+ Page 9 + --------------------------------------------------------------------------- ####### ######## ######## ########### ### ### ## ### ## # ### # Interpersonal Computing and ### ### ## ### ## ### Technology: ### ### ## ### ### An Electronic Journal for ### ######## ### ### the 21st Century ### ### ### ### ### ### ### ## ### ISSN: 1064-4326 ### ### ### ## ### January, 1995 ####### ### ######## ### Volume 3, Number 1, pp.9 - 16 --------------------------------------------------------------------------- Published by the Center for Teaching and Technology, Academic Computer Center, Georgetown University, Washington, DC 20057 Additional support provided by the Center for Academic Computing, The Pennsylvania State University, University Park, PA 16802 This article is archived as BRACK IPCTV3N1 on LISTSERV@GUVM (LISTSERV@GUVM.GEORGETOWN.EDU) --------------------------------------------------------------------------- ACCESSING THE DATA TO CREATE THE KNOWLEDGE: A CASE STUDY FROM FORESTRY C.L. Brack and A.C.F. Trevitt, Department of Forestry, School of Resource and Environmental Management, Australian National University Canberra, A.C.T. 0200, Australia. INTRODUCTION There appears to be a new problem facing managers. This problem is a new form of paranoia (Annon, 1993) and stems from the belief that somewhere in the vast mountain of data available in private and commercial databases and other information sources is exactly the item of data needed. The new management fear - information paranoia - is that this data will never be found. Schrage (in Annon, 1993) argues that the computer and modern technology has caused the productivity of managers to drop substantially as they search for this elusive data. Our concern, and the focus of this paper, is that this paranoia is no less likely to be affecting foresters. For example, US Vice-President Gore's report on the USDA Forest Service (quoted in Olson 1994) noted that an administration manual that used to fit in a shirt pocket now takes up an entire wall of bookshelves. The USDA Forest Service has also seen this major problem area. In an overview of their resource inventory needs (USDA 1992), four + Page 10 + long term issues "stand out". Ranking alongside the issues of pollution, dwindling forest resources and increasing population pressures, is a concern about the "proliferation of information". However, we believe that the issues and opportunities raised here in relation to the Forestry profession are generally applicable. Foresters in the US are daunted in their search for an understanding of public perception as they attempt to review newspaper articles. A search on the key word "forest" in electronic databases available through FOCLC/FirstSearch or CitaDel found over 8000 newspaper and magazine references in the US in the last 12 months! Even restricting the search to refereed journals available in international libraries would find thousands of articles through CD-ROM or on-line electronic services like Uncover. Another substantial quantity of data is also available through newly developing digital publications appearing in the Internet. Smith (1993), who himself has published through electronic media, reported that there are over two million computers hooked into the Internet with an estimated 10,000 biologists reading and participating in some of the 250 newsgroups or 100 mailing lists. Computer and associated technology has meant that huge quantities of data can be made available. But acquiring or getting access to this data is only the first step - if the process stops here, then productivity will dramatically drop and communication costs dramatically rise as all the databases are searched. Schrage (1993) said that you can only consider the computer as a data processing machine if you also consider the motor vehicle simply as a gas processing machine! The computer, and its related technology, must be used to find new relationships with this data just as the motor vehicle was responsible for the creation of new communities and relationships like the suburbs and the shopping mall. Data accessed by computers can be transformed into "information" (i.e. data in a useful form) by selecting, sorting and linking the relevant data to allow patterns to be found. Once the patterns are found, they can be understood and applied as "knowledge" to allow improved actions and decisions to be made. Computers can help with all three of these stages: gathering data, linking it to form information, and applying it as knowledge. The earlier examples demonstrate the usefulness of the computer in gathering or accessing huge volumes of data. Saarikko (1994) gives directions to access some of the many electronic publications, discussion groups, electronic archives and databases as well as discussing network access and navigation tools. + Page 11 + This paper reviews initiatives taken in Forestry education at the Australian National University (ANU). Two case studies are discussed. They demonstrate the use of the computer to link data from a range of sources to make it informative and help in the creation and presentation of knowledge. A final section discusses these approaches in the wider field of ongoing professional development and communication. INFORMATION CREATION: A CASE STUDY There are few textbooks suitable for the study of fire science and management in the Australian environment. However, this does not mean that there is a lack of data relating to this topic (Trevitt, 1994). There is a considerable quantity of data circulating within the forestry profession and elsewhere in the form of historical documents; pamphlets; loose-leaf articles; recent internal reports; operational fire management plans; guidelines; as well as professionally refereed scientific papers. This data is invaluable, but is largely an uncompiled and inaccessible archive of miscellaneous materials. To overcome some of these access problems, the Forestry Department at ANU, began compiling some of the more readily obtainable documents into computer files in 1992. Both text and graphical material were included in this compilation. This information was then made available to students of the Fire science and management course in 1993. Students could access this information from a network of Macintosh computers with user-friendly menus (Trevitt, 1994). The value of having these course materials and data in an