific management of the materials is needed along with an integrated indexing system. Duplication should be avoided by sharing between units.108

Foreign copyright owners will no doubt appreciate that advice. Other notable provisions are requirements for national defense S&T intelligence units to invite subject matter experts in the relevant disciplines “on leave or retired” to participate in their S&T intelligence work (Article 24) and for annual increases in the defense S&T budget (Article 26).

A more recent source on CDSTIC lists similar tasks, such as building information resources, developing and using S&T information, managing the information, training individuals in the field of S&T information, integrating information organizations, providing policy consultations on weapons and equipment modernization, and safeguarding S&T information. The Center has supported China’s nuclear weapons, ballistic missile and satellite programs  , Yinhe (Galaxy) military supercomputer, and 863 program with its library of 38 million articles and 80-odd databases holding over four terabytes of information gleaned from American, Japanese,

Russian, and British publications, military reports and standards.109

Tied closely to CDSTIC is the Beijing Document Service  ,

BDS), founded in 1978 with cooperation from the Beijing Science and Technology Association  . CDSTIC provides most of BDS’s funding and is also its heaviest user. In addition to its own document holdings, BDS is the network administrator and service center for

COSTIND’s China Engineering and Technology Information Network (CETIN), a one-stop shop for foreign military technology information.

For example, CETIN provides a hotlink to a “(sub)network on foreign quality and reliability information”  that gives detailed information on standards, test results, and maintenance issues of foreign (mostly American) military equipment.110 Another link leads to “foreign military engineering and equipment”  ,111 while a third labeled “C3I” takes you to “electronic countermeasures”  .112 One can do full text searches or look for particular military technologies, such as mine detection, camouflage, and bridging devices.

CETIN periodically highlights foreign military technologies of interest. On January 11, 2010 the site featured Northrop Grumman’s AN/ALQ-162 countermeasures set. Several other American ECM devices were described in great detail. There was a “pictorial description of the American national missile defense system”   and a side bar with links to 41

“specialized areas” such as radar and radar countermeasures, laser weapons, computer warfare, network centric warfare, stealth and anti-stealth technology, and GPS systems, all in the Chinese language and all focused on foreign – especially American – systems.

As valuable as information on existing foreign equipment is to China’s technology planners, equally important are the insights gained from foreign patent exploitation. Although China did not enact its own patent law until 1985, it was busy exploiting foreign patents some 30 years before that. Indeed, according to a 2005 article celebrating 50 years of patent work, China’s S&T workers early on adopted the mottoes “Scientific research work cannot be separated from the support of foreign patent literature” 

and “Patent literature is a treasure trove for

science and technology” .113

Recognizing these facts, the CAS National Science Library set up a Patents Section   in 1956 and started gathering foreign patents literature for use by China’s S&T workers. In 1965, ISTIC set up a Patents Library  , which took over the NSL Patents Section’s holdings and “began continuously expanding the scope of its foreign patents literature collection, turning patents literature into the main source of China’s S&T information.” The S&T information offices that were springing up all over China at that time also gathered foreign patents literature and made it available to local patrons.114

In 1980, the China Patents Office   was formed, which the following year took over ISTIC’s Patents Library and renamed it the Patent Documentation Library  . It retained this name after the Patents

Office was reborn in 1998 as the State Intellectual Property Office 

 , SIPO) and is now China’s largest and most complete

repository and service center for patent documents and information.

The Patents Library provides “multi-faceted, multi-level, multi-channel” patent document and information services, including collecting, reading, consulting, searching, and training. Patrons begin work in the reception hall, where they register, consult with library staff, and learn how to use the search facilities. The patent search room has 70 computer terminals that customers use free of charge. Half of them search the library’s holdings and the other half are linked to international patent databases. The stacks hold 3,760 shelves of patent specifications from 28 countries. The Library also offers consignment services, including patent reproduction, searching, and analytic services delivered personally or by mail, phone, fax, email or online.115 SIPO in addition operates 47 local patent information centers across the country.

