by Rodney Hoinkes
The interest in networked multimedia for teaching has grown exponentially in the last few years. In a large part, this has been fueled by the emergence and promise of hypermedia on the World-Wide Web (WWW), a collection of distributed servers on the internet, each offering public access to a variety of information resources. The most critical break from previous network communication technologies has been the direct integration of imagery and text, allowing more varied, complex, and contextual ideas to be transmitted from one place to another almost anywhere on the globe. This is the first time that the spatially-oriented fields have had a common vehicle for publishing their work in an effective electronic networked manner.
Since the facilitating technology has been put in place, an exponential growth in data, information, and sometimes knowledge is starting to appear. The process of gaining insight into how it may influence education has now tentatively begun, both for the teacher and the student.
This paper will describe a number of direct observations from ongoing development and usage of the World-Wide Web Virtual Library for Architecture and Landscape Architecture, and observations from teaching a variety of courses which have utilized the World-Wide Web in Architectural and Landscape Architectural design.
Keywords: World-Wide Web, education, learning, teaching, internet, architecture, landscape architecture
Few educators would say that their students were "taught" something by having access to chunks of information in multimedia databases. Many however, are of the opinion that such access has the potential to facilitate a wider range of thinking and dialogue, key constructs in building a students ability to learn, rather than just be taught (Barrett, 1992). Multi- and hypermedia must not be looked upon as an either-or situation (with respect to more traditional teaching methods), but as a new participant in the educational process (Hammond, 1993).
Multimedia can provide for random access to information, allowing a multitude of perspectives or readings on the information at hand. This ensures that learning is an active process on the part of the student, reminding them that one solution and way of looking at ideas is not the only one. Many ideas have complex relationships to others ideas, relationships that linear presentations have difficulty representing - particularly those that are spatial and temporal in nature. Students are provided with an environment that allows them to learn at a speed more appropriate to them, as well as applying a personal outlook to their inquiry. This offers a means of building experience with ideas - personal experience - and through personal experience comes the building of trust and confidence in the understanding of ideas (Papert, 1993). Rather than facts and rules alone, students actively gain experience in learning, a process that is increasingly important in a world where the static notions change too rapidly to be of much long-term benefit.
The global village, virtual communities, and collaborative work - all promises of the new networked age. We can communicate and collaborate with colleagues around the world, join international teams to solve problems or research ideas of mutual interest and share our ideas with the world while sipping cappuccino on the balcony. These concepts are nothing new for some of you who, like myself have been using electronic mail and file transfers to talk and work with others for years. What is new however, is a change in the general accessibility and usefulness of these networks resulting from the merger of network communications with the capabilities of multimedia. No longer are we limited to text in email or the transferring of files with only their names as context. We can finally merge our verbal and graphical representations and indicate more complex non-linear relationships. Collaborators no longer have to have a separate copy of the information on each machine, and it only running on one type of system (as is the traditional mold of multimedia products). The widespread introduction of HTML (Hyper-Text Markup Language) abstracted the content of the material from any specific hardware or software, and the WWW (World-Wide Web) provided a common means of machines accessing such material over the internet - a platform-neutral format and protocol. Network-based resources have always been of interest, allowing individuals to share ideas and material around the world, but until the introduction of the World- Wide Web, they had largely kept information in separate packages of either text or images, or fell back on traditional approaches of combining them but being platform-specific.
Since the beginning of the '90s, the World-Wide Web has existed, initially as a common means for discussion and publication within the realms of High Energy Physics, originating at CERN in Switzerland. This provided a means of rapid publishing in a field that could not afford to duplicate effort and costs. In 1993, the NCSA (National Center for Supercomputing Applications) released a graphical browser for accessing information on the web which fueled an exponential growth in the use of this environment. Statistics indicate that in June of 1993 there were 130 servers while now there are 27,000, with the population said to be doubling every two months (Matthew Gray, 1994 and Business Week 2/27/95 p.80). It is clear that this is not simply a passing fad, with exponential growth and most new operating systems being shipped with software that allows web access, the web is rapidly becoming a common environment in our computing lives.
This does however have implications for our computational systems, primarily in the realms of bandwidth, but also in terms of fundamental access and content structuring. It is one thing to transfer a few bytes of text here and there and transmit it to local machines by modem when the user connects but the inherent interactivity of multimedia places heavy demands on networks. Fast modems (14.4k and up) and direct network protocol support (SLIP, PPP) are the minimum, fast 10MBit/sec direct connections the baseline. But none of these come close to the bandwidth typically utilized on a local machine between hard disk and monitor for multimedia endeavours.
