Learning Through the Internet: Lessons from the GNA-VSNS Biocomputing Course

Francisco M. De La Vega <fvega@gene.cinvestav.mx>
Department of Genetics and Molecular Biology
Mexico City 07000, Mexico


The global networking infrastructure can be an excellent medium to deliver education and training to satisfy the increased demand for educational services. This involves the implementation of online learning environments that should provide learners with the necessary resources to explore alternative pathways and develop their own style of learning with a wider choice in what, when, and how to learn. Two major paradigms of online learning environments currently exist on the Internet: the autonomous learning paradigm, which relies on knowledge servers scattered through the globe providing resources for self-paced learning, and the online apprenticeship paradigm, which relies on access to experts and peers via computer networks. In this paper I describe our experience with the online apprenticeship model, as implemented in the Globewide Networked Academy (GNA)--Virtual School of Natural Sciences (VSNS) Biocomputing course, which was delivered through the Internet in the early summer of 1995. In this course, we used three Internet resources: e-mail distribution lists, a WWW hypertext book, and a virtual classroom set up within the text-based virtual reality conferencing system for biologists, BioMOO. I discuss the advantages and disadvantages we found in this type of education with the current state of the art of the relevant technology. While most of the advantages are widely recognized, most disadvantages arise from technical problems and the immature state of some services.

1. Introduction

There is an increasing demand in educational services among the general population. This ranges from K-12, university curricula, and postgraduate courses, to corporate training, specialty knowledge and service learning. However, the availability of these resources and the time people can invest in these efforts is reduced. An alternative to traditional classroom education to supply the needed educational services, is to use the now widely available global Internet to design and implement online learning environments. Nowadays the availability of Internet access is widespread from the workplace, school or home, pointing to this medium as an ideal delivery system for distance learning on a global scale. This kind of delivery is quite different from the traditional approach of distance education, which involves costly dedicated teleconferencing systems and rooms. In comparison, the use of the Internet to provide online learning environments requires a much lower entry cost and can reach a much wider audience.

Online learning environments are an instance of computer-mediated communication (CMC). CMC and networking in general can promote long-distance collaboration between students and trainers across the globe. The integrated use of this technology offers many educational opportunities. However, effective design is essential to the success of any online educational enterprise, and a considerable amount of time and effort needs to be invested in items like resource allocation, creation of a syllabus, selection of the learning methodology, production of the course material, and evaluation planning [1]. Procrastination and loss of motivation pose a serious threat to effective learning via CMC. Thus, the design phase should consider these facts, and accordingly develop a stimulating environment. The ultimate goal is to provide students with the resources necessary for independent exploration and learning.

The type of education promoted by CMC allows learners the freedom to explore alternative pathways to find and develop their own style of learning. Students are allowed to take considerable control over their learning in terms of how they schedule both personal study time and time for any group interactions, how much personal contact they have with the instructor, and how they contribute to the class [1].

2. Educational paradigms in online learning

I have identified two major paradigms for online learning environments currently available on the Internet: the autonomous learning approach and the online apprenticeship approach.

The autonomous learning approach relies on the existence of online resources or "knowledge servers" that provide study material to be covered on a self conducted basis. The material found ranges from online articles, class notes, link repositories, and tutorials, to more sophisticated self-paced learning programs. This kind of learning closely resembles the correspondence or programmed instruction education promoted by the open university many years ago. This approach is also called Computer Assisted Instruction (CAI) and is based on easily assimilated units which are placed in a sequence that the learner can move through in easy steps. A unit of knowledge is presented, followed by a problem in the form of a test. If the learner solves the test correctly, he moves to the next bit of knowledge and the next test [2]. Much of the research on the use of the Internet and the WWW to develop online learning environments has focused on this paradigm, and there are numerous examples of its implementation ([3-5]).