Finding space for its massive holdings has been a problem, as the following comments suggest:

Workers at the Patents Library have been overworked trying to manage the truckloads   of foreign patent specifications that arrive. Their biggest headache was finding room for it all. Today the paper documents have been largely replaced by CDs and computers. Although the number of customers has grown, most people access the library’s services remotely. Library workers spend less time finding and copying documents and more time helping customers analyze the patents, spot trends, and find loopholes to avoid patent fees, for example, by researching how a product’s patent is treated by other countries.116

Challenges caused by volume led to the establishment of an “Automated Work Division”   within the Library and in 1993 to a separate China Patent Information Center   also subordinate to SIPO. The new center is charged with building SIPO’s automated systems; processing, disseminating, searching and providing consultancy services for patents and other IPR information; “exhibiting, mediating, developing, implementing, using patented technology and organizing test production of patented products”; and information engineering and related services. It has offices for searching and consulting, translation and information processing.117 Some 140 technical staff members are on call.118

Just as patents provide China’s S&T cadre critical insight into foreign technology, so do foreign technical standards offer short cuts to development. The National Library of Standards  , which collects foreign technical standards and supports development of China’s own, was established in 1963. Prior to that, its functions were carried out by a Library of Standards within ISTIC. In 1978, it became part of a new Institute for

Standardization and Integration   within the Office of National Standards  and in 1989 it was put under the State Bureau of Technical Supervision’s  Standards Information Center 

 . In 1999, its functions were transferred to the China Center for Standards Research , which was reborn in 2003 as the China National Institute of Standardization  , the present parent of NLS.119

The Library supports research by the Standardization Administration of

China   established in 2001 to exercise “unified management, supervision and overall coordination of standardization work.”120 NLS’s parent body CNIS is affiliated, in turn, with the General Administration of Quality Supervision, Inspection and Quarantine of the PRC  . CNIS has overall control of foreign standards research, as suggested by the following:

The main responsibilities of CNIS are to conduct all-round, strategic, and comprehensive research of standardization during the development process of economy and society, to research and develop comprehensive fundamental standards, as well as to provide authoritative standards information services.121

SAC, on the other hand, is China’s public face to the international standards community and is tasked to:

represent China in the International Organization for Standardization (ISO), International Electrotechnical Commission (IEC) and other international and regional standardization organizations; organize the activities of the Chinese National Committee for ISO and IEC; organize domestic sectors and local areas to participate in international or regional activities on standardization; sign and implement international cooperation agreements in standardization, and examine, approve and implement international cooperation and exchange projects on standardization.122

Information on foreign and domestic technical standards is provided over the China Standards Service Network  , owned jointly by NLS and a National Standards Document Sharing Service Platform  

 123 The “platform” was built by CNIS to collect, edit, and publish information on standards, facilitate standards research, provide full text delivery of standards documents, and perform other online services, such as consulting, training, and translation. Some 1.3 million titles are available.124

Other organizations are also involved in exploiting foreign standards. A directory posted to the “measurement forum”   website125 lists names and contact information of offices that provide information on particular categories of standards. Many of the offices are inside the technical ministries (shipbuilding, aviation industry, etc.), including the

Ministry of Electronics’ Fourth Institute for Standardization, which handles “US national military standards.”126

Foreign standards are also scrutinized by China’s National Institute of

Metrology  , which like CNIS is subordinate to the General Administration of Quality Supervision, Inspection and Quarantine. The value of foreign technical standards to national development is acknowledged by the Institute:

NIM participates actively in international and regional key and supplementary comparisons to ensure our national standards’ equivalence and their traceability to the SI, and provides the country with a very important technical basis for her involvement in international trade and economic globalization.127

NIM sits at the top of the standards chain ensuring that the tools and processes used to model foreign specifications are up to the task, so that “Chinese” products work and are marketable internationally. A key purpose of the Institute’s new campus that opened in 2009 is to attract “elite” scientists for “global collaboration” to “serve national strategies.”128 In China, foreign experts help design processes that ensure foreign technology is faithfully copied. 

