Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
2. Part 2: Advanced applications
By Information Technology (IT) we mean the combination, first of software with computing equipment: computers and input/output peripherals (visual display units, sensors, actuators, etc.), and then this software/hardware computing system with human operators and users (as in administrative and end-user systems) and/or electro-mechanical gear (as in computer-integrated manufacturing, railway train traffic regulation, etc.).
By Advanced Applications we mean computing systems that apply state-of-the-art information technology to novel end-user problems. In this part, I will review state-of-the-art applications, while part 3 reviews state-of-the-art technology.
2.1 Background
The UNU/IIST seeks to be a bridge between the industrial and the developing world for new applications.
Several presentations at the Kyoto Symposium have outlined specific application technological means such as: databases, data banks, communications networks, electronic libraries, information retrieval, multi-media, cognitive (expert) systems, natural language processing, etc.
It can be foreseen that the UNU/IIST will attempt to organize advanced courses on: Database and Information Retrieval Systems; Knowledge- and
Expert-based Systems and Data Engineering; Visualization of Data, Hypermedia, Data Communication Networks and ISDBN, etc.
2.2 Specific Application Disciplines
Each of the following subsections addresses one of the topics covered by our colleague speakers. From these you may get the erroneous impression that the UNU/IIST will assist in transfer of material technology. The UNU/IIST will not be so occupied. The UNU/IIST will not advise the developing world on which hard, material technologies to adopt. Instead, the UNU/IIST will consult on the software technological consequences of most such hard technologies, will conduct technology management awareness courses, will train in their software components, will help develop these, and will educate in their development.
2.2.1 Database and Knowledge Systems
We speak here of a spectrum: from conventional "old-fashioned" databases - in which data ("knowledge") are represented in a form that does not readily permit the kind of reasoning usually experienced in deductive, inductive, and "abductive" systems- to knowledge bases where such reasoning is readily provided for.
In creating a database, one often imperceptibly transforms the knowledge gained, when analysing the application, into some efficient physical storage representation - thereby "compiling" away those knowledge facets that might be useful in heretofore unforeseen contexts.
In creating a knowledge base, such generally adaptable knowledge is preserved. Execution time is longer, but the time it takes for humans to formulate a new application and get it "running" is generally very short.
In either case (of information-base creation), the requirements developer spends a very long time analysing the domain of interest. In the former case, the knowledge gained is specialized with reference to the specific, usually narrow, foreseen application - and hence most is lost with reference to unforeseen, future, related applications.
Take an example: one can write a software system for the regulation (monitoring and control) of a railway system in either of three ways.
In the first, using the ALG approach, all knowledge about the specifics of railways is hidden in the code of the programs, in their compiled constants, and in the structure of the code itself, in the specific composition of statements.
In the second case, one represents the railway system implementation as a database: all the tracks, the trains, the schedule, etc., are data structures, and computation proceeds through interpretation. The railway system specific interpreter is, however, general in that the same interpreter, written once, can be used to handle any country's railway system. Execution is slower than in the former case, but one can easily introduce changes to the railway network, trains, schedule, etc.
The compiled, first version represents a so-called partially evaluated version of the second. In either case, the interpreter does not reason about train regulation (based on logic) but, typically, arithmetically computes based on classical mathematical laws of kinematics and operations research.
Instead of either of these two approaches, one could represent these laws in logical form - in addition to, but in the same form as, the representation of the railway system components. "Computation" now proceeds by means of a general meta-interpreter, also called an inference machine, the same for all kinds of systems and not only railway systems.
This "tri-partite" view is novel, not scientifically, but in being accepted in everyday industrial practice.
UNU/IIST wishes to propagate awareness of possibilities and of techniques relevant for all three cases. And the UNU/IIST will attempt to arrive at some kind of "didactics" and "pragmatics"-based classification criterion that advises the designer on which of the three design approaches to apply for given applications.
2.2.2 Communication Networks and Distributed Systems
Techniques for implementing secure and fault-tolerant communication networks and distributed systems include cryptographics and verified protocol designs. It seems inescapable to have any computing system being developed today involve networking and distribution.
Typical networks are either Local Area Networks (LANs) or Wide Area Networks (WANs), or are heterogeneous combinations thereof. LANs are typically based on fibre optics, whereas WANs typically involve radio telescope communication via satellites.