In contrast, learning activity can also be accomplished and fostered by coaching or help offered by people who know the target skill (instructors) or who are in the process of learning (peers), by several CMC techniques. Individual learning combined with the support of knowledgeable others through processes of modeling, observation, and successive approximations is called apprenticeship [6]. Online apprenticeship, thus refers to apprenticeship mediated by access to experts and peers on computer networks. This is usually characterized by the following five attributes [7]: One-to-many and many-to-many communication; asynchronicity or time independence; place independence; text based presentation, and computer mediation. In this environment, online apprentices can build and share knowledge through goal-oriented learning interactions with peers, experts, mentors, and through the access to specialized sources of information [7]. This category of learning environments can also include periodic synchronous meetings either in a virtual reality environment or simply by Internet and WWW "chat" applications. This usually takes the form of an online virtual class or guest lecturing [8], which can allow interaction of learners with experts.

The online apprenticeship methodologies have also been called "Network Learning" (a term coined by Linda Harasim), to emphasize the fact that CMC can both enhance traditional forms of education and enable the development of entirely new forms of educational interactions and opportunities. Although network learning is mediated by the networked computer, the process itself is composed of human interactions and the formation of global learning communities is promoted. A carefully designed environment that provides instances of collaboration, coaching, reflection, and exploration is essential to support network learning. To individuals willing to work within the constraints and the potentials of CMC, network learning can provide a very powerful learning option [1].

3. A case study of online apprenticeship: The GNA-VSNS biocomputing course

3.1 Brief account of the BCD course.

A prototype online course on biocomputing was delivered via the Internet in the early summer of 1995 [9]. The course lasted 11 weeks, and was offered free of charge. It was organized by the Biocomputing Division (BCD) of the Virtual School of Natural Sciences (VSNS) [10], which is a member school of the Globewide Network Academy (GNA) [11]. The GNA is a federation of educational and research institutions whose purpose is to provide a central location in which students and teachers can find each other, and to provide administrative support services to aid instructors. A main reason for the existence of GNA lies in the belief that the traditional departmental structure of universities is reaching the end of its useful life and that new organizational structures are needed if universities are to provide education for the masses with the diminishing resources that are available to them [11]. GNA consists of independent members and affiliated schools (like VSNS), each having their own internal finances, support, course offerings and curriculum policies.

The BCD course brought together 34 students and 7 instructors from all over the world, and covered the basics of biological molecular sequences analysis [9]. Five authors from Germany and the USA prepared a hypertext book which was the core of the course material. The hypertext book included direct links to sequence analysis and databank search services available over the Internet. The course was aimed at students with mixed backgrounds in molecular biology, biomedicine, or computer science, to improve their skills in this area with an interdisciplinary curriculum. Special emphasis was placed on the use of various Internet tools, and the development of new teaching tools [9]. The application process was carried out through a WWW-forms interface and was highly selective. This allowed almost self-selection of the students, the acceptance of most applicants, and a number of participants we were able to handle [9].

The subject matter of the course was covered in small study groups (four to seven people) led by an instructor who reviewed a weekly chapter of the hypertext book, exercises and homework. Question and answer sessions were synchronously held weekly in a virtual classroom specially built at a text-based virtual reality environment. A basic premise in the design of the course was to use only the most basic and widely available Internet services and freely obtainable software, which would ensure a wider audience. For a more detailed account of the course, its philosophy, syllabus, and implementation, see De La Vega, Giegerich and Fuellen, 1995 [9].

3.2 Internet resources used in the BCD course

3.2.1 Automated e-mail distribution lists

Seven mailing lists were set up to provide students, instructors, and interested parties with an asynchronous information and discussion medium. Each list was devoted to a specific topic: a general discussion list, curriculum issues, technical issues, administrative matters, faculty discussion, and course updates sent to externals (see [14]). Apart from these public lists, private lists for each of the study groups were set up at the University of Bielefeld, Germany. These lists enabled discussion of group-specific issues, like scheduling of synchronous sessions or homework. A periodic course newsletter was also edited by the main course coordinator (Georg Fuellen, University of Bielefeld) and published every one to three weeks using the general list [9].