electronic format was generally appreciated by the students with usefulness being rated as fair, good or better by 83% of those that responded to a questionnaire (Trevitt & Sachse-Akerlind, 1994). Making these information resources freely available to the students allowed the course to progress from a traditional subject oriented curriculum towards a skill- oriented Problem Based Learning approach (Baud and Fellitti, 1991). Students were encouraged to use the information technology and the course resources to retrieve, examine and visualise the weather patterns relevant to fire management. Such skills can be very important when controlling a wildfire, but the opportunity to develop them rarely presents itself. The factors that produce 'blow up' days conductive to conflagration scale fires are rarely experienced, but the relevant data can be made available. These data, in conjunction with appropriate practical-based problems could be used to examine and test some of the class developed and prevailing fire management policies thereby fostering skill development in policy making. + Page 12 + This type of data and access is also useful to forest managers or researchers generally, and in 1993 an international information retrieval network called FireNet was established (Green et al, 1993). At this stage, FireNet operates independently of the electronic course resources referred to above, but it is our intention to merge them in the near future. FireNet can be accessed from computers anywhere in the world through a number of protocols. For example, through the World Wide Web protocol at the following address: "http://www.anu.edu.au/Forestry/fire/firenet.html". For a discussion of this, and similar protocols, see Saarikko (1994) and Smith (1993). While FireNet is an information exchange network, it also permits on-line storage and retrieval of information. Its initial contents included a bibliographic database, some software packages, access to current weather satellite images, and various documents including those prepared for the Fire Science and Management course. This information can be updated and improved and new linkages formed to improve the generation of knowledge. KNOWLEDGE: A CASE STUDY The above case study demonstrates the power of computer technology to access and store vast quantities of data, and then make them available in a meaningful manner. However, it is still difficult to distil this data and the observed patterns into a meaningful set of relationships that can be remembered and used. For example, the substantial efforts made in the development of tree shape and taper models can be utilised to improve volume estimation through taper sampling approaches like importance sampling (Gregoire et al., 1986); centroid sampling (Wood et al 1990); and control-variate sampling (Van Deusen 1987, Van Deusen & Lynch 1987). These techniques are extremely efficient and estimate tree volumes precisely and without bias from only one diameter measurement. However, an understanding of the concepts involved in taper sampling is often difficult to convey. It is also unlikely that simply having easy access to various case studies where the techniques are applied, will lead to an improved understanding. An understanding of the techniques, and their differences, requires students to visualise a tree being represented by the three-dimensional shape produced by a rotation of the plot of a proxy taper function. The sampling techniques are designed to relate these two shapes and the quality of this relationship is used to predict the tree volume and precision of the estimate. + Page 13 + In an experiment carried out during the 1994 undergraduate Forest Mensuration course at the ANU, importance sampling was explained in a conventional manner using a series of overhead projections displaying a proxy taper function, the transformed proxy sectional area function and then the corresponding proxy volume function. This explained how the proxy volume function was related to a sample tree A sample height was selected (proportional to the volume below that height) and the sectional area at this sample height was measured on the sample tree and then the proxy sectional area and proxy volume functions updated by the ratio of the tree sectional area : proxy sectional area to produce a corrected volume estimate. The following week, a digital animation of the same concept was shown. This animation showed a tree and generated an appropriate taper function. The taper function was rotated to form a three- dimensional shape and students could easily visualise how they were related. The appropriate sample height was selected and again students could see the corresponding position on the sample tree, so the ratio correction looked logical. At the completion of the animated movie, the original overhead transparencies were again shown to the class. A question and answer period indicated that the majority of students understood the process. Approximately one month later, students were asked to describe the importance sampling technique during their mid-term examinations. All students who attempted this question drew pictures or diagrams that were derived from the digital animation - none of the overhead figures were used. It appeared that the improved visualisation, three-dimensional effect and movement of the animation assisted the students in understanding and recalling the underlying concepts of this relatively complex sampling technique. PROFESSIONAL DEVELOPMENT FireNet has grown from less that 80 subscribers in July 1993, to over 360 in December 1994. This growth demonstrates that an electronic network can be used by professionals as a source of data, information and knowledge to the professional. This type of knowledge resource can continue to evolve and grow as new users access, link and incorporate new data and information sources. Similarly, the Forest Mensuration information resources can be made increasingly available and comprehensible through the use of computer facilitated storage and presentation. However, ensuring that these resources and facilities remain part of the answer and not another data black hole to trap the unwary, is a time-intensive task. A group of people or institutions is needed to compile and link the information as it becomes available and relevant, and to ensure that + Page 14 + its presentation is useful. Green and