Articles

Urban Design on the Internet:

RUDI, a case study in practice

A Paper By

Alan Reeve, Rowena Rouse, Catherine Tranmer, Bill Worthington.

RUDI is jointly run by the University of Hertfordshire and Oxford Brookes University

http://rudi.herts.ac.uk/

Introduction

Urban design would seem like an appropriate testing ground for developing an Internet based resource , constituted as it is quite naturally of a complex range of materials and media, from theory to practice, from simple text to 3D representations. This was, indeed, an initial rationale behind the funding and establishment of RUDI (Resource for Urban Design Information see Fig. 1). It would answer, that is, to an anticipated demand for diverse data, information, polemic and instances of good practice, and be an experimental site providing sufficient (hyper)textual depth to challenge the most innovative theoretical and technical structures. After all, architects in education and practice have for some time been exploring the new medium and its glossy possibilities.

But the RUDI team wanted to provide something with more coherence, depth and, untrendy as it might seem, seriousness than most sites appeared to be offering at the inception of the project a year ago. This was the big issue whose scale has become apparent only slowly as the project has developed. It was not just a matter of defining a hitherto healthily unbounded subject, or of constructing a sufficiently comprehensive, iterational and intelligent knowledge structure - even if the available search engines had been able to cope with all of this, which they were not. It was also a matter of the relative unfamiliarity of urban designers with the new information technology of the web, and the apparently baroque difficulties of acquiring, assembling, reinterpreting and publishing material of any worthwhile content in this brave new medium. The Web has been fantastically over hyped in the last few short years. The reality is that using it to deliver services of any quality comparable with those readily available through the older and seemingly archaic and dry delivery systems - like books and journals, housed in libraries with relatively easy to follow cataloguing systems - raises some fundamental questions about the transferability of information from the old technology to the new, as well as broader issues to do with the technical culture of users - particularly of design and academic professionals.

This short paper reviews RUDI from the inside, as it were, reflecting on the issues above, drawing out some tentative lessons and conclusions about the limitations and possibilities of the Internet as a medium for delivering real content for those interested in the design of the built environment.

Figure 1. The RUDI Home page

Background to RUDI

(see RUDI Welcome Page for full project details)

RUDI was started in January of 1996 under the eLib programme, funded by the JISC on behalf of the UK Higher Education Funding Councils, and the British Library Research and Innovation Centre, with hardware support from Sun Microsystems Ltd. It is jointly run by the Engineering Research and Development Centre (ERDC) at the University of Hertfordshire from where the project is managed and where the server is located, and the Library at Oxford Brookes University, where material is selected, acquired and edited. In the the first instance it has funding for three years, to the tune of around Ł110,000 per annum.. This funding supports a well configured server (with 37Gb of capacity), data acquisition costs, and approximately two and a half full time staff. After the three years of dedicated funding the project is intended to become self supporting through a number of mechanisms to be developed as part of the programme.

The joint application of the two institutions arose in the first instance because both have become acknowledged centres of excellence in their respective fields: the ERDC for its expertise in computing in relation to information delivery systems and software along with expertise in digitising information in the subject area, and Brookes for its research and teaching in Urban Design, Planning and Architecture and for its Library's collection of relevant material. Of course, a by-product of this cooperation is, hopefully, a greater understanding between the two cultures of information technology and design represented in the two universities.

Aims

The aims of the project from the outset were similarly multi layered. The principal justification at a one level was and remains to provide a significant hypermedia collection of material on the topic of urban design of use to academics, practitioners and others. As a part of this objective is the value added possibility that the resource is dynamic and capable of being edited 'on-line', and the further possibility of allowing direct authoring by contributors. After eighteen months work in acquiring, marking up and configuring material in the subject area, RUDI can claim to have shown itself to be an exemplary resource in terms of these objectives: already there is a wealth of material - from journal articles, case studies and bibliographies, to a catalogue of other planning, architecture and urban design sites, an interactive discussion area, and various UK Government produced material.

These explicit aims and justification for the resource, clearly, have implicit and more subtle objectives with somewhat more problematic outcomes. The first of these is the development of a content structure with an appropriate searching and cataloguing mechanism which can deal with different forms of document, arising from different types of author, containing different kinds of material and potentially overlapping in terms of users expectations, interests and heuristic readings. Put more simply, the nature of both subject and material type requires a highly sophisticated search system to cope with any endogenous and exogenous demands that might be placed on it. The question becomes, where to draw the line for the purposes of providing a reasonably usable search tool, Fig 2).

Figure 2. The current Search page

Limits to Searching

This then has been the first real challenge to the project, and anyone out there who has been attempting to create a properly searchable site will undoubtedly have had similar experiences. The tools are not yet available, although they are being developed. For example, RUDI currently uses the 'Netscape Verity' search engine which allows full text boolean searching. This is a powerful instrument, however herein lies its weakness. For example, a boolean search for Oxford and traffic retrieves all occurrences of these words within a document whether or not they have a syntactic relationship. This means that users have to plough through numerous documents to find those that are actually of relevance.