There is a tendency to vastly over-sophisticate, especially the WAN networks. Too many optional features and too general communication capabilities, for example, between any two ground stations involving only one intermediate satellite call usually take such systems economically out of the reach of most nations.
The National Informatics Centre (NIC), part of the Indian Government's Department of Electronics, has developed a low-cost WAN called NIC-NET. NICNET claims to connect about 500 computer centres all over India. There is a central, "expensive" computer centre, in New Delhi, and each of some 500 ground stations communicate via one satellite (India-Sat) directly with this centre. If the communication is intended for some other ground station, then the centre passes it on. NIC-NET then combines this network communication with seemingly reasonably homogeneous, conventional databases at each ground site. Together the network communication plus the databases form a typical distributed system.
It seems that NIC-NET offers a very interesting solution to a global problem, one that is usually frowned upon in the industrial world for being "too simple" (industrialists would claim), but one that might appeal to countries and regions of the developing world by being low cost.
The tendency to over-design networked, distributed systems is a typical result of engineering - of electronic engineers putting too much emphasis on getting maximum functionality and flexibility out of the hardware. The result is oftentimes that the software part of these systems is hard to get right and that the end-user is offered too many options.
The UNU/IIST wishes to propagate state-of-the-art techniques both for the design of sophisticated, not necessarily low-cost, as well as of simple, low-cost, secure, and fault-tolerant (distributed) network systems. The UNU/IIST will emphasize that systems that reflect user requirements, as it seems the NIC-NET does, are provably correct, secure, and fault tolerant, and that such systems offer flexibility through implementing user concepts, not by implementing computing systems concepts that the users must then combine, often in tricky ways, to achieve user functions!
We finally remind the reader of the desire of the UNU/IIST to provide for an electronic mail service network connecting (at least) all the UNU/IIST Organic Network affiliates (see last paragraph of the introduction to section 1.7). We have such "e-mail" nets to greatly enhance international cooperation - and I personally believe it extremely important that our colleagues, in academia and in industry in the developing world, be integral parts of such "gateway interconnected" networks.
2.2.3 Electronic Libraries and ISDN
Here we are talking about a special application area within information base systems - and hence the question arises as to whether such systems should be databased or knowledge oriented.
ISDN stands for "integrated services digital network." With the ISDN the possibility exists of freely mixing structured (tables) and unstructured (text) data, structured (drawings, diagrams) and unstructured (photos), still and live (video) pictures, and voice in any one communication.
Whereas the "narrower" issue of electronic databases may primarily be application driven, ISDN is a classical case of a technology- and monopoly-driven development. There still seems to be some international uncertainty with respect to the introduction and spread of ISDN.
The focus that the UNU/IIST could emphasize is that of software implications of the ISDN with respect to end-user applications. Once the ISDN becomes popular there will be a great need for a richly faceted software industry that can provide a multitude of applications without which the ISDN would not be economically feasible.
Again I remind the reader that the UNU/IIST eventually would like to establish a form of electronically accessible database in connection with its Software Catalogue (see subsection 1.4.5.). Establishing such a facility could eventually become a "carrier" for a UNU/IIST engagement in the ISDN.
2.2.4 Higher Education Management Information Systems
There seems to be a demand, voiced through Unesco, that the UN system help, specifically African universities, in jointly developing and installing Management Information Systems (MISs) for Higher Education (colleges and universities).
The application is a typical (LAN, and possibly WAN-based) distributed information planning, control, and base system. Typically such systems can principally be databased, rather than knowledge based, with no loss in generality.
The UNU/IIST is currently in the process of formulating for Unesco a proposal for the requirements development and software development of such a system. The UNU/IIST will propose that it coordinate a feasibility study, to be followed by a demonstrator project, and hopefully a prototype technology transfer product development project.
A university MIS usually enables the following functions: (i) budgeting (planning and approval) and accounting; (ii) personnel; (iii) student registration; (iv) course/class planning, scheduling, and registration; (iv) examination planning, monitoring, and control (including student and grade registration, external reviewer liaison, etc.); (v) research project planning, monitoring, and control; (vi) central and distributed library database; (vii) services management (purchasing, invoicing, facilities management, etc.); and so forth. A computerized university management system thus requires software that helps in planning, regulatory control, and information dissemination - hence it may contain both classically, i.e. formally developed, as well as conceptually developed components- all integrated.
