PART FOUR: COMMUNICATION AND CALCULATION 752 contractors, such as NASA for space programme records; Bell Telephone Laboratories for maintaining Nike-X data; and the Southern Pacific Railway for waybills. The Ampex professional system was the standard for commercial television studios for several decades. Much less expensive recorders using one inch (2.5cm) wide tape attracted educators and training establishments which wanted to experiment with this alternative to film production; these VTRs use helical recording (tracks at a slant instead of almost vertical). Many helical recorders were in use during the 1960s, and Philips soon marketed a videocassette recorder (VCR), which was much easier to operate for the inexperienced. Philips one-hub cassettes were not very reliable, and image quality was inferior. Sony brought out a two-hub system using 3/4 inch-wide (19mm) tape, called U-Matic. These recorders sold for $2000 or more, but were very reliable, versatile and provided a high-quality colour picture, and U- Matic became the standard for educational and industrial use. During the 1970s, Phillips, Sony and JVC (Japan Victor Corporation) brought out rival two-hub consumer systems using 1/2-inch (12.5mm) tape. By the early 1980s the most popular was JVC’s VHS (video home system); Sony’s Betamax had only 10 per cent of the market, and the Philips V2000 was scrapped. By 1985, inexpensive VHS recorders capable of recording up to four hours of TV on one cassette, and with built-in timers for unattended recording were dominant. In the early 1970s, mass memory systems based on lasers and holography (see p. 737) were developed. W.J.Hannan of RCA invented a home system, Selectavision, which used clear vinyl tape for the storage and playback of colour television programmes; images were stored holographically. Precision Instruments Company developed an optical terabit memory (1 terabit=10 12 bits), the Unicon, based on burning 1µm holes in metal-coated polyester tape with a laser. However, such optical media, like electromechanical ones, had a major disadvantage-data could be recorded only once, and could not be erased. Erasable and reusable magnetic systems were preferred for active mass storage, with optical systems relegated to archival support, usually for images. Computer data storage The earliest form of magnetic mass-data storage was magnetic tape, which had been originally developed for the analogue recording of sound waves (see p. 724). However, even with multi-track digital recording, which allowed the bits encoding each character to be stored in parallel, tape had to be scanned sequentially. Computers also needed rapid random access to small amounts of data. Magnetic cores strung on a grid of wires gave the highest speeds, but were very expensive on a per-bit basis (one bit per core); therefore, magnetic INFORMATION 753 drums were introduced for back-up mass memory. In 1956, IBM introduced RAMAC (Random Access Method of Accounting and Control for Automatic Computers), the first magnetic disk system. Up to 100 rigid disks were mounted on a common spindle which spun at high speed; data were written and read (computerese for recorded and reproduced) by magnetic-sensing heads flying in and out only microinches above the disk surfaces. In 1962, IBM marketed the first cartridge magnetic disks, which could be mounted on a disk drive for access, but removed for storage—a similar system to that employed for magnetic tapes. About 1970, IBM introduced a non-rigid magnetic disk which has come to be called the floppy disk, or simply floppy. This has become the preferred medium for off line storage of data and programs for small computers. The original floppies were 8in (20.3cm) in diameter and could store 100,000 bytes (100 kbytes) on a side; the encoding system, data density and file format were set by IBM, becoming a de facto standard. However, when microcomputers became popular in the late 1970s, a 5 1/4in (13cm) version was introduced, soon dubbed the minifloppy. Unfortunately, formats were never standardized, so that by the early 1980s there were a hundred different ones: single-density, double-density, even quad-density; and both single-sided and double-sided. Despite this diversity, the 5 1/4in floppy has become the universal external storage medium for microcomputers, providing from 200 kbytes to more than 1 megabyte (1 Mb) per disk. In the early 1980s, further miniaturization took place: a 3 1/2 in (9cm) diameter disk, sometimes called a microfloppy. These disks can store 1.5 Mb and are more reliable because they are enclosed in a rigid case. Also, in the late 1970s, rigid disks of 8in (20cm), and then 5 1/4 in (13cm) diameter were introduced to provide much greater mass-storage capacity for microcomputers. They are called hard disks to distinguish them from floppies, and are usually not removable, being housed in a dust-free casing. They spin at much higher speeds than floppies, and their plated-metal surfaces are capable of much higher recording densities, providing users with from 20 to more than 100 Mbytes of on-line data access. Optical Media The dominance of magnetic media for the storage of all types of information is now being challenged by optical media, particularly those based on the use of lasers for recording and reproduction. The first videodisc system was developed by Philips in the Netherlands in the 1970s for the storage and playback of still and motion pictures; all information is stored digitally in tracks as microscopic pits (‘micropits’), and read out by means of reflected laser beam. Thomson CSF, a French company, developed a similar format, but data PART FOUR: COMMUNICATION AND CALCULATION 754 is recorded as microscopic surface blisters, and read out transmissively. Recording techniques are based on pulse-width modulation, and both use discs the same size as LP records, 12in (30.5cm) in diameter. The Philips system was originally intended for home use; consumers would buy videodiscs of their favourite movies, just as they bought pop songs or other music on gramophone discs. However, instead of about five minutes of music (on 45rpm records) or thirty minutes (on 33 1/3 rpm