The professionalization of open-source intelligence

Chinese exploitation of foreign open-source S&T information is an allconsuming, systematic endeavor aimed at accelerating China’s scientific development. Transforming open-source S&T information into useful intelligence is pursued not just to keep tabs on foreign threats, as is done by western services, but to circumvent the cost and risk of indigenous research. Another distinguishing characteristic of Chinese open-source operations – beyond the scale itself – is the diversity of venues through which information is gathered, from scanning technical literature to analyzing patents, reverse engineering product samples, and capturing conversations at scientific meetings. Nothing is overlooked.

In intelligence circles the term “open source” connotes information obtained for fee or free without violating legal statutes. But there is a huge difference between what is needed to openly monitor foreign technical programs and the apparatus needed to covertly model foreign R&D as the basis for one’s own development. Not only does China invest far more effort in open-source collection than other countries, the “back-end” components – analysis, customer interaction, and feedback to collectors – also play a much larger part, as befits a nation whose progress depends more on adaptation than innovation.

This qualitative distinction between the roles of open-source intelligence in the PRC and in other countries is also apparent in the professionalization of China’s S&T open-source cadre. Whereas western services typically regard open source as a poor cousin to “real” (clandestine or technical) intelligence, China staffs its OSINT129 organizations with top-line career personnel, backed by an industrial organization with its own trade journals.

We know of no other nation where open-source intelligence enjoys this level of support.

As noted above, the China Society for Scientific and Technical Information – the state-backed professional body for open-source S&T workers – emerged in 1964 as an “academic and non-profit institution” to promote open-source S&T exploitation.130 All of China’s national S&T information organizations are corporate members. CSSTI has branches in every province and major city of China as well as 11 committees to represent technical and other disciplines.131

CSSTI’s tasks are to promote open-source intelligence research, provide consulting and other services “to meet various information requirements of the nation,” strengthen links between the S&T intelligence network’s central and local units and between the organizations and their individual members, and acknowledge outstanding personal achievements in S&T intelligence – in essence to create a “home” for China’s S&T information workers.132 Information on open-source S&T tradecraft is shared at annual meetings and through its periodical publications: the Journal of the China

Society for Scientific and Technical Information   and the China Information Review  .

Organizationally, CSSTI is a component of the China Association for Science and Technology  , an alleged NGO that plays a major role in facilitating foreign S&T technology transfers (see Chapter 5). CSSTI’s headquarters are located within ISTIC’s building at 15 Fuxing Road in Beijing. The Society is chaired by a Party figure from the science ministry but is run by ISTIC’s incumbent director. Board members include representatives from ISTIC, ISTIC’s quasi-independent Shanghai chapter, the National Science and Technology Library, the Academy of Machine

Information , and the Department of Electronic Information  within the PLA’s General Armament Department  133

CSSTI also manages the Society of Competitive Intelligence of China 

 , one of its 11 disciplinary committees  . SCIC

claims 400 corporate and more than 800 individual members from among “China’s 20,000-plus intelligence research and information consulting person-nel.”134 “Competitive intelligence” organizations worldwide typically focus on business intelligence in general. In China the emphasis is on S&T intelligence as indicated by SCIC’s subordination to two S&T organizations (CSSTI and CAST).

Formed in 1995, SCIC studies the theory and practice of competitive intelligence, popularizes its techniques, offers consulting services, helps companies gain competitive advantages, cooperates and exchanges information with other professionals worldwide, and protects the legal interests of competitive intelligence professionals.135 It holds annual meetings, training sessions, and has its own journal, Competitive Intelligence  . Beyond its connections within China, SCIC “has established a long-term partner relationship of cooperation and exchange with America’s Society of Competitive Intelligence Professionals.”136

We conclude this chapter on Chinese open-source

“information/intelligence”   with a few words on the etymology of this italicized term, which is problematic not only for translators but for Chinese as well, who are aware of its different connotations in translation. “Information” has neutral or positive associations, while “intelligence” to those outside the field conjures up a negative cloak-and-dagger image. We have resisted the impulse to color our readers’ perception by using one term or the other in translation and rely wherever possible on China’s own English usage, for nomenclature especially.