The ambitious aim is to create, in one or more developing countries, on each of two or three continents (Africa, Asia, Latin America), specialized software houses whose expertise is exactly in such college and university MISs. The long-range hope is that such software houses, by being specialists in a very application-specific domain, can become competitive in the global market.
2.2.5 Education Technology
Much is being said about computing system learning and instruction technology: primary and secondary schools, high school education, trade school training, profession-oriented instruction, etc.
Here we are dealing, in general, with systems that require computing components that reflect national language translation and cognition and hence will typically be knowledge based. Usually it is believed that the computer-human interfaces should additionally include multimedia technology.
It is not clear to me - I am not an "expert" in the field - whether the requirements issues are well understood:
- There are up to four "players" in any such educational technology system: (i) there is the computer system with all its "gadgets"; (ii) there is the student; (iii) there is the teacher; and (iv) there is the subject - the topic being taught or learned.
The sequence in which I listed these four "actors" is usually reflected in many proposals and actual products. That is: often the multimedia components of the computer hardware "swamp" the intellectual content of the subject field. It also seems that we do not always expose, let alone exploit, all the possible interfaces between the four "players" consistently.
- In this connection, there are different emphases to be made: either the teacher is supported in the education process vis-à-vis the students, or the students are supported in their learning process. Thus there are times in an expert system like educational technology when it is unclear who is being the expert!
- When an educational technology system is offered it naturally appears that definitive means of "didactic" and "pedagogic" nature are being provided. But is that the case? Do there exist generally accepted criteria for good "didactics" and good "pedagogics" at the operational level required by educational technology?
Thus the role of cognition seems far from settled.
- The few systems that we have seen demonstrated have usually been developed by one person. Most often, this results in less than professionally engineered and programmed systems, which in addition emphasize only one of the four cornerstones of the educational technology-based teaching/learning process. By less than professionally engineered systems I mean for example: systems featuring disturbing, ad hoc interfaces by the users and technology. By less than professionally programmed systems I mean for example: obvious, generally adaptable knowledge-based techniques are usually omitted in favour of standard, "hard-wired," non-flexible coding tricks.
The sum total of the above suggests that this exciting and undoubtedly much-needed educational technology has not been launched as the result of commercially sound or publicly high priority endeavours. Industries and businesses are not embarking on the production and marketing of educational technology. If it exists, yes; then such companies will willingly offer it via their marketing channels, as something that will enhance their other, commercially viable products. Ministries of education in several countries are likewise "holding their breath" - regarding, it seems, educational technology as an "addon luxury." Contributing to the attitude of the commercial field is that of reduced government funds for education, and contributing to the attitude of education ministries is that of widespread resistance among teachers to the new technology. It seems, however, that, for example, efforts in Bulgaria and Russia are exceptions that offer promising possibilities.
The UNU/IIST is not interested in becoming involved in controversial issues, but the UNU/IIST is potentially interested in establishing, possibly as a Unesco-sponsored project, a series of increasingly ambitious feasibility, demonstrator, and technology transfer product development software projects, combined with training courses about existing technology.
It is most likely an over-ambitious aim that such studies can bring together, on a global scale, the best expertise in order that some "standardizable" consensus can be obtained, and such that this pedagogics, didactics, and cognition consensus could form the basis for a distributed software "village" industry in the developing world, where each "cottage" specializes in some field and subject. To wit: secondary school mathematics, university biochemistry, trade school auto mechanics, etc.
The UNU/IIST is interested in offering its good offices in an international, cooperative effort that critically examines the problem of educational technology and proposes portable structures for implementation.
2.2.6 Multimedia Technology
It has been implied that multimedia technology may divert the developer and the user from the "real thing"; namely, that of reflecting, respectively utilizing, by means of computer software, deep knowledge about the problem domain.
The UNU/IIST will attempt not to forget issues of multimedia technology and will be ready to undertake projects and train in multimedia issues. Should the UNU/IIST be so involved, then it is likely that it will emphasize: (i) how to choose such technology based on proper requirements studies, (ii) how to let its functions be determined by considered use of cognition, (iii) how such technology influences technical solutions, and (iv) techniques for their programming.