To overcome these problems, RUDI has been experimenting with metadata and its associated tools. Metadata is the internet equivalent of a traditional library catalogue record. A metadata record will be created for each discrete item of information on the resource. The metadata will include standardised subject keywords and standardised place names. A separate metadata search engine (in this case 'Netscape Catalogue Server') will be used to search this data. By indexing the resource in this way, it is hoped to overcome the inherent problems associated with natural language searching.

At the outset of RUDI, metadata tools were very primitive. This has therefore been an important and pioneering aspect of the project, in so far as the RUDI team have been able to develop focused searching methods such as keyword and location specific searching, and automatic searching from an evolving controlled vocabulary.

The implications of this aspect of the development of such a resource are potentially uncomfortable: and for any one considering creating a site of equivalent breadth and depth need to be thought about at the outset. Embedding keywords or some other type of metadata in documents retrospectively can be a massive task, and the superimposition of templates on material already configured is not without its technical problems. The key to successful structuring and searching is to get the search mechanism in place at the outset. This makes describing and structuring documents much more manageable and efficient compared with a post-hoc approach. Again, RUDI has gone a long way in exploring these issues and successfully answering their challenge.

What RUDI offers: the philosophical and practical issues for urban design and related subjects.

The problem of creating efficient and appropriate tools for searching is doubly difficult in a field such as urban design where there is, for example, no existing and generally accepted set of subject terms or even categories. This is hardly surprising given the political and ideological volatility and vulnerability of the subject, its susceptibility to changes in academic as well as practice fashion and fads. There is at least an apparent contradiction, that is, between the healthily anarchic and anti-positivist attitude in a subject like design, and the technologically strict, binary structures and systems of the new information technology which are, on the face of it, highly unreflexive. It is perhaps a philosophical as much as a technical question whether this apparent contradiction can or even ought to be overcome, or whether it might not actually be healthier to simply accept it as the fly in the ointment of the new age of information.

On a more mundane note, work on RUDI has also brought to light the real problems of material acquisition and translation. Aside from issues of taxonomy or classification, actually getting hold of material and rendering it into a style and format which can be used on the Internet raises a number of complex and challenging issues. In our experience RUDI is an unusual and rare Internet site in an important respect. Unlike the majority of other sites in planning (for example the exemplary PAIRC) or architecture (for example Archinet), we are more than a gateway to other gateways or smaller resources. One major aim of RUDI all along has been to disseminate and make available real material with real content. Of course there are other sites which in part do this (for instance the Congress of the New Urbanism , which has links to the work of a number of practices closely linked to the movement). However, such sites are either predicated on the values and outlook of a particular lobby, pressure group or other limited constituency with a proselytising rationale; or they are commercial or governmental sites, again each with their own particular and structuring raison d'etre. RUDI is not like this. As a basic resource for the whole field of design in the built environment addressed to any one with an interest in that field it has no special access to material published elsewhere, or to a ready supply of new documents generated for other purposes. It has to rely that is on what it can author itself (for example case studies such as Gloucester Green Fig. 3), or what it can persuade authors and other publishers to give it (for example, Caring for Our Towns and Cities), or, material which comes to hand which is copyright free or where copyright can be agreed with publishers (for example, the Urban Design Group Source Book and the Urban Design Quarterly Fig. 4).

Figure 3. Sample case study top page: Gloucester Green

Figure 4. Top page: Urban Design Quarterly

The consequence of this is that RUDI has had to be pioneering in its approach to copyright. The project has a standard publishing agreement which Hertfordshire University's lawyers have produced largely without precedent. This must be completed and signed before material can be published on the site. The unfortunate but unavoidable consequence is that instead of the whole process of authoring and publishing being more rapid and immediate (the so called and much vaunted space-time compression possibilities of the information age) it is in this respect more bureaucratic and often slower than for other publishing media.

Likewise with the issue of authoring and translating for the Web, as it might be called. There is nothing immediate let alone instantaneous about marking up, OCR'ing (Optical Character Recognition) or scanning. The constraints of the standard software and the unpredictability of what browser any one user is looking at or how it is configured, means that often only an approximation to the original can be given.

Largely because of the built in and uncertain user limits - for example is the surfer running a 286 or does she/he have an mmx pentium processor inside with a browser that can support tables, VRML, etc. - the RUDI team often have to make necessary compromises and go for an assumed common denominator. This can be highly frustrating when creativity runs ahead of technical capacity.

Technical Competence and the Culture of Use

More seriously, the tests we have conducted as part of the project into users' perceptions and actual use are illuminating in terms of the technical competence that is out there. It is clear that while some academics and even practitioners are competent and confident in using the new information technologies, many are only tentatively acquainted with the intricacies of browsing and navigation. When designing for a screen of, optimistically, 14 inches, there is little room to give blow by blow instructions. Designers and academics can, on the whole, be assumed to have some imagination and the capacity to interpret and experiment in order to navigate, but there is clearly a need for research into how best