A Wireless Future:College And University Libraries Unplugged This paper was presented at the 1996 CAUSE annual conference. It is part of the proceedings of that conference, "Broadening Our Horizons: Information, Services, Technology -- Proceedings of the 1996 CAUSE Annual Conference," page 5-8- 1+. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage. To copy or disseminate otherwise, or to republish in any form, requires written permission from the author and CAUSE. For further information, contact CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301; 303-449-4430; e-mail info@cause.org. A WIRELESS FUTURE: COLLEGE AND UNIVERSITY LIBRARIES UNPLUGGED by Clifton Dale Foster Auburn University Libraries Auburn, Alabama ABSTRACT The development of electronic resources and "virtual libraries" offers many new opportunities for accessing voluminous amounts of information. The research described here chronicles efforts to take this concept one step further by testing wireless data communication in educational and library environments. Wireless data communication is revolutionizing many jobs and services, much as the introduction of desktop computers did fifteen years ago. This developing technology offers great potential for broadening access to library services in ways never before conceived. Wireless data communication may be the next logical step in the evolution of library information systems. Three basic types of wireless data technologies are discussed--cellular, packet-radio, and wireless local area networks. The latest prototype projects demonstrating wireless applications in educational and library settings are described. The technical feasibility of accessing library online catalogs, commercial online databases, and the Internet using wireless technologies is evaluated. Other uses of wireless communication in libraries and educational institutions are proposed. Sources for additional information on wireless libraries, including the Wireless Libraries Homepage at Auburn University, are provided. INTRODUCTION The introduction of wireless data communication is revolutionizing many jobs and services, much as the introduction of desktop computers did fifteen years ago. This developing technology offers great potential for broadening access to library services in ways never before conceived. Wireless data communication may be the next logical step in the evolution of library information systems. Just as many librarians were unfamiliar with the potential of desktop computers in the early 1980s, we again find ourselves on "the outside looking in" when it comes to wireless data communication. The explosive growth of the Internet and the continued proliferation of electronic information resources place libraries square in the middle of revolutionary changes in the way information is stored and accessed. These dramatic changes in wire/cable-based electronic information delivery are paralleled by developments in the wireless realm. If librarians are to take advantage of the great potential offered by wireless data communication, it is imperative that we understand at least the rudiments of this technology. This paper describes the basic types of wireless data transfer technologies currently available: cellular, packet- radio, and wireless local area networks (LANs).[1] The latest demonstration projects to apply wireless technologies in educational and library settings are discussed, including the Library Without a Roof Project at the University of South Alabama, the Wireless Andrew Project currently underway at Carnegie Mellon University, and the Ricochet Wireless Project at the University of California at Santa Cruz. A survey of the technological landscape is provided, along with an evaluation of the feasibility of accessing library online public access catalogs (OPACs), commercial online databases, and the Internet using wireless technologies. HOW DO YOU SAY "WIRELESS"? Perhaps the most familiar form of two-way wireless communication technology is analog cellular communication. In this system, cellular data transmission is much like cellular voice transmission in that it uses the same analog radio signals (825-845 MHz and 870-890 Mhz) routed through circuit- switched networks. In the simplest configuration, a user can plug a desktop or laptop computer equipped with an analog modem into a cellular phone using a cellular data interface. This method was used with some success by Steven J. Bell and Michael Halperin in testing wireless online searching capability.[2] Several public libraries have used existing analog cellular voice services to connect bookmobiles to online catalogs.[3] While certainly one of the least expensive wireless options in terms of equipment expenditures, analog cellular has three major drawbacks--transmission costs, speed, and reliability. With cellular airtime fees ranging from $.