Transmission Medium Bandwidth Estimates Local Bus ~30MBytes/sec ATM ~7.5MBytes/sec Ethernet ~1.25MBytes/sec 14.4k Modem ~0.0018MBytes/sec
It seems clear that multimedia has its potential benefits but with the disparity between local multimedia and network multimedia, the interest seems hard to justify and the growth of this environment unwarranted.
Perhaps this is so on technological scales, but on social ones, the reasons for acceptance are powerful. Users are no longer confined to their interaction with a machine, rather it can become a medium for communication with other people and other ideas around the world. The inherent limitations on bandwidth often means material is abstracted in more useful and quickly digestible manners - the glitz of multimedia is sheared away to focus on the content. This also means material is potentially faster and easier to develop and publish. Furthermore, content has been separated from the specific machine interface and allowed the material to become accessible to more people working in heterogeneous environments. Suddenly anyone can publish on the network and anyone can read it, eliminating much of printing and shipping costs.
The Centre for Landscape Research (CLR) at the University of Toronto School of Architecture and Landscape Architecture has long been developing and integrating highly interactive multiple media design exploration tools into the curriculum of the school (Danahy, 1990, 1991). In 1992, these tools were applied to a public multiple media exhibit "Opening the Gates of 18th Century Montreal", developed with the Canadian Centre for Architecture.
Here, the historically reconstructed 3d models were integrating with historical paintings of the city and its social life along with thematic mapping of its ethnic and commercial changes over the century (Hoinkes and Mitchell, 1994). These were presented in an interactive kiosk within the traditional exhibition setting. The exhibit proved to be highly successful (and was extended several months), with the kiosk allowing the public a more personal interaction with the historical information, actually touching digitized copies of the imagery to zoom in for more detail or strolling down the reconstructed main street - possibilities that have not existed in conventional settings. While the CLR, acting as experts running the system, had employed these technologies with the public many times in the past, the possibilities for allowing people direct access to the material to built their own experience and interpretations had made a powerful impact.
In the fall of 1993, while discussing these concepts with other colleagues at the university, mention was made of the World-Wide Web (WWW) and HTML (Hyper-Text Markup Language), where some simple multiple media work was being done to deliver course material to students in a more self-directed manner than can be obtained in some classes with up to 300 students. At that point, the concept was mearly one of interest but of no practical use as our design work is inherently three-dimensional and temporal, requiring a higher level of interactivity than the text and bitmap world could support.
The spring of 1994 found that situation changing due to necessity. While I was working on my doctoral thesis at Harvard, my particular teaching skills were required in a design studio in Toronto. As is increasingly the case, time and money would not permit me to continually fly back and forth. At this point we turned to a variety of solutions - adding improved internet connectivity to our primary teaching software, and communicating course work through the World-Wide Web. While our own software was used for direct tutorials and design reviews with students during class time, the web was used to post course notes, tutorial handouts, historical and site imagery, and to regularly compile example imagery of student work. This provided the students with access to my knowledge out of class time (when traditionally they could have come to me for direct questioning), and me access to their progress whenever I had time to look in on their work (as an instructor could do traditionally by passing through the studio area and seeing what was tacked up on boards or left on tables). While developing the online documents for the students, I found myself looking around the web - surfing - for useful related material that I could point them to. Indeed, I found a few references of some use to them and was easily able to link them in much as an instructor could put certain books on reserve in the library - but without the limitation of only one student being able to access it at any time or taking it out of circulation for other students. Seeing the potential for a variety of links to other documents that would be relevant to my students and the hap- hazard method of finding them, I decided to devote time to the establishment of a primary resource database for the fields of Architecture and Landscape Architecture on the internet.
The World-Wide Web Virtual Library is an attempt at a distributed subject catalogue of information on servers around the internet. This was brought about to aid users facing the endless list after list that was appearing. Each server had links to the ten servers it knew, each in turn having maybe one or two new links that the previous one did not - it was almost impossible to find anything or even to know if something you were interested in even existed on the net. The Virtual Library Subject catalogue for Architecture and Landscape Architecture was established at the Centre for Landscape Research (CLR), in June 1994 as a response to my previous explorations. Each link in the database is a reference to a document on a server somewhere on the network, with the universal addressing scheme hiding the specific location from the immediate user. The document references are compiled from personal surfing, other lists, and new server information being emailed directly to me. At that time, there existed between 50 and 80 hypermedia document topics related to Architecture and Landscape Architecture, generally as introductions to particular schools or research groups, and an average of 150-200 network accesses to it per day. At the present time (February, 1995), the database of links numbers just over 600 and an average of 5-700 accesses daily.