3.2.2 The WWW Hypertext Book

The hypertext book was assembled from chapters specifically contributed by five authors; most were edited in different formats (e.g., TeX) and later converted to HTML. The collected material was made available by a WWW server at Bielefeld, and two mirror sites in the USA and UK [14].

The GNA-VSNS Biocomputing Course WWW home page contained pointers to homework assignments posted by WWW or e-mail, archives of the mailing lists, transcripts of the faculty meetings, a newsletter archive, and other useful information related to the course. Additionally, each study group set up a WWW home page pointing to their personal home pages, when available, and group-specific announcements. Homework problems were assigned each week and distributed. Many students posted their answers on the WWW. Thus, WWW was a very important resource used in the course for asynchronous collaboration [9].

The entire hypertext book and the course archives remain available from the course WWW home page for further reference by the global academic community [9]. It is hoped this book can become a valuable resource for training in Biocomputing, usable by other courses, and/or as a source for autonomous learning.

3.2.3 Virtual classroom

An important tool of GNA-supported education efforts are the so-called MUDs (Multi-User Dimensions) or, more exactly, their object-oriented programmable variants, the MOOs [11]. MUDs are text-based, virtual reality environments that can be used as electronic conferencing systems, i.e., they allow synchronous, real-time communication among groups of people located in diverse parts of the globe. MOOs offer the additional possibility of constructing virtual educational environments and learning objects that are useful for interactive sessions [11]. The MOO-based professional environment used in the BCD course was BioMOO [12], an environment designed as a meeting place for biologists, founded and maintained at the Weizmann Institute by Gustavo Glusman. Among the unique features of BioMOO is "BioWeb," a WWW interface that allows the use of graphical objects to perform the usual commands, easy interaction with objects or characters, and integration of HTML links and resources within the MOO context [13]. A virtual complex for the BCD was constructed, including classrooms and instructors' offices. A set of tools was created for the didactic needs of the instructors, including simple blackboards, "slide" projectors, and text recorders. The latter were used routinely to keep transcripts of all weekly sessions [9]. Personal tutoring of students was possible during the instructors' "office" hours. Furthermore, two guest lectures with authorities in the study field were successfully conducted. The experience of discussing with otherwise difficult to reach experts, was very motivating for the learners.

Clearly the MOO currently imposes many usability restrictions. They are slow tools with a primitive textual interface and awkward for novices to use, though most of the students were eventually comfortable in MOO usage by the end of the course. While there are efforts devoted to improve MOO usability and to develop graphic interfaces to them (e.g., BioWeb), a completely redesigned VR-tool would be desirable. Efforts in this direction have been undertaken by commercial enterprises giving rise to a series of products that are competing to establish a de facto standard (e.g., Worlds Chat, Global Chat, etc.). In spite of the disadvantages inherent to text-based, virtual reality environments, and the fact that synchronous meetings on a global scale pose many scheduling difficulties, we found the help of a synchronous CMC system invaluable to enhance the learning process. The interactive sessions held at the MOO provided the study groups with the required feedback and coaching characteristic of online apprenticeship. They also provided a community feeling to the study group members, difficult to attain by asynchronous communication, which has been suggested as an important factor for motivation [15, 16]. It is expected that further technical developments in this area would enable the practical and widespread use of bandwidth-effective virtual reality systems in network learning.

3.3 Evaluation and perspectives of the course

The BCD course proved to be a successful experience in distance learning, as indicated by the collective judgment of organizers, instructors and students, expressed in a final closure MOO meeting and a mid-term survey [17]. Probably due to the stringent application procedure for the course, the drop-out rate was small, as compared to other previous free network learning experiences [9]. By the end of the course all 7 study groups, and 25 of the original 34 registered students, covered the text in considerable depth, and sustained lively discussions in the MOO.