[18] to $.50 per minute, connection charges add up quickly. Since analog cellular systems are designed primarily for voice, they tend to be very slow at carrying data. Speeds of up to only 9,600 bits per second are possible with any degree of reliability. Reliability of data transmission is often a factor, since static in analog voice channels can lead to garbled data on the computer screen. Microcom, Inc. has gone a long way to enhance the speed and reliability of analog cellular capability by developing the proprietary MNP 10 cellular networking protocol. MNP 10 is available in most of today's cellular-ready modems and data interfaces. Still, for applications requiring high-speed data rates, analog cellular comes up short. Digital cellular technologies were developed to overcome some of the shortcomings of analog signals in delivering data. Two major types of digital cellular transmission technology are available for use in standard circuit-switched networks--Time Division Multiple Access Code (TDMA) and Code Division Multiple Access (CDMA). Both translate analog data into digital bits. They differ in the way they code the data. By translating analog signals into digital format, greater transmission speeds and greater accuracy of data are possible. The digital cellular technology receiving the most fanfare these days is the long-awaited Cellular Digital Packet Data (CDPD) service. Instead of using standard circuit-switched telephone networks designed primarily for analog transmissions (like TDMA and CDMA), CDPD translates data into digital packets and allows these packets to channel-hop among the ever-changing openings in cellular voice traffic. CDPD was developed and is supported by a consortium of IBM, AT&T, GTE, and various regional Bell telephone companies. CDPD has the advantage of offering greater transmission data speeds and greater reliability than any of the circuit-switched cellular networks. The availability of any of the digital cellular transmission technologies is totally dependent upon the type of system installed by local cellular carriers. CDPD networks are still quite new and have not received widespread installations as yet across the U.S. and Canada. Packet-radio communications hold much potential for data transmission but have received little support from the library community. Two nationwide commercial packet-radio systems exist--Ardis and RAM Mobile Data. Unfortunately, these commercial services offer little support for practical library applications. The real potential for packet-radio networks in libraries is through locally developed systems. In the mid-1980s, Edwin B. Brownrigg, Clifford A. Lynch, and Rebecca L. Pepper examined the applicability of wide-area packet-radio networks in various library automation and higher education contexts.[4] Their research concluded that the current state of packet-radio development at that time was unsuitable for the specific networking applications they attempted. However, their results indicated that a suitable device for library applications could be engineered and that packet-radio networks held much potential in library automation. Recent technological developments may make packet-radio more functional for library networking applications. Wide-area packet-radio is a digital mode of communications used extensively by amateur radio enthusiasts. In a packet- radio data transmission system, the computer modem is replaced by a terminal node controller (TNC). The telephone is replaced by a radio transceiver, and the phone network is replaced by radio waves. Packet-radio takes data sent from the computer and sends it via radio to another radio station similarly equipped.[5] The operation of a packet-radio station is transparent to the end user; connect to the other station, type in a message or command, and it is sent automatically. The TNC divides the message into packets, keys the transmitter, and then sends the packets. While receiving packets, the TNC automatically decodes, checks for errors using built-in error detection schemes, and displays the received messages. Like other digital modes, packet-radio has the ability for many users to use the same frequency channel simultaneously. Since packet-radio is most commonly used at the higher radio frequencies (VHF), the range of the transmission is somewhat limited. Generally, transmission range is limited to "unobstructed line-of-sight" plus approximately 10-15%. The transmission range is influenced by a number of factors including: transmitter power; the type, length, and location of the antenna; frequency used; and the existence of obstructions (hills, buildings, etc.). Transmission range can be extended by using a TNC as a packet relay station, sometimes called a digipeater. This allows for greater range by stringing several packet stations together. AX.[25] is considered the defacto standard transmission protocol for amateur packet-radio use and is even recognized by many countries as a legal operation mode. The Transmission Control Protocol/Internet Protocol (TCP/IP), commonly used over the wired Internet network, is also available for packet-radio. The KA9Q NOS program (also called NET) is the most commonly used version of TCP/IP in packet-radio operations. For libraries, the major advantages in packet-radio networks are that transmission costs are minimal (once initial operating equipment is purchased) and that no Federal Communication Commission (FCC) licenses are required to operate within prescribe frequencies. The drawbacks are that coverage is limited to line-of-sight and, at this time, specialized library computer applications taking advantage of packet-radio technology do not exist. This is certainly an area in great need of further research and development.