This provides a wide variety of accessible architectural information on the web under a variety of categories. The primary classes to date include organizations, topical works/exhibits, software information, and a miscellaneous category. Organizations account for ~30% of the links while published material accounts for ~50%, software information and notes ~10% and a miscellaneous mix (News Groups, Mailing Lists) accounts for the remaining 10%. The general tendency seems to first to establish a presence on the web, have your name and basic organizational information accessible. Several people have asked me why their organization is not in the Virtual Library listing. When they realize that the listing is only for those with web-accessible information, many have gone back and pursued the process of making their organization accessible. This first stage is often about recognition, being in the yellow pages so people know you exist, referrals by chance seeing of your name, a marketing tool, and staking your land in the cyberspace equivalent of a land rush. A second phase that often follows in educational settings is the course handout and tutorial - most commonly technical in nature and related to computer software introduction and tutorial, which can be directly accessible while working on the computer. A common third phase (and often the first or second with governmental agencies) is the addition of more general content that is not site-specific but of value to the general topical community. This includes reports, papers, and exhibits by both faculty and students. This now accounts for 50% of the links in the database and an equal 50% of the accesses through the Virtual Library. These links are often the more interesting multimedia content on the net, one part research/technical papers, one part site documents, and one part historical project documentation and reconstructions.
This Virtual Library differs from several other Architectural and Landscape Architectural resource indices on the web in a number of ways. The primary difference is that the subject is maintained by a group within the field, having a personal and professional interest versus a general library trying to track everything. As an analogy, compare a local reference library to a University's specific Architectural one. Furthermore, most indices provide only links to primary documents such as a School's Home Page but rarely to the interesting documents within it, meaning that a user must randomly search all schools to find out if they have any specific information within them. The Virtual Library in Architectural and Landscape Architecture attempts to list all documents of content, even below the organizational level.
As an added capability, all entries in the library are keyword-accessible using the language of the disciplines. This is not possible for general indices to due to specific knowledge and time constraints. As a result, most other general indices contain between 20 and 40 references versus the 600 in this specific one.
There are however problems with the document references found within the library. The current problem with these documents is frankly, they are conceived of largely as documents. There is not enough content yet for useful multiple links, fully exploiting hypermedia capabilities and there is not enough support and experience with interactivity over low-bandwidth links to make the information overly compelling as multimedia. Why then are more and more groups publishing information and more and more groups accessing it? The issues seem to be ones of economy of scale and infancy of tools with regards to publishing. The domains of Architecture and Landscape Architecture are largely educational or professional and primarily small-office or small-class oriented. In the professional realm, there is a strong focus on the projects at hand and little value in products for learning, only those that have immediate value to their work - a poor candidate for a commercial multimedia market. On the other hand are those in the educational domains who provide broad backgrounds, lectures in history covering centuries and dozens of architects and hundreds of details. For these purposes individual projects or architects are only of a certain value, often not enough to justify the costs for specific multimedia material that may run on a few machines today but will be outdated with new operating systems or hardware only two years down the road. This paints a rather grim but I will argue realistic view of a non-market for multimedia in these fields. On the other hand, the network-based multimedia of the World-Wide Web exploits the networks that most educational institutions have in place, is accessible from almost any type of computer system (now and in the foreseeable future), and is relatively inexpensive. At this time, CD-ROM Multimedia technology has existed for more than five years and is only now starting to become popular and useful; in the same vein, the web has only become known over the past year and is only beginning its journey through the technological horizons to usefulness. The impact of this is that the tools for content creation are in their infancy but with commercial solutions finally arriving. This will provide the means for people to publish the document-style content more readily but does little to address the hypermedia and interactivity potentials. For these we must simply have some patience; the hypermedia cross-referencing potential grows as the body of content does, providing an ever-richer environment for individual exploratory learning, while the techniques for interactivity are only now being pushed and stretched towards useful bounds.
The world of networked multimedia (WWW) is still largely one of promise and a start at implementation. In educational terms we can see potentials in "discovery learning" and in technical and growth terms, potential for a rich future body of content on which to draw. In the mean time what does it offer of substance to education? To this end, I will engage in some conjecture drawn from the teaching use I have been party to as well as usage and commentary of users of the Virtual Library in Architecture and Landscape Architecture.