LPs), videodisc enthusiasts would get up to an hour of sound, colour motion pictures or TV programmes. A number of manufacturers have marketed laser videodisc players, including Philips; Magnavox, a US company; and Japanese companies such as Pioneer and Sony. However, videodisc marketing not only had to contend with three incompatible broadcast television standards (see p. 746), but also with the well- entrenched videocassette industry which so dominates the home video market that it threatens to replace motion-picture theatre distribution as the primary way in which the public sees movies. Therefore, although multinational companies such as General Motors and IBM adopted videodisc technology for industrial training, it was a failure in the consumer market. The worst drubbing was taken by RCA; by the time they withdrew their contact-stylus, capacitivesensing system, they had spent about $100 million on its development and marketing. For education and training, a single videodisc can store up to 54,000 individually addressible, colour, still images; or a mixture of stills and motion sequences. Archival videodisc systems are an ideal distribution medium for large colour slide collections, and are becoming a useful tool for museums. For example, the Smithsonian Institution’s Air and Space Museum in Washington, DC, the world’s most visited, has put its archives of Wernher von Braun, the rocket pioneer, on a single videodisc for sale to the public. They are also making their photographic collections (almost a million items) available in this format. Even more important than the vast storage capacity of videodiscs is the possibility of student interaction using a built-in microprocessor or external microcomputer. Interactive videodisc is the most promising method so far devised to implement the long-sought promises of CAL (computer-assisted learning), alternatively known as CAI (computer-assisted instruction) in the United States. Apart from text-oriented ‘drill-and-practice’ lessons, CAL has been too expensive in cost per student-hour; however, by making available the distribution of 50,000 images at a selling price of $20, videodisc-based CAL has the potential of becoming the least expensive form of audiovisual instruction ever devised. The main obstacle is the high cost of developing well-designed and tested instructional programmes—it may take a professional team of educators and technologists 100 hours or more to develop a one-hour programme. INFORMATION 755 Optical Data Recording Another use for laser-optical technology is the storage of digital data, a natural extension because all types of information (pictures, sound, text, data) can be recorded by laser-based technology. More than a gigabyte can be stored on one side of a 12in (30.5cm) DOR (digital optical recording) disk, ten times more than on a similar-sized magnetic disk; however, random access times and transfer rates are much slower. For very large memory requirements, multidisk systems are available, such as Philips’s Megadoc, which uses a jukebox mechanism holding up to 64 optical disks. Within a quarter of a minute, users can access one record (equivalent to an A4-sized page) out of a store of 15 million such records. Megadoc could provide the all-electronic library, which, unlike present automated bibliographic systems, would provide immediate access to facsimiles of original documents. At the other extreme is the unit record, particularly important in transaction processing systems. Such systems are characterized by the storage of only a small amount of information, which must be updated frequently, in each record. Drexler Technology in the United States has developed the Laser- Card, which can store up to 400 pages of digitized text (2 Mbytes) on a creditcard-sized piece of polycarbonate. This medium can store ten thousand times more data than magnetic-strip plastic cards, and lasts twenty times longer. Promising applications include individual, pocket-sized medical histories; maintenance manuals and other reference publications; and software for microcomputers. Philips and Sony, working together, developed a laser-read optical format for audio recording, which was introduced in 1979 as the Digital Audio Disk (DAD). Without the incompatibility problems of television, Philips and Sony were able to persuade other manufacturers to adopt it. Stereophonic music recordings using this medium were introduced to the public about 1983, under the name Compact Disc (CD); they are 12cm (4 3/4 inches) in diameter, and 1.2mm thick. Although their dynamic range is 96 decibels, 30 times greater than on LP records, CDs provide virtually noise-free listening during quiet passages. Also, because the recording is read by reflected laser beam, they are virtually indestructible. Sound to be recorded for CD use is sampled 48,000 times a second, each sample being encoded with 16-bit resolution. With the addition of errordetecting bits, the digital band-width approaches 1 Mbit per second, fifteen times greater than PCM-based telephony. A compact disc can provide uninterrupted play for up to an hour; any track can be selected and played automatically; and sequences of selections can be pre-programmed by the user. Soundwaves are encoded by PCM, the basic information carrier being pits burned by laser into the master, each of which is only 0.16µm high and 0.6µm wide. Such dimensions are those of light wavelengths, which range from PART FOUR: COMMUNICATION AND CALCULATION 756 0.4µm (violet) to 0.7µm (red) —hence the iridescent colour patterns that CDs exhibit, characteristic of lightwave interference. This is the greatest density of artificial information storage yet achieved in a consumer product, and the error-correction capability of the code used is such that a 2mm hole drilled in the record will not affect playback. In 1985 the same technology was made available for the storage of digital data. Virtually the same players can be used as for music reproduction, using a switch to bypass the digital-to-analogue conversion required by audio amplifiers and loudspeakers. Each CD-ROM (for Compact Disc Read-Only Memory) can store up to 500 Mbytes of data. This