Croft (1994) consider that this collaboration can also make use of the Internet network by the formation of a Special Interest Network (SIN). A SIN is the modern equivalent of learned societies and provides a thematic focus for information, stability, organisation, quality control and standardisation (Green and Croft 1994). Information exchange and communication is an obvious function of a SIN, but the provision of a standardised, organised and quality controlled database by the SIN is particularly valuable. This database would not only promote the co-ordination and exchange of scientific information, but could also help ensure efficienct research by directing people away from a re-invention of the wheel, and towards the application of results. For example, each forestry organisation in Australia has its own set of practices for fire management and mensuration. In fact, each state in Australia has its own railway track gauge which significantly hampers progress! Often, the different practices are the result of ad hoc decisions or ignorance of other options. Even simple mensuration terminology is often re-invented, causing needless confusion and requiring the development of appropriate codes (Australian Forestry Council, in prep). As idealists, we wish to learn from history! Active SINs may be involved in the discussion, evaluation, modification and standardisation of practices (and definitions) to a universal advantage. As improvements are found, relevant material can be added to the database for wide circulation. Case studies could also be easily added simply by writing forest service internal reports in a standard format for inclusion in the database, e.g. Standard Generalised Markup Language (American National Standards Institute, 1991). The database, collaboratively managed by the participants of the SIN, and carefully managed for quality control and standards could serve as the text book for the training of student foresters and as the source of ongoing professional development for any forester with access to the Internet. Conclusions A forest management system (or indeed any information system) must not simply provide MORE information (Brack 1991). This will simply increase information paranoia. The goals of any information system should be to: allow better information use; support the user's knowledge and effectiveness; enhance the user's view of the whole problem; be responsive to changing needs and be user friendly (Brack + Page 15 + 1991). Fortunately, appropriate computer technology, a computer network and the basic standards for a Special Interest Network to create a database environment now exists, where such goals may be met. Recent experiences at ANU in developing Fire Science and Management course resources, FireNet and a prototype Mensuration resource database suggest that information paranoia can be overcome. Computers and technology can facilitate sorting data, compiling information and personalising knowledge to hence enhance interpersonal communication. REFERENCES American National Standards Institute (1991) Electronic manuscript preparation and markup: American national standard for manuscript preparation and markup. National information standard series 1041-5653. Annon (1993) Information Paranoia. Your Computer. April 1993: 16-18. Australian Forestry Council (in prep) Code of mensuration practice. Prepared by Research Working Group #2: Mensuration and Management. Brack, C.L. (1991) Knowledge-based approaches to forest operation scheduling problems. PhD thesis. University of British Columbia pp 132. Green, D.G. and J.R. Croft (1994) Proposal for implimenting a biodiversity information network. In Linking Mechanisms for biodiversity information. Proceedings of a workshop for the biodiversity network. Base de Dados Tropical, Campinas, Sao Paulo, Brasil. Gregoire, T.G., H.T. Valentine and G.M. Furnival (1986) Estimation of bole volume by importance sampling. Canadian Journal of Forest Research 16: 554-557. Olson, B. (1994) Oracle announcements. Your Computer January 1994: 12. Saarikko, J. (1994) Forestry Information Resources on the Internet. Paper presented at the EUFORIS'94 conference. Available through FTP or URL: "http://www.metla.fi/~saarikko/doc/forestry-resources.html" Smith, U. (1993) The biologist's guide to Internet resources, volume 1.7, 10 November 1993. Available through FTP or URL: "gopher://calypso- 2.oit.unc.edu/11/../.pub/academic/ecology+evolution/bioguide". + Page 16 + Trevitt, A.C.F. (1994) Computer-facilitated learning in forest fire management. In: M. Ryan (ed) Proceedings of Asia Pacific Information Technology in training and education conference. Brisbane, Australia. 28 June - 2 July. pp723-728. Trevitt, A.C.F. and G. Sache-Akerlind (1994) A District fire committee simulation in the professional forestry curriculum: a case study of computer-facilitated problem- based learning. In: S.E. Chen, R. Cowdroy, A. Kingsland and M. Ostwald (eds.), Reflections on problem-based learning. Australian Problem Based learning netwrok, Sydney. pp 347- 369. USDA (1992) Forest Service Resource Inventories: An overview. USDA Forest Service Economics and Recreation Research Paper. September 1992. Van Deusen, P.C. (1987) Combining taper functions and critical height sampling for unbiased stand volume estimation. Canadian Journal of Forest Research 17: 1416- 1420 Van Deusen, P.C. and T.B. Lynch (1987) Efficient unbiased tree-volume estimation. Forest Science 33(2): 583-590. Wood, G.B., H.V. Wiant, R.J. Loy and J.A. Miles (1990) Centroid sampling: a variant of importance sampling for estimating the volume of sample trees of radiata pine. Forest Ecology and Management 36: 233-243. ------------------------- BIOGRAPHICAL SKETCHS ---------------------------------------------------------------------- 5. Copyright Statement ---------------------------------------------------------------------- Interpersonal Computing and Technology: An Electronic Journal for the 21st Century Copyright 1995 Georgetown University. Copyright of individual articles in this publication is retained by the individual authors. Copyright of the compilation as a whole is held by Georgetown University. It is asked that any republication of this article state that the article was first published in IPCT-J. Contributions to IPCT-J can be submitted by electronic mail in APA style to: Gerald Phillips, Editor IPCT-J GMP3@PSUVM.PSU.EDU