Firstly, let me state that everyone expects too much from such infant technology, and has trouble with it when it does not live up to expectations. What this means is that we should more fully understand the approach and underlying concepts behind such software before trying to fit the round peg into the square hole. The web is not the glitzy multimedia environment we are accustomed to from CD-ROM, it is "just- in-time" publishing, rapidly developed, quickly posted, and thankfully can be just as quickly updated. It is largely free, often haphazardly organized, and only marginally interactive at this time. On the other hand, it is easy to add to, in terms of content and interactive capabilities, and is an enormous cross-disciplinary resource.
In that first studio in the spring of 1994, thankfully expectations were not very high, everyone was just trying to make a less-than-ideal situation work. Being a first attempt, the techniques worked just well enough to survive the class, both on my part and from the students perspective. What did occur of more lasting benefit, are some outlooks on the nature of networked multimedia in education; the web must become second-nature to the students, a sketch-book that they continually refer to for new ideas, links, and updates, not simply another place to have to look. The web is not a short-term investment, it is a long- term one, a new environment for learning, less so one for teaching. This means that educators must change their outlook as well, from oracles describing rules to guides in pursuing ideas. Educators themselves must explore and learn about learning - how one builds a knowledgebase as an act, not simply the transfer of the knowledgebase itself. This requires more time on the part of instructors as does the preparation of material in electronic form, especially anything multimedia in nature. Educational institutions must realize this and provide for the time and additional support resources that are needed, all of which can payoff quickly. Think globally, act locally, can also apply to information; if each school spent the time to post one document of common interest, dozens if not hundreds of resources would suddenly become accessible that all could share.
Since those initial exploits, the web has become a standard part of our school's educational practice. The capability of accessing the information from anywhere and any type of machine has drawn great acceptance for its development. This medium of expression has also allowed us to pursue collaborations with other groups on campus that routinely utilize similar information but remained isolated due to differences in software and hardware platforms. We can now explore the pooling of common resources for our students, particularly in the realms of imagery and mapping where Architecture, Landscape Architecture, Fine Arts, Geography, Environmental Studies, and others overlap. We are also making digital web-based documents the common product of studies undertaken by our students, building a long- term collection for future reference rather than reports which are never seen again after a term ends.
At the global scale of the Virtual Library, it is possible to start exploring what users are interested in on the web. At the beginning of December 1994, I put a metering program on all of the references in the subject catalogue to identify what people are accessing. While this does not indicate why they are accessing the links, it does start to indicate the general tendencies of users, not whether they might be 'surfing' or specifically 'searching', but what reference types they are interested in. What has become obvious over the last two months of reporting is that image-based resources (digital image databases, multimedia descriptions of architectural projects or sites) are a heavy access point, occupying 50-60% of the library use.
As these are the primary links that students may get educational value from (but certainly not the only ones) they are more significant to this inquiry. From my own studios and classes it seems clear that there are two modes of exploration that the students are engaged in, search-oriented and exploratory. As specific work is assigned, students often try to find relevant information using the various lists or search keywords as the server may provide. The fact is that this rarely results in success at this time - finding something you are specifically looking for is not yet a likely prospect in Architecture or Landscape Architecture both due to the scope and language used in searching.
This is evident from my own students and the number of searches that do not result in direct or indirect (same topic but with different keyword encoding) matches. Many of these searches often switch to exploratory use if they did not start from there, selecting interesting and usually divergent topic areas. In most cases the web is useful for expanding ones thoughts, building a wider and more diverse experience in ones education (or at least the potential thereof). In design this is useful and oft-used tactic, to question the problem scope itself and see if other interests need to be considered; e.g. should we design to accommodate the road or should we be exploring solutions to the automobile and where might that lead us. Text and imagery can trigger associative connotations, leading to a wide range of subject explorations, and while most Architectural libraries are self-contained and students would have to go and search in other physical locations, the web connects divergent thoughts and links in a manner that allows for free exploration, interactive and intuitive thought where network-bandwidth provides.
While there is no definitive evidence that supports this as a "better" educational model, it is also not obvious that it is any worse - only different. This has generally been the approach to design studio education - one absorbs a multitude of ideas and develops their own learning/designing strategies - it has not been held as a common precept in formal lectured education. If the material from such courses is intended to directly influence design, than should it not follow similar models - or at least allow for them? The dichotomy of lecture and studio is an interesting one to evaluate against teaching and learning. Multimedia is starting to make the either/or approach a thing of the past - both strategies to education can be accommodated and explored for their meaning in educational processes.