is 100 times greater than state-of-the-art floppy-disk technology. However, as has been true of many new media, finding the right place for CD-ROM technology is a problem. What is the best application for a read-only (soon to be write-once, read- only) storage system which has a data storage capacity equivalent to 1000 printed books? One market in which many CD-ROM-based services are interested is the publication of bibliographical databases. Such databases were originally available only in printed form, but as publishers accumulated vast amounts of machine-readable text they realized that a market existed for on-line access to frequently updated versions of the same material. Customers use computer terminals or microcomputers connected to large, central computers via telecommunications, so remote database searching is expensive and demands considerable training. CD-ROM promoters believe this new technology could be more cost-effective: users can interact with the medium without entailing constantly accumulating communications costs, and the difficulties of ‘logon’ procedures which require accurate keying of long sequences of characters are eliminated. Present CD-ROM storage capacity is too small for retrospective searching applications, which may require access to data files going back 10–20 years, and the discs cannot be updated. However, as CD-ROM technology advances these problems will be mastered, so that by the early 1990s publishers will at last have a convenient, compact and economical alternative to print—heralding the era of electronic publishing. FURTHER READING Introduction Dummer, G.W.A. Electronic inventions and discoveries: electronics from its earliest beginnings to the present day, 3rd edn (Pergamon Press, Oxford, 1983) Grogan, D. Science and technology: an introduction to the literature, 4th edn (Clive Bingley, 1982) INFORMATION 757 Ryder, J.D. and Fink, D.G. Engineers and electronics: a century of social progress (IEEE Press, New York, 1984) Timekeeping Landes, D.S. Revolution in time (Harvard University Press, Cambridge, 1983) Counting, calculating and computing ‘The computer issue’, Science, vol. 228, no. 4698 (26 April 1985) Shurkin, J. Engines of the mind: a history of the computer (Norton, 1985) The telegraph McCloy S.T. French inventions of the eighteenth century (University of Kentucky Press, Lexington, 1952) The telephone Shiers, G. (Ed.) The telephone: an historical anthology (Arno Press, New York, 1977) The gramophone Hope, A. ‘A century of recorded sound’ and ‘From rubber disc to magnetic tape’, New Scientist (22/29 December 1977, pp. 797–9; 12 January 1878, pp. 96–7) Radio and radar Barnouw, E. A history of broadcasting in the United States, 3 vols. (Oxford University Press, New York, 1966, 1968, 1970) Briggs, A. A history of broadcasting in the United Kingdom, 3 vols. (Oxford University Press, London, 1961–5) Susskind, C. ‘Who invented radar?’, Endeavour, New Series, vol. 9, no. 2 (1985) pp. 92–6 Photography Pollack, P. The picture history of photography, 2nd edn (Harry N.Abrams, Inc., New York, 1970) Desilets, A. et al, La photo de A à Z (Les éditions de l’homme, Quebec, 1978) Okoshi, T. Three dimensional imaging techniques (Academic Press, New York, 1976) PART FOUR: COMMUNICATION AND CALCULATION 758 Facsimile and television Barnouw, E. Tube of plenty: the evolution of American television (Oxford University Press, New York, 1975) Satellites Clarke, A.C. The making of a moon: the story of the earth satellite program, 2nd edn (Harper & Brothers, New York, 1958) McDougall, W.A. The heavens and the earth: a political history of the space age (Basic Books, New York, 1985) Information storage today Information takeover (videocassette with printed Fact File, New Scientist/Quest Video, 1985) PART FIVE TECHNOLOGY AND SOCIETY 761 16 AGRICULTURE: THE PRODUCTION AND PRESERVATION OF FOOD AND DRINK ANDREW PATTERSON INTRODUCTION From the time that the first steps were taken to obtain food by the cultivation of the land, the proper practice of that technology has been of concern to the whole of the community that depended upon it. The need to record the methods of the experts so that their knowledge might be utilized by those less proficient, or might be passed to future generations, was felt by many early authors. As a result various literary sources have been left to us, from which we are able to gain glimpses and sometimes great insights into the agricultural practices of the past. The state of preservation of this literature, and the range of its content, has affected the quality of knowledge available. The most complete material in terms of subject matter and timespan comes from the Chinese. Classical Mediterranean writing has provided quite detailed information on technology, although little of social or economic interest. Middle Eastern sources have provided tantalizing glimpses of laws of property and husbandry, and also volumes of account books and records of palace or state stores. In contrast, the mediaeval period has left us with only snatches of technical and economic information, and our knowledge of European agriculture is very scanty until the eighteenth century. Only in very recent times has any consideration or study been given to those areas of the world outside the European influence, and outside the knowledge left by the written word. In addition, archaeology has provided extensive data with the retrieval of the actual equipment used, and with the discovery of pictorial representations of equipment and practices. In the past twenty years it has also occupied itself with . food by the cultivation of the land, the proper practice of that technology has been of concern to the whole of the community that depended upon it. The need to record the methods of the experts. glimpses and sometimes great insights into the agricultural practices of the past. The state of preservation of this literature, and the range of its content, has affected the quality of knowledge. based on the use of lasers for recording and reproduction. The first videodisc system was developed by Philips in the Netherlands in the 1970s for the storage and playback of still and motion