[6] The final type of wireless technology offering major potential for libraries is wireless local area networks (LANs). Wireless LANs do the same thing as conventional computer LANs, only without the need to lay costly phone lines or coaxial cables throughout a building or campus. Three types of wireless LANs are available--infrared, microwave, and spread-spectrum.[7] Infrared data transmission is a well developed, mature technology and is enjoying great popularity. By using intense light beams to transmit data, it permits faster transmission speeds than many other wireless LAN technologies. Infrared devices are less expensive than other wireless LAN devices, but not as cheap as standard network cards. The drawbacks of infrared systems are that they are limited to line-of-sight transmission, sending and receiving units must be closely aligned, and the data path must be kept unobstructed. Microwave signals allow high-speed data transfer rates and have the advantage of penetrating walls. They can also support applications requiring full-motion video transmission and point-to-multipoint broadcasting. Microwave-based LANs have the same drawbacks as infrared systems in that transmission is limited to distances up to 15 miles, sending and receiving units must be closely aligned, and the data path must be unobstructed. Microwave systems have been used with success in deploying wide area networks (WANs) for multi-campus educational institutions. With the current state of microwave technology, these systems can offer substantial cost savings over leased lines and fiber optic connections. A spread-spectrum system is one in which the transmitted signal is spread over a wide frequency band, much wider, in fact, than the minimum bandwidth required to transmit the data being sent. Some systems use "frequency hopping" to jump around the assigned array of channels. Other systems use a faster direct-sequence spread-spectrum (DSSS) technique. The constant changing of frequencies helps speed transmission rates and increases data security. Spread-spectrum systems operate in the industrial, scientific, and medical (ISM) bands of the radio spectrum (902-928 MHz, 2.4-2.484 GHz and 5.725-5.85 GHz), which do not require FCC licenses. AT&T's WaveLAN and Telxon's Arlan are examples of two types of spread-spectrum systems currently available. Spread-spectrum signals can penetrate walls and can reach transmission speeds of up to 3 Mbps and a range of up to 1,200 feet. Wireless LANs will get a tremendous boost with the soon to be released IEEE 802.[11] wireless LAN standard.[8] The lack of standards, high prices, and perceived low transmission speeds have been the main issues suppressing wireless LAN development.[9] The new standard will allow similar wireless LAN technologies to interoperate; it will also enable wireless LANs to work with existing wired LAN technologies such as Ethernet and token ring. DEMONSTRATION PROJECTS Projects demonstrating the application of various wireless data technologies in libraries are relatively few. The first efforts, already mentioned, began more than 10 years ago and identified both technological and financial barriers to success. Recently, three major pilot projects have readdressed this issue in university settings. In November 1993, the University of South Alabama Library initiated the Library Without a Roof Project.[10] This project was designed to test the feasibility of using hand- held personal digital assistants (PDAs) equipped with cellular modems in a library environment.[11] The project was conducted with the assistance of AT&T, BellSouth Cellular, and Notable Technologies, Inc. Apparently, this was the first systematic effort to connect a PDA to library online public access catalogs (OPACs), commercial online databases, and the Internet using a digital cellular communications link. The EO 440, a personal digital assistant developed by AT&T, was chosen as the hardware platform since, at that time, it was the only PDA to offer a cellular modem option.