The mid-term survey, conducted through a WWW form interface [17], revealed some interesting trends. The main problem the students faced was that the time required to keep up with the course pace was more than what they expected; most of them devoted two to five hours a week. Secondly, they experienced several network delays and/or failures. It was also noted that while the MOO sessions took too much time, few accomplishments actually resulted from them. However, most students found the course interesting and appropriate to their expectations and the group interaction enjoyable and useful. They reported being interested in further learning with this approach. Instructors also experienced a series of problems: homework was sometimes not delivered or delayed by the students, the mixed backgrounds resulted in problems due to the different goals of the biologists vs. computers scientists, and some inexperienced students initially found difficulty with the use and configuration of the network clients (e.g., WWW browser, MOO client, etc.).

A second edition of the course is programmed for the spring of 1996. Improvements in the hypertext book content, exercise sets, updated hyperlinks, and increased use of guest lecturing and multimedia elements are expected. More references to printed published papers and reviews in the area will be used. Pilot experiences with VRML didactic material, hypertext collaboration environments (e.g. HyperNews [18]), and CU-SeeMe [19] desktop videoconferencing are foreseen. More objective criteria to evaluate the course methodology and the usefulness of the learning experience would be desirable [15]. As a long-term goal, the depth and breadth of the hypertext book and course offer shall be expanded considerably to involve contributions from many more researchers and teachers. However, such an expansion can only take place if the project is managed by full-time staff, and/or run as a commercial enterprise, because the organizational load is considerable and the 1995 course has already stretched the time limits of the organizers [9].

4. Advantages and disadvantages of network-mediated distance learning

Some advantages and disadvantages of distance learning using Internet resources were apparent during the development of the course and have been discussed elsewhere [9]. Here, I will review some pros and cons of online apprenticeship as experienced in the BCD Biocomputing Course.

4.1 Advantages of network learning

From an educational standpoint, learners have wider choice in what, when, and how to learn. There is more freedom to explore alternative pathways to learning. Also, this kind of learning encourages interdisciplinarity by introducing people with shared interests coming from very diverse backgrounds.

There are several obvious practical advantages of network learning. There is no need for displacement from the workplace or home to attend the learning session.. Discrimination regarding gender, citizenship, cultural origin, etc., is avoided. Asynchronicity ensues better integration of the learning process to other activities. Some topics covered by online education are not taught at local universities. Nomadic instruction [9] helps to overcome scheduling problems of traveling instructors. Wherever they were, instructors usually found Internet access, could still respond to e-mail, and classes were given from conference sites.

Online learning provides savings in travel costs of instructors, can optimize the time of instructors and students, and allows access to remote knowledge and facilities. It is easy to integrate Internet resources with hypertext educational materials, and their content can be easily updated, indexed, and referenced by many Internet indexing engines. Often, the resulting educational resources stay accessible to the global community.

4.2 Disadvantages of network learning

The main disadvantages are related to the efforts required to overcome the technical problems that the available tools still pose for the nonexpert. There is still a need for more friendly and widely available network and client software for the tools used during the learning process (e.g., the MOO). Learners also have to cope with unreliable and slow telecommunication links. The integration of heterogeneous distributed material from diverse origin, and sometimes doubtful quality, into the course materials also poses many difficulties. Low stability of several WWW hypertext links further complicates the update of the references within materials. The WWW is by design a passive and non-proactive environment [15], though by complementing with other tools (e.g., mail and MOO), this deficiency can be eased. Furthermore, approaches aimed at providing more interactivity to the WWW (e.g. CGI, Java, JavaScript) can provide tools for the design of WWW-based proactive environments.