As the web expands, the scope of material for directed searches will become more useful, assuming that the search language can be clearly expressed to the users. Further, the range of multimedia material that will be accessible will eclipse what conventional publishing can provide, allowing for interactive exploration of computer constructions or reconstructed works that are three or four-dimensional in nature. We are already finding this to be useful in explaining student designs to critics that are participating from elsewhere in the world, and it has shown itself useful and publicly accessible in a context such as the Montreal project.
For basic web access, one needs internet access (direct or through a protocol such as SLIP or PPP), and a WWW Browser such as NCSA Mosaic, Netscape Navigator, MacWeb, and many others. For those with only textual access, a browser known as Lynx is commonly available. To access interactive three- dimensional models, the Centre for Landscape Research supplies a modified version of Mosaic known as CLRMosaic (for Silicon Graphics Workstations, at ftp.clr.toronto.edu in /pub/sgi/clrmosaic).
The CLRnet World-Wide Web server, which serves the local research and teaching needs, several Architectural and Landscape Architecture organizations, and the Virtual Library in these fields, is currently a Silicon Graphics, Inc. Indy workstation with 32MB of RAM, and 1GB of storage space. Currently the server is also used for student work and statistics from others on the network indicate a potential of ~50,000 server accesses/day for this level of server (before being overloaded). Our current usage is in the range of 12-15,000 accesses/day. The statistics break out into approximately 50/50 usage between internal access (local machines) and external access (all other machines accessing information on this server) at this time. The heavy internal use is accounted for by several of our courses which require students to publish their work on the World-Wide Web. To date (since May 1994), the server has delivered over 1,000,000 files, one half coming between May and December 1994, and the other half in only January and the beginning of February 1995. It is beyond us to accurately predict the future usage, and most graphs of World-Wide Web usage indicate a continuing exponential growth (since 1993).
Some of the more significant World-Wide Web documents that are mentioned in this article are accessible through the following network addresses:
Document URL (Universal Resource Locator) Virtual Library Subject Listing http://www.w3o.org/hypertext/DataSources/bySubject/Overview.html Architecture Virtual Library http://www.clr.toronto.edu/VIRTUALLIB/arch.html Landscape Arch. Virtual Library http://www.clr.toronto.edu/VIRTUALLIB/larch.html CLR Home Page http://www.clr.toronto.edu/clr.html
Barrett, Edward. 1992. "Sociomedia: Multimedia, Hypermedia, and the Social Construction of Knowledge." (Ed. Edward Barrett) MIT Press (Cambridge: MA).
Danahy, John. 1990. Irises in a Landscape: An Experiment in Dynamic Interaction and Teaching Design Studio, in "The Electronic Design Studio." (Eds. McCullough, Mitchell, and Purcell) MIT Press (Cambridge: MA).
Danahy, John. 1991. The Computer-Aided Studio Critic: Gaining Control of What We Look At, in "CAAD Futures '91." (Ed. Gerhard N. Schmitt) Vieweg, 1992.
Hammond, Nick. 1993. Learning with Hypertext: Problems, Principles and Prospects, in "Hypertext: a psychological perspective." (Eds. C. McKnight, A. Dillon and J. Richardson) Ellis Horwood (West Sussex: UK).
Hoinkes, R., and Mitchell, R. 1994. Playing with Time: Continuous Temporal Mapping Strategies for Interactive Environments, in "Proceedings 6th Canadian GIS Conference". (URL: http://www.clr.toronto.edu/COLLAB/CCA/ContTemp/gispaper.rtf.binary.html)
Gray, Matthew. 1994. Growth of the World Wide Web. WWW (URL: http://www.netgen.com/info/growth.html)
Papert, Seymour. 1993. "The children's machine: rethinking school in the age of the computer." BasicBooks (New York: NY).
Rodney Hoinkes is an Adjunct Lecturer in Computation, Architecture, and Landscape Architecture at the University of Toronto and the Head of Design Applications within the Centre for Landscape Research. He is currently completing his Doctorate in Design at the Harvard University Graduate School of Design, exploring methods of visually representing design processes to better facilitate network collaboration efforts. Rodney is also the maintainer of the Virtual Library subject areas of Architecture and Landscape Architecture on the World-Wide Web. He can be reached by email to: rodney@clr.toronto.edu.