[12] Laptop computers were not used since the keyboard configuration precluded the specific applications to be tested. This application required a totally self-contained, portable, pen-based system with cellular data communications capability that did not need a table or extraneous peripherals to operate. Using communications software supplied by Notable Technologies, Inc., librarians were able to connect with the online public access catalog within the library building. This setup allowed the librarians to roam the stacks with the PDA in hand. The same configuration of hardware and software was used to access several commercial online database services and "surf" the Internet outside the library building. While sitting beside a serene lake on campus, users could search LEXIS/NEXIS, OCLC FirstSearch, and DIALOG Information Services. The communications software was pre- configured with the necessary telephone numbers and logon scripts to simplify the connecting process. Connectivity to the Internet was available through dial access to the university's Digital VAX mainframe computer linked to the Alabama Supercomputer Network. Another wireless test project commenced in the Fall of 1994 at Carnegie Mellon University. Known as Wireless Andrew, this project plans to make the full functionality and services of Carnegie Mellon's on-campus wired computing system (known as Andrew) available by wireless means.[13] Services include e- mail, access to stored audio and imaging data, file transfers, access to the library and other databases, and full Internet services. The project is funded by a recent National Science Foundation grant award and by a service grant from major partner Bell Atlantic Mobile Systems. This project is developing a communications infrastructure based on two separate but interconnected wireless networks. Using AT&T's WaveLAN wireless LAN technology, a high-speed (2 Mb per second) wireless data network is used to provide coverage over a large portion of the campus. Outside the wireless LAN service area, a lower speed (19.[2] Kb per second) CPDP service is being developed to provide coverage of parts of the Pittsburgh metropolitan area. Users access the networks using laptop computers equipped with wireless LAN interface cards and on-board CDPD modems. Software to allow seamless movement between the wireless LAN and CDPD networks is being developed. A third related project began in 1995 at the University of California at Santa Cruz. Researchers there recently completed a successful test of a wireless computer network which enabled students and faculty to get online from anywhere on campus. The campus-wide network was developed by Metricom Inc. using their proprietary Ricochet technology.[14] The nine-month pilot project, which involved 15 students on the 2,000 acre campus, is among the most comprehensive attempts at exploring how the fledgling technology could be used in higher education. The Ricochet network employed spread-spectrum packet radios installed geographically in a mesh topology. Metricom mounted 50 radio transceivers on the tops of street lights and buildings. The company provided students in the pilot project with special wireless modems, which are rectangular black boxes--a little longer and heavier than a remote control for a television set--that connected to a computer in the same way as a traditional modem. Each had a two-inch rubber antenna that transmitted signals to the transceivers located around the campus. The system proved reliable during the university's trial period. Students were able to connect to the campus network and to the Internet from virtually anywhere on campus. The speed of the connection, an important factor for the transfer of large graphical image files, was at least as good as what was available to dormitory residents who must use the telephone to dial into the campus network. Transmission speeds ranged from 14.[4] Kb per second to 34 Kb per second, depending on network traffic. Computers wired directly to the campus Ethernet network were much faster, sometimes reaching speeds of 10 Mb per second. Users of the wireless network also had full Internet connections, which meant that they could look at World Wide Web (WWW) pages as well as review their electronic mail. Metricom currently has the Ricochet wireless network installed on nine otheruniversity campuses across the country. A WIRELESS FUTURE? Besides exhibiting the novelty of doing library research while lounging beside a campus lake, these pilot projects demonstrate the tremendous potential of wireless technology in providing "untethered" library access. For researchers, the use of PDAs and wireless connectivity within library buildings eliminates unnecessary steps in the research process. Users can conveniently search the online catalog anywhere in the building, even in the book stacks. This freedom of movement eliminates unnecessary foot traffic to/from the OPAC terminals. It also reduces the need for patrons to print or write bibliographic information on paper. For librarians, using wireless PDAs can make assisting patrons easier and more efficient. Rather than taking the patron to the OPAC terminal, it is now possible to take the terminal to the patron. Hand-held wireless computers can also be used by librarians to automate inventory checking and the collection of serials usage data. One of the most attractive components of wireless connectivity is the ability to break the physical restraints of buildings or land-line telephones. Wireless connections to information resources provides a means for marine biologists to access directly library information or other computer databases while on a boat offshore; likewise, facsimile copies of drawings or photographs can be sent directly to onboard computers without the need to return to shore. These same services would be available to archaeologists and other researchers involved in field research many miles from the nearest library. Another advantage offered by wireless networks is the ability to provide computer network access in buildings not suitable for wired networks. In situations where concrete walls or historic structures are involved, wireless provides the best solution to network connectivity.