There are also many logistical problems. Due to the open and free nature of the BCD course, learners found it difficult to commit the appropriate amount of time to the course. There is still a lack of acceptance of network learning in the standard curricula. How will institutions and employers evaluate a virtual course included in a curriculum vitae? Mostly, this issue is related to a lack of awareness of network learning by the general public. Promotion of the accomplishments made in this field and support/certification by widely recognized academic institutions is essential to enhance the acceptance of this kind of education in standard curricula. How to identify and certify students that are never contacted in person (IRL, in real life) is still an open question. This is also related to the difficulty in tracking and assessing students' achievements to provide the appropriate feedback [15]. There are some research efforts in producing tools to track students and provide automated feedback based on expert systems that are promising [3,4].

Finally, an organizational overhead is clearly apparent by the organizers of these kinds of efforts. There is a need for specialized staff to carry out the different tasks needed during the development of the course material and the learning process. Many current efforts are carried out as pilot experiments on a part time-basis without dedicated staff or funding. Awareness by funding agencies of the need and importance of these pilot experiments or of full-scale implementations is still scarce. Alternatively, commercial enterprises can also be developed to fund the necessary research in these early stages of network learning with the aim of future profits.

5. Conclusions

We expect that Biocomputing will continue to serve as a major field of research on distance learning via the Internet. While the minimum technology necessary for these enterprises is available, new technologies are still emerging and thus it may be too early to make reliable conjectures about its future. As important as the technical problems are, the educational issues still to be solved to make this kind of education much more efficient and widespread. The experimental nature of computer-mediated distance education results in much work overhead for organizers and somewhat unpredictable outcomes. More studies and new technologies are needed to reduce the workload and the instructor/learner ratio, a main goal of this kind of education. While problems still remain, this approach can be very useful and attractive, if only for the intrinsic benefits of computer-mediated communication.

About the author

Francisco M. De La Vega is an Assistant Professor at the Department of Genetics and Molecular Biology of the Center for Research and Advanced Studies of the National Polytechnic Institute (CINVESTAV-IPN), in Mexico City. He is also Head of the Biosequence and Structure Analysis Unit, a computer resource for the molecular biology community of CINVESTAV and regional academic institutions, and webmaster of CINVESTAV's WWW server. e-mail: fvega@gene.cinvestav.mx; URL: http://www.gene.cinvestav.mx/~fvega


I wish to greatly acknowledge the help and advice received by Georg Fuellen and Robert Giegerich (University of Bielefeld), main coordinators of the course, and the enthusiastic participation of the hypertext book authors and instructors. Invaluable support provided by the consulting students is also greatly appreciated. In particular I wish to acknowledge the help of David Pisano and Chris Lilley in implementing the mid-term survey and compiling its results. We thank the Research Center for Studies in Structure Formation (RCSF), Jaime Prilusky, Gustavo Glusman, and the other BioMOO "wizards," the Globewide Network Academy and its Virtual School of Natural Sciences, Ken Schweller (CollegeTown MOO), the Molecular Biology Computational Resource at Baylor College of Medicine and the Manchester and North Training & Education Centre for technical assistance. Support for the course was received from the Stifterband fuer die Deutsche Wissenschaft, Essen, Germany. I also wish to acknowledge the support provided by CONACyT (Mexico) grant 4826N, the Department of Genetics and Molecular Biology of CINVESTAV-IPN, and the Aspen Center for Physics, for the facilities provided during the summer of 1995. The critical reading of this manuscript by Mandy Caird and Heinz Hemken, is greatly appreciated.


[1] Berge, Z. and Collins, M. (1995) Computer Mediated Communication and the Online Classroom in Distance Learning. In: Berge, Z.L. and Collins, M.P. (Eds.), Computer Mediated Communication and the Online Classroom. Vol. III: Distance Learning. Hamptom Press, pp.1-12.

[2] Tiffin, J. and Rajasingham, L. (1995) In search of the virtual class. Routledge, New York, 204 pp.

[3] Lai, M.C., Chen, B.H. and Yuan, S.M. (1995) Towards a new educational environment. World Wide Web Journal 1(1)221-230.