It seems that many small software houses could potentially base their project services or their marketable products on specialization in the use of specific media in specific application areas. It seems that one does not need vast armies of programmers in order to develop appropriate software.
2.3 Material versus Intellectual IT Worlds
It is normally the material quantities, the physical components (including the "smart, gimmicky, colourful" user-interface equipment), the high-speed execution, the large-volume storage, and the low price that enable ergonomically and economically feasible access to IT.
And often that is naively all that IT journalists see and report on!
The real advances in our improved understanding of problems and in our algorithmic or deductive ability to solve this problem using computers are harder to communicate and digest.
The UNU/IIST will try to present a balanced picture.
3. Part 3: Advanced technologies
In the final part of the talk I express the desire of the UNU/IIST to become a showroom for demonstrating advanced computer-based technologies.
In addition to the "soft technologies" of methodologies and the "less soft" applications (parts 1-2), the UNU/IIST also aims at featuring, as an integral part of its operations, some of the underlying technologies implied by the presentations at the Kyoto Symposium. Thus, the UNU/IIST is expected to become a showroom where awareness courses and "hands-on" experience can illustrate front-of-the-wave technology in a context not readily available elsewhere in the developing world.
This then represents the third axis by means of which the UNU/IIST hopes to be able to help the developing world improve access to information technology.
3.1 Specific Material Technologies and Tools
From the next subsections you may get the erroneous impression that the UNU/IIST will assist in the transfer of material technology. Remarks made in subsection 2.2 apply. The UNU/IIST, however, may, in Macau, feature some of the hard, material technologies - some of which are now mentioned.
3.1.1 Supercomputing
This subsection is somewhat speculative! The UN system has a great need for gathering and evaluating socio-economic data - but lacks facilities for the kind of supercomputing sometimes called for when performing extensive regional and global modelling and simulation (based on such data and with large-scale mathematical [operations research analytic and statistical] models). The developing world itself has no such well-identified supercomputer centre.
Modern technology now provides dramatically new, and, in some cases, rather lower priced supercomputers than heretofore experienced. The industrial world the United States, Japan, and Europe - now has several such supercomputer centres established outside the nuclear physics sector.
The UNU/IIST would like to take the initiative to establish such a centre, for example as part of the UNU/IIST, and located in some developing country, for example, but not necessarily, in Macau.
The supercomputer centre could be a joint effort with a supercomputer manufacturer (XXX) and be announced as the "XXX/UN Supercomputer centre."
That centre could be the place where the entire UN system could satisfy its demand for supercomputing. Its presence in a developing country and close to many other developing countries could stimulate local build-up of expertise in supercomputing. The centre could also be a place where universities of the developing world fill their demand for supercomputing facilities. Its presence in one of their countries should provide a more sensitive affinity to the needs of the developing world.
It is envisaged that each user of the global (or Asia-Pacific regional) centre link up with that centre via satellite communication - much like the NICNET mentioned in section 2.2.2.
The UNU/IIST can train users in installing, operating, and applying existing software for the modelling and simulation of large-scale models, and for visualization of socio-economic and scientific data. The UNU/IIST can make policy planners and university administrators aware of the many possibilities supercomputing offers. The UNU/IIST can educate operations researchers, experimental scientists, and policy planning staff to develop their own large-scale (simulation) models. The UNU/IIST can finally coordinate regional and global efforts in large-scale modelling and simulation.
The above are "loose" plans, "dreams" perhaps "in the eyes of the beholder"!
3.1.2 Multimedia Laboratory
The UNU/IIST is interested in establishing, and many clients of the UNU/IIST expect it to establish, a Multimedia Laboratory - a showroom featuring the latest multimedia technology: hardware plus software, including tools and applications. Various conditions in the developing world prevent the ready availability of such technology. The UNU/IIST could repair such a situation.
The UNU/IIST can then, within the multimedia laboratory, give training, awareness, and education courses, in usage, product management, and actual software development of applications that are based on multimedia technology. The idea is to have easy access to this technology, provide a means for software companies of the developing world to skip a product generation or two, and step right into one of the commercially most viable forefronts.
3.1.3 Other Technologies
It might be useful to feature several other technologies at the UNU/IIST Telematics, CIM (Computer Integrated Manufacturing, incl. Robotics), etc.
3.2 Conclusion
There is imminent danger that in focusing on the "hard" technologies, applications often become artificially technology motivated and driven. On the other hand, the newest versions of these technologies often enable applications that were previously not feasible.