[15] As digital cellular and packet-radio networks become more available, public libraries will be able to connect bookmobiles to online catalogs less expensively and more effectively. Wireless connectivity not only improves access to the library's own holdings, but it also opens up the entirety of "cyberspace" to bookmobile users. From bookmobiles, patrons can have access to the online catalogs of academic libraries, electronic texts, and other Internet resources. For campus-wide information systems, these projects demonstrate the feasibility of supplementing, or even replacing, wire-based connectivity.[16] The possibility of using packet-radio, spread-spectrum, or microwave technology for connecting desktop/laptop computers or PDAs to university mainframe computers offers greater flexibility and certain cost savings. By using campus-wide wireless networks, college and university administrators could save millions of dollars they would otherwise spend on wiring buildings and rooms. Indeed, the mobility offered by wireless networks may be better than traditional wired versions at meeting the computing needs and habits of students and faculty. On a funding level, wireless applications are particularly attractive to libraries since, once the initial equipment is purchased, ongoing operating and maintenance costs are minimal. Wireless equipment can be funded through modest capital expenditures or through grant funding. The aforementioned projects demonstrate that wireless data communications can have a place in libraries and educational institutions. They have proven that such applications are feasible, at least technically feasible, today--right now. The increasing installed base of CDPD cellular systems and spread-spectrum based wireless LANs should reduce communications costs associated with these technologies and make them more economically attractive for educational institutions. It should be noted that wireless data technologies are relatively new and that the commercial market for these products is still in an evolutionary stage. At this point, it is difficult to make specific recommendations regarding products or service. For example, AT&T's premature decision to halt production of the EO significantly curtailed the availability of the most viable PDA for library applications. Apple Computer's long-touted plans to provide wireless LAN capabilities for the Newton MessagePad may ultimately prove to offer a worthy device for client/server networks in libraries.[17] Many more new products and improvements to existing services are on the horizon. The potential pay-off for further research and development in wireless library applications is tremendous. Packet-radio and wireless LANs in particular offer librarians green pastures for future research efforts. Recent technological and industry advances favor increased commercial development in these areas. If the great potential of wireless data communications in higher education is to be achieved, librarians and educators will need to take the initiative in molding its development. Librarians should experiment with off-the-shelf products and configure them in unique ways to solve specific problems encountered in library work. It may be a matter of merely taking seemingly dissimilar pieces of a puzzle and joining them in different ways to create new and exciting wireless library applications. FOR MORE INFORMATION Additional information on wireless data communication in libraries can be found on the Wireless Libraries Homepage maintained at Auburn University Library. This site contains an interactive bibliography of printed publications and Internet resources relevant to wireless library applications. The WWW address is: http://www.duc.auburn.edu/~fostecd/docs/wireless.html. For a broader view of wireless research and wireless products, not all focusing necessarily on libraries or educational applications, consult the University of Washington's World-Wide Web Virtual Library: Mobile and Wireless Computing Homepage. The WWW address for this site is: http://snapple.cs.washington.edu:600/mobile/mobile_www.html. NOTES 1 Satellite transmission and data transmission over National Television Standards Committee (NTSC) broadcast television signals are not included in the discussion because of their excessive costs or current lack of applicability in library environments. 2 Steven J. Bell and Michael Halperin, "Testing the Reliability of Cellular Online Searching," _Online_ 19 (September/October 1995): 15-24. See also Steven J. Bell, "Online Without the Line: Cellular Technology for Searching On the Go," _Online_ 15 (September 1991): 15-25. 