[4] Hubler, A.W. and Assad, A.M.(1995) CyberProff: An intelligent human-computer interface asynchronous wide area training and teaching. World Wide Web Journal 1(1)231-238.

[5] Blumberg, R.B. (1994) MendelWeb: An electronic science/math/history resource for the WWW. Proceedings of the Second International WWW Conference '94. Chicago, Ill, October 1994. Volume I, pp.449-457.

[6] Teles, L. (1993) Cognitive Apprenticeship on Global Networks. In L. Harasim (Ed.), Global Networks: Computers and International Communication. Cambridge. MIT Press.

[7] Harasim, L. (1990) Online Education: An environment for collaboration and intellectual amplification. In L. Harasim (Ed.), Online Education: perspectives on a new environment. New York. Praeger.

[8] Cotlar, M. and Shimbukuro, J.N. (1995) Stimulating learning with electronic guest lecturing. In: Berge, Z.L. and Collins, M.P. (Eds.), Computer Mediated Communication and the Online Classroom. Vol. III: Distance Learning. Hamptom Press, pp.105-128.

[9] De La Vega, F.M., Giegerich, R. and Fuellen, G. (1995) Distance Education through the Internet: The GNA-VSNS Biocomputing course. In: Hunter, L. and Klein, T.E. (Eds.), Proceedings of the 1996 Pacific Symposium on Biocomputing. World Scientific Publishing Co., pp. 203-215.

[10] Virtual School of Natural Sciences WWW Home-Page URL: http://uu-gna.mit.edu:8001/uu-gna/schools/vsns/index.html

[11] Butts, C., Reilly, C., Speh, M., and Wang, J. (1994) WWW and the Globewide Network Academy. Proceedings of the First International WWW Conference, Geneva, May 1994. URL: http://www1.cern.ch/PapersWWW94/speh.ps; URL (GNA): http://uu-gna.mit.edu:80001/ uu-gna/

[12] Anderson, C. (1994) Cyberspace Offers a Chance to do "Virtually" Real Science. Science 264:900-901. BioMOO Home-Page, Weizmann Institute, Israel. URL: http://bioinfo.weizmann.ac.il:70/1s/biomoo

[13] BioWeb MOO/WWW Interface, implemented by Gustavo Glusman, Eric Mercer, Jaime Prilusky and Paul Hansen. URL: http://bioinformatics.weizmann.ac.il/bioweb/

[14] GNA-VSNS Biocomputing Course Home-Page. URL: http://www.techfak.uni-bielefeld.de/bcd/

[15] Brewer, P.W., Gibson, E.J., Dholakia, A., Vouk, M.A., and Bitzer, D.L. (1995) Advanced Computer-Based Education on the World Wide Web. Poster Proceedings of the 4th International World Wide Web Conference. O'Reilly & Associates, Inc.pp. 34-35. URL: http://renoir.csc.ncsu.edu/MRA/Reports/AdvEduWeb.html

[16] Archee, R. and Hill-Duin, A. (1995) The World Wide Web and Distance Education: Convergence or Cacophony? Proceedings of the AUUG'95 & Asia-Pacific World Wide Web Conference & Exhibition. URL: http://wwww.csu.edu.au/special/conference/apwww95

[17] Pisano, D.P. and Lilley, C. (1995) Summer 1995 BCD Student Questionnaire Results. URL: http://www.techfak.uni-bielefeld.de/bcd/Adm/eval95.html

[18] LaLibertie, D. (1995) Collaboration technologies on the Web. URL: http://union.ncsa.uiuc.edu/HyperNews/get/www/collaboration.html

[19] Dorcey, T. (1995) CU-SeeMe Desktop VideoConferencing Software. Conexions, Vol9(3), March. URL: ftp://cu-seeme.cornell.edu/pub/CU-SeeMe/html/DorceyConnexions.html