The problem with qingbao (Japanese: jōhō) is this: not only does the Chinese term cover multiple English words but the Chinese themselves use it inconsistently. It is not simply a case of polysemy (e.g., “intelligence” in English, the distinct meanings for which Chinese have no trouble sorting out). The concept denoted by qingbao is broad enough to encompass both “intelligence” and “information.” And if you focus on the basic meaning, without its emotive nuances, there is not much difference between the two, from a layperson’s view especially.

As a result, use of the term qingbao varied for decades in China’s intelligence community. According to Cheng Jixi, professor at the Beijing

Science and Technology University, in the mid-1950s China built a “system for collecting, translating, and studying foreign S&T materials,” i.e., the apparatus we described above. Needing a name for it, China translated the Russian word for “information” with Chinese “qingbao,” which was used in China to depict the full range of activities from clandestine collection (espionage) to straightforward open-source monitoring to one’s particular knowledge of events.137 By the 1970s, however, China had also adapted the word “xinxi”   to translate the English word “information.” So “information” ended up with two Chinese words that were used interchangeably.

Some Chinese open-source workers saw qingbao as a subset of xinxi meaning “processed or distilled information” and believed qingbao should be used exclusively for “intelligence”.138 Huo and Wang adopted this position and their usage eventually became the norm. Interestingly, it also parallels how western services treat the two terms today. In Huo and Wang’s words:

There are currently a considerable number of people who appreciate the fact that, even though information is the primary source and basic vector for intelligence, information nevertheless is not intelligence. One must go through a catalyzing and activating process in order to extract intelligence from information.139

More exactly:

Information is knowledge put into material form or symbols. Intelligence is knowledge needed to solve specific problems that has been extracted from information. Information is the source from which intelligence is extracted.140

As such, “information”  is the target of open source collection and “intelligence”  is its output:

The primary task of collection work within the broader modern S&T intelligence setup is to collect information that serves as an intelligence source, and to provide raw knowledge for intelligence analysis and synthesis work.141

More recently, ISTIC’s Chen Jiugeng affirmed Huo and Wang’s distinction, but added an important nuance:

“Qingbao” (intelligence) is a special class of “xinxi” (information) used to keep secrets that pertain to national security, social stability, and business competition. “Qingbao” includes both “original”   and “processed”   “information”  . And processed information does not necessarily become “intelligence”—that depends on one’s goal and intended use.142

That is, “information” becomes “intelligence” depending on who uses it for what. Another factor governing choice is psychological impact, which played a role in the State Science and Technology Commission’s 1992 decision to change ISTIC’s and other S&T agencies’ names from “qingbao” to “xinxi” despite the nature of these organizations. As Chen put it:

Eliminating the sensitivity and suspicion that the public – especially in Taiwan, Hong Kong, and among overseas Chinese – has toward the old practice of calling publicly shared and used S&T information “S&T intelligence” will help expedite the spread of knowledge, technology exchanges and foreign cooperation.

The nuance is important. It matters to China’s foreign supporters whether they simply share “information” with the PRC or support that state’s “intelligence” efforts. Similarly, whether foreign firms with research labs in China are “cooperating” or “betraying” their own nation’s interests depends on which end of the stick you hold. In the following chapter we address the issue of foreign high-tech investment in China and its implications for the transacting partners.

Notes

1 John Lewis and Litai Xue, China Builds the- Bomb, Palo Alto, CA: Stanford University Press, 1988, pp. 223-224.

2 Guan Jialin   and Zhang Chao   ("Review and

Outlook for Scientific and Technological Information Undertaking of China"),  

(Qingbao Kexue), 25, no. 1, January 2007, p. 2; Chen Zeqian   and Bai Xianyang 

  ("The Locus of Development for China's S&T Information Enterprise"),   (Xiandai Qingbao), December 2007, p. 12.

3   ("1956-1967 Long-term Plan for the Development of Science and Technology"). www.cdstm.cn/?action-viewnews-itemid-14784-page-l.

4 Guan and Zhang, "Review and Outlook for Scientific and Technological Information Undertaking of China."

5 Miao Qihao, "Technologial and Industrial Intelligence in China: Development, Transition andPerspectives." In Prescott and Gibbons:, eds, Global Perspectives on Competitive Intelligence, Alexandria, VA: Society of Competitive Intelligence Professionals, 1993.

6 ISTIC's: Chinese name was changed again in 1992 from   to 

. The rationale for the shift from   (qingbao,

intelligence/information) to   (xinxi, information) is discussed later in the chapter.

7 Guan and Zhang, "Review and Outlook for Scientific and Technological Information Undertaking of China." In 1959, the university became a department within the University of Science and Technology of China  

8 Ibid.

9 The magazine was entitled   ("Instructional Materials on S&T Information Work").

10 Guan and Zhang, "Review and Outlook for Scientific and Technological Information Undertaking of China."

11 Ibid.

12 Chen and Bad, "The Locus of Development for China's S&T Information Enterprise."

13 Miao Qihao, 1993, p. 51.

14 Guan and Zhang, "Review and Outlook for Scientific and TechnologicalInformation Undertaking of China" and Chen and Bai, "The Locus of Development for China's S&T Information Enterprise."

15 Guan and Zhang, "Review and Outlook for Scientific and Technological Information Undertaking of China."

16 1978–1985   (Regulations on National S&T Development 1978–1985).

State Science and Technology Commission, 1977.

17 Guan and Zhang, “Review and Outlook for Scientific and Technological Information Undertaking of China.”

18 Chen and Bai, “The Locus of Development for China’s S&T Information Enterprise.”

19 Guan and Zhang, “Review and Outlook for Scientific and Technological Information Undertaking of China.” The authors claim that half of the 1,038 databases operated by China’s 64 state-level ministries and committees, and 27 provinces and autonomous regions, from the early 1980s to mid-1990s were S&T information databases.

20 Chen Jiugeng, “Actual Strength of S&T Information Service System in China,” China Information Review, no. 10, 2006, pp. 17–22.

21 Qihao Miao, 1993, pp. 49–53. There was also a large (but indeterminate) number of S&Tnovelty searching centers   throughout China to guide technical projects by foreign literature retrieval; see Guan and Zhang, “Review and Outlook for Scientific and Technological Information Undertaking of China.”

22 Paraphrase of SSTC document   (“Opinions on

Restructuring and Strengthening National S&T Information System Document Work”), January 1989.

23 The expansion is:  . ISTIC is now called   its

English name remains unchanged.

24 The S&T Intelligence Bureau is now the China Defense Science and Technology Information

Center  . COSTIND expands to “Commission on Science and Technology

Industry for National Defense.” Further down in the notice the Science Commission assigns coverage of US government AD and NASA documents to this unit. “AD” means “Armed Services Technical Information Agency Document.”

25 Now called the Patent Documentation Library   and the China Patent Information Center  , both part of the State Intellectual Property Office  

26 Now the China National Institute of Standardization, National Library of Standards 

 

27   (“Opinions on Restructuring and Strengthening

National S&T Information System Document Work”), January 1989. Huo and Wang’s description of the three levels and their respective functions matches the SSTC prescription. See Huo Zhongwen and Wang Zongxiao, Sources and Methods of Obtaining National Defense Science and Technology Intelligence, Beijing: Kexue Jishu Wenxuan Publishing Company, 1991.

28 www.nstl.gov.cn/index.html.

29 Xia Chengyu   (“A New Role for Leaders of S&T

Information Departments under the New Circumstances”), in   (Science &

Technology Progress and Policy), 2001.1, pp. 104–105. Guan and Zhang also date the third

(modern) phase of the system’s development from 2001. See Guan and Zhang, “Review and Outlook for Scientific and Technological Information Undertaking of China.”

30 MOST statistics cited by Chen, “Actual Strength of S&T Information Service System in China.” Chen states the figures do not include institutes within the Aviation Industry Corp, China State Shipbuilding Corp, China National Nuclear Corp, China North Industries Group Corp, and other companies working on sensitive projects not suitable for public disclosure.

31 Ibid.

32 Ibid.33 Ibid.

34 Chen and Bai, “The Locus of Development for China’s S&T Information Enterprise.”

35 Bruce Gilley, “China’s Spy Guide: A Chinese Espionage Manual Details the Means by Which

Beijing Gathers Technology and Weapons Secrets from the United States,” Far Eastern Economic Review, December 23, 1999, p. 14. See also Huo and Wang, Sources and Methods of Obtaining National Defense Science and Technology Intelligence. An English-language translation by the US government completed in 2000 and verified by the present authors has been posted to <www.fas.org/irp/world/china/docs/sources.html>.

36 Huo and Wang, Sources and Methods of Obtaining National Defense Science and Technology Intelligence, p. 32. All citations refer to pagination in the translation.

37 Ibid. p. 5.

38 Ibid. p. 7.39 Ibid. p. 8.

40 The authors’ efforts to treat their topic scientifically sometimes border on banal. For example,on page 151 we learn that “Intelligence consumer studies refer to studies of the consumers of intelligence.” In the same spirit, the authors launch into an explanation of human needs that begins with Maslow’s hierarchy and ends with references to “Bradford’s Law of Grade

Distribution” and “Zipf’s ‘Least Effort’ Principle.”

41 “Research on consumer demand for intelligence is the basis for information collection and isone of the key research areas in collection science. Without consumer demand, collection loses its significance. . . . The difficulty with research on consumer demand lies in the fact that hardly any consumers are skillful or adept at communicating their real needs.” Ibid. p. 14.

42 Ibid. p. 68.

43 Ibid. p. 69.

44 Ibid. p. 19.

45 Ibid. pp. 42–44.

46 Ibid. pp. 57–58.

47 Ibid. p. 24.

48 Ibid. p. 72.

49 Ibid. p. 75.

50 Ibid.

51 Ibid. p. 78.

52 Ibid. p. 81.

53 Ibid.

54 Ibid. p. 82.

55 Ibid.

56 Cited in ibid. p. 85.

57 Ibid. p. 85.

58 Ibid. pp. 88–89.

59 Ibid. pp. 90–91.

60 Ibid. p. 93.

61 Ibid. p. 95.

62 Ibid. pp. 98–99.

63 Ibid. pp. 108–118.

64 Ibid. p. 120.

65 Ibid. p. 147.

66 Ibid. p. 182.

67 Ibid. pp. 184–185.

68 Ibid. p. 221.

69 Ibid. p. 224.

70 Ibid. p. 228.

71 Ibid. p. 53.

72 Ibid. p. 230.

73   (“National Science and Technology Information Development

Policy”), State Science and Technology Commission, January 1992. Accessed at: www.srstc.gov.cn.

74 “Home Page,” National Science and Technology Library, www.nstl.gov.cn/index.html.

75 Ibid.

76 Liansheng Meng and Yan Quan Liu, “The Present and Future of China’s National Science andTechnology Library: A New Paradigm of Sci-tech Information Resource Sharing,” New Library World 106, no. 7 (2005), 343–351.

77 Ibid.

78 CALIS is sponsored by the Ministry of Education for the entire range of academic disciplines.

It has centers at seven regional locations, plus a “Northeast Regional National Defense

Document Information Service Center”  . Established in 1998, the system promotes joint acquisitions and sharing through a digital network. Some 500 libraries participate (www.calis.edu.cn/calisnew).

79 Chen, “Actual Strength of S&T Information Service System in China.”

80 CAS Document Information Center   was its name from 1985 to 2006.

81 Tracing their name changes through time can be challenging. The Lanzhou Branch Library,founded in 1955 as the “LCAS Lanzhou Library”  , was renamed in 1957 the “CAS Northwest Branch Library”  , and renamed again in 1959 the “CAS Lanzhou Branch Library”  . During the Cultural Revolution it was known as the “CAS Gansu Provincial Library”  . In 1973 it was renamed the “CAS Lanzhou Library”  , and in 1987

redesignated as the “CAS Lanzhou Documentation and Information Center”    while continuing to use its former name the “CAS Lanzhou Library.” In

1997 it became the “CAS Scientific Information Center for Resources and Environment”   while retaining “CAS Lanzhou Library.” Today its website (www.llas.ac.cn) sports three names: CAS National Science Library, Lanzhou Branch Library

 , and the legacy names CAS Scientific Information Center

for Resources and Environment, and Gansu Province S&T Library  

82 www.hudong.com/wiki/  and Chen, “Actual Strength of S&T Information

Service System in China.”

83 www.las.ac.cn.

84 www.hudong.com/wiki/ 

85 Chen, “Actual Strength of S&T Information Service System in China.”

86 www.las.ac.cn.

87 Ibid.

88 www.gmachineinfo.com.

89 www.cmisi.com.cn.

90 www.cncic.gov.cn.

91 Chen, “Actual Strength of S&T Information Service System in China.”

92 www.istic.ac.cn.

93 Chinese normally order relationships from superordinate (on the top or left) to subordinate(toward the bottom or right).

94 www.istic.ac.cn.

95 Ibid.

96 www.chinabaike.com.

97 www.istic.ac.cn.

98 Ibid.

99 www.chinabaike.com.

100 He Defang, “As for Indigenous Innovation, Information Should Go Ahead of Rest,” China Information Review, no. 10, 2006, pp. 12–13. 101 www.hudong.com/wiki/ 

102 Ibid.

103 www.cqvip.com.

104 In 2008, COSTIND became the State Administration for Science, Technology and Industry for

National Defense   and part of the newly formed Ministry of Industry and Information Technology.

105   (Regulations on National Defense Science and Technology

Information Work), PRC State Council, July 1984.

106 Ibid. Article 8.

107 Ibid. Article 9.

108 Ibid. Article 13.

109 home.cetin.net.cn, accessed May 2006 and January 2010.

110 home.cetin.net.cn/storage/cetin2/QRMS/index.htm.

111 home.cetin.net.cn/storage/cetin2/xw/indexgcb/indexgcb.htm.

112 home.cetin.net.cn/storage/cetin2/xw/xxz/xxz.htm.

113 Wu Xuanzhou  , “50 Years of Patent Documents”  , in Zhongguo Faming yu Zhuanli, April 2005, qkzz.net/Announce/Announce.asp?BoardID=13200&ID=120156.

114 Ibid.

115 www.sipo.gov.cn/sipo2008/wxfw/.../t20080416_381338.html.

116 Wu “50 Years of Patent Documents.”

117 www.cnpat.com.cn.

118 www.sipo.gov.cn/sipo_English/news/official/200904/t20090417-453529.html.

119 Chronology reconstructed from information posted to baike.baidu.com/view/2473995.html and www.zjj315.gov.cn/b25.asp. NLS is also called the National Standards Document Center 

 .

120 www.sac.gov.cn.

121 www.cnis.gov.cn. Our italics. The terms are standard euphemisms for foreign collection.

122 Paraphrase of information on www.sac.gov.cn.

123 www.cei.gov.cn/homepage/gov/zgbzqbzx.htm.

124 www.cssn.net.cn.

125 www.gfjl.org.

126 www.gfjl.org/thread-14428-1-1.html. The information was reportedly posted on November 19, 2007 and was accessible as late as January 2010.

127 www.nim.ac.cn.

128 Ibid.

129 OSINT (Open Source INTelligence).

130 edu.istic.ac.cn.

131 Ibid.

132 www.cssti.org.cn (defunct as of November 21, 2009).

133 Ibid.

134 www.scic.org.cn.

135 Ibid.136 Ibid.

137 Cheng Jixi   (“Rethinking the Change of ‘Qingbao’ to ‘Xinxi’”) in   (Journal of Academic Library and Information Science), 1998.2.

138 Ibid.

139 Huo and Wang, Sources and Methods of Obtaining National Defense Science and Technology Intelligence, p. 10.

140 Ibid.

141 Ibid. p. 12.

142 Chen Jiugeng   (Journal of Intelligence) 19.1, January 2000.