3 For a description of three such projects, see Catherine Suyak Alloway, "On the Road with Online: The Online Bookmobile," _Wilson Library Bulletin_ 66 (May 1992): 43-45; Mary Lou Pratt, "A Bookmobile Driver's Story," _Wilson Library Bulletin_ 66 (May 1992): 46-47; and Lori Logsdon, "Bookmobile Online Circulation Via Cellular Telephone," _Computers in Libraries_ 10 (April 1990): 17-18. 4 Edwin B. Brownrigg, Clifford A. Lynch, and Rebecca L. Pepper, "Packet Radio for Library Automation," _Information Technology and Libraries_ 3 (September 1984): 229-244. See also Clifford A. Lynch and Edwin B. Brownrigg, "Progress in Wide-Area Packet Radio," in _Proceedings of the 49th Annual Meeting, American Society for Information Managers_ (Medford, NJ: Learned Information, Inc., 1986), 167-177. 5 For detailed descriptions of the technical aspects of packet-radio, see Greg Jones, ed., _Packet Radio: What? Why? How?_ (Tucson, AZ: Tucson Amateur Packet Radio Corporation, 1995) and Clifford A. Lynch and Edwin B. Brownrigg, _Packet Radio Networks: Architectures, Protocols, Technologies, and Applications_ (Oxford: Pergamon Press, 1987). 6 For an example of continuing research and development in this area, see Edwin Brownrigg, "Continuing Development of California State Packet Radio Project," in _Proceedings of the ASIS 1992 Mid-Year Meeting_ (Silver Spring, MD: American Society for Information Science, 1992), 97-100. 7 Angela Gunn, "Wireless Communications: Connecting Over the Airwaves," _PC Magazine_, August 1993, 359-384 8 Kaveh Pahlavan, Thomas H. Probert, and Mitchell E. Case, "Trends in Local Wireless Networks," _IEEE Communications Magazine_, March 1995, 88-95. 9 Jodi Cohen, "Wireless LANs Hope To Get Second Wind With Upcoming Standard," _Network World_, 12 June 1995, 32. 10 Clifton Dale Foster, "PDAs and the Library Without a Roof," _Journal of Computing in Higher Education_ 7 (Spring 1995): 85-93. 11 Personal digital assistants (PDAs) are handheld computers which primarily allow use of Personal Information Management (PIM) applications such as keeping track of daily appointments and telephone numbers. Data entry is by using a "pen" to write on the screen, much like you would write on paper. Using sophisticated handwriting recognition algorithms, this information is converted into machine readable text. Instead of typing commands on a keyboard, users "tap" on dialog boxes and menu buttons to execute applications. For an examination of the early prospects of PDAs in libraries, see Donald T. Hawkins, "Have You Seen Your First PDA Yet?" _Online_ 17 (March 1993): 81-83. 12 The EO Personal Communicator 440 was equipped with a 20-MHz Hobbit microprocessor, 8 MB of ROM and 4 MB of RAM. With the cellular modem and nickel-cadmium battery, it weighed just over three pounds and measured 11" x 7". 13 Robert L. Jacobson, "Carnegie Mellon U. Urged to 'Exploit Technological Advances'," _The Chronicle of Higher Education_, 26 October 1994, A26-A27. See also Alex Hills and David B. Johnson, "A Wireless Data Network Infrastructure at Carnegie Mellon University," _IEEE Personal Communications_, February 1996, 56-63. 14 Thomas J. DeLoughry, "No Wires: An Alternative to Cabled Computing Passes a Test at Santa Cruz," _The Chronicle of Higher Education_, 7 July 1995, A15-A16. 15 For descriptions of two projects using wireless networks in historic structures and concrete buildings, see Karen Cummings, "Wireless Technology Brings the State Library of Iowa Online," _Computers in Libraries_ 15 (November/December 1995): 26-27; David Storm, "A Sprawling Library With Concrete Walls Looks for Better LAN Technology," _InfoWorld_, 30 March 1995, 68; David Storm, "Friday the 13th: A Horrific Day for Testing Wireless Network Solutions," _InfoWorld_, 27 March 1995, 80; and David Storm, "Wireless Links Prove Worse Than a Card Catalog for Giving a Reliable Connection," _InfoWorld_, 3 April 1995, 60. 16 For an overview of wireless technology as it relates to campus-wide networks, see Frank H.P. Pearce, "Wireless Communications - Come in Dick Tracy!" in _Proceedings of the CAUSE Annual Conference_, 1993 (Boulder, CO: CAUSE, 1994), 369-378. 17 Related to this is Apple Computer's recent FCC petition to request that a portion of the radio spectrum be set aside for unlicensed use under the National Information Infrastructure (NII). The petition calls for the creation of a new band of frequencies for high capacity, unlicensed wireless data, commonly known as the "NII Band." Specifically, Apple proposes that the Commission allocate 300 MHz of spectrum in the 5 GHz range, comprising the 5150-5300 Mhz band (which has been allocated throughout most of Europe for "HIPERLAN" unlicensed wireless local area networks) and the 5725-5875 MHz band (which currently is used by unlicensed technologies authorized under Part 15 of Title 47 of the Code of Federal Regulations; industrial, scientific and medical devices; and amateur operators). See Federal Communications Commission, _Petition for Rulemaking: The NII Band. RM-865_ (Washington, D.C.: Federal Communications Commission, 1995).