In part 1 I emphasized the methodology aspects: techniques for achieving applications. In part 2 I emphasized developing-world needs: applications that solve problems. And in part 3, I have emphasized the hard technology: ultimate "carriers" of such applications - carriers often requiring special development techniques, novel technologies that often remove constraints on previous usage.
I have thus sketched three axes of possible UNU/IIST offerings - each aimed at helping developing countries improve their access to information
(including computing) science and technology. The first axis stressed "soft-soft" methodologies, the second stressed "soft" applications, and the third stressed "hard" technologies. The UNU/IIST emphasizes the intellectual possibilities afforded through any of these axes.
Obviously the UNU/IIST cannot feature a complete range of access provisions-as outlined in parts 1, 2, and 3 of this paper. Which will actually be pursued depends on policy priorities within the UNU/IIST and on funding opportunities originating outside the UNU/IIST.
The UNU/IIST is, on the one hand, to be an institute located in and offering many activities in Macau. On the other, the UNU/IIST will strive to see the establishment, if needed, of similar institutes, perhaps specializing in specific IT areas elsewhere in the developing world. The UNU/IIST will be ready to advise, on the basis of many years of experience with industrial world IT (including software technology) programmes, regions of the developing world should they wish to embark on similarly successful R&D programmes.
1. Bjørner, D. (1991). Proposed Strategy, Tactics and Operational Plans: The United Nations University (UNU), International Institute of Software Techology (IIST), "The Macau Institute". Dept. of Comp. Sci., Techn. Univ. of Denmark, 13 and 31 July, 5 Aug. 1991.
2. Bjørner, D. (1992) "Trustworthy Computing Systems: The ProCoS Experience." In: 14th ICSE: Intl. Conf. on Software Eng., Melbourne, Australia. ACM Press, 11-15 May 1992.
3. Bjørner, D., C.A.R. Hoare, and H. Langmaack, eds. (1992). The ProCoS Project, volume 5XX of Lecture Notes in Computer Science. Heidelberg: Springer-Verlag.
4. Bjørner, D., and J.F. Nilsson. (1992). "Algorithmic and Knowledge Based Methods-Can They Be Unified?" In: International Conference on Fifth Generation Computer Systems: FGCS '92 ICOT, 1-5 June 1992.
5. Brooks, F.P. (1986). "No Silver Bullet: Essence and Accidents of Software Engineering." In H.-J. Kugler, ed. Information Processing '86. Amsterdam: North-Holland Publ. Co. IFIP World Computer Congress, pp 1069-1076.
6. Harel, D. (1992). "Biting the Silver Bullet." IEEE Computer 25 (1): 8-20.
7. Ramani, S. et al. (1989). International Institute for Software Technology, A Proposed UNU Research and Training Centre. UNU, June 1989.
The potential of information technologies for international cooperation
Abstract
1.
The new technologies
2.
Information and knowledge
3. Activities of some international bodies in
information technologies cooperation
4.
Educational strategies
5.
Developing countries
6. Negative tendencies and illusions
References
Blagovest H. Sendov
The paper emphasizes information technologies as an object of international cooperation. The activities conducted by a number of international organizations active in this area are outlined. Reflections follow on the identification of appropriate thrusts for international education and training activities concerned with new information technology, in particular with regard to the developing countries.
It is impossible to work together without exchanging information. In this respect, information technologies may be considered as basic instruments for cooperation. But when we discuss the potential of information technologies for international cooperation, it is necessary to distinguish two different aspects: (1) information technologies as a tool for, and (2) information technologies as an object of international cooperation. In this paper I emphasize the latter.
When considering information technologies as an object (and objective) of international cooperation, I shall briefly discuss the notion of "information" as a substance to be processed through these technologies.
As examples of successful ongoing international cooperation in information technologies, I review some activities of the International Federation for Information Processing (IFIP), the Intergovernmental Informatics Programme (IIP) of Unesco, the Committee on Data for Science and Technology (CODATA) of the International Council of Scientific Unions (ICSU), and the project of the International Association of Universities (IAU) called the "University-based Critical Mass System for Information Technology" (USIT). This selection is based on my personal acquaintance with these organizations. There are so many other governmental and non-governmental, international and regional organizations active in information technology that it would be difficult even to list their names.
Education is an important objective for international cooperation in information technologies. There are many important problems in devising strategies of education for the information age that concern all countries. I will discuss some of these problems with emphasis on needs for radical change.
Collecting and using different types of information (in the form of data) are also important problems of international cooperation, especially in science. Among the areas needing attention are standardization, accessibility, reliability, and, last but not least, the cost of obtaining and processing different kinds of scientific data.
When speaking about "new technologies" we usually imply information technologies. From the vantage point of the approach to information, the history and foreseeable future of mankind can be viewed as consisting of the following eras [7, 19]:
(1) Neolithic era. Man acted on the basis of learning how to make use of the potential provided by Nature to have food available in sufficient amount and whenever needed.
(2) Industrial era. Man acted on the basis of learning how to use the laws of Nature to have energy available in sufficient amount and whenever needed.
(3) Information era. Man acts and will act on the basis of how to use Nature and its laws to have information available in sufficient amount and whenever needed.
The technologies of the industrial era, or the old technologies, are based on the transformation (processing) of energy. More generally speaking, the technologies of the industrial era related to the transformation of physical matter. The technologies of the information era, or the new technologies, are based on the processing (transformation) of intellectual product.
There is one substantial difference between material products and information products. To multiply a material product, one needs comparable material resources. But multiplication of an informational product does not require a proportional expense. This correlation is significant in planning strategies for international cooperation in information technologies.
The information technologies comprise basically the storage, processing, and transmission of information. These branches of information technology have roots in the history of knowledge and learning. The book, for example, represents an accumulation of knowledge and culture. Instruments for the automatization of arithmetic operation have been used for centuries. But only after information technologies were based on electronics and new physical principles were the new information technologies born.
These three types of information technologies have their specifics. Historically, they were developed as independent technologies. The electronic computer became the integration medium for these three branches of information technology. Today it is impossible to imagine communications without computers. But while the computer has a central role in the new information technologies, these are not to be identified with computers only.
When using information, very often we talk about the information as knowledge. But it would be wrong to equate them. According to the Random House Dictionary of the English language (1971), "information" is:
1. Knowledge communicated or received concerning a particular fact or circumstance.
2. Information, knowledge, wisdom are terms for human acquirements through reading, study, and practical experience. Information applies to facts told, read, or communicated which may be unorganized and even unrelated .... Knowledge is an organized body of information; or the comprehension and understanding consequent on having acquired and organized a body of facts.... Wisdom is a knowledge of people, life, and conduct, with the facts so thoroughly assimilated as to have produced sagacity, judgment, and insight.
In the first approximation there is no substantial difference between information and knowledge. It is accepted that knowledge is "structured information." Today, the information industry is producing so-called "knowledge based systems." A point I would like to make here is the difference between "knowledge" stored in a computer and "knowledge" possessed by a human being. The "knowledge" in a computer is information structured according to given rules that are precisely defined and well known. The "knowledge" possessed by a human being is information structured through the learning process, and the rules for this structuring are not known. It is natural to expect that the learning process, or the rules for structuring that transform acquired information into human knowledge, is individualized. That means that we have to differentiate between "knowledge" of a human being and "knowledge" stored in the memory of a computer [14]. The first may be called simply knowledge (or natural knowledge), and the second artificial knowledge.
On the other hand, the notion of knowledge is closely related to the notion of "intelligence," which means the capacity for reasoning, understanding, and for similar forms of mental activity. Knowledge is a necessary condition for every manifestation of intelligence. I have to stress here that intelligence has been considered as a unique characteristic of a human being. That is why when the idea of imitating man's intelligence was born, the term "Artificial Intelligence" was accepted as the most correct description.
The successes and shortcomings of Artificial Intelligence are well-known. Critics like Hubert Dreyfus [4] many years ago called attention to its difficulties and restrictions. But in one aspect the enthusiasts of Artificial Intelligence are on the safe side. They have addressed the fundamental task of imitating intelligence, and this imitation, good or bad, they called Artificial Intelligence. Their aim is not to create intelligence artificially, but to create Artificial Intelligence. It is quite natural that artificial intelligence be something different from natural intelligence. Let us mention that in the material world we are happy with a lot of artificial materials such as artificial wool, artificial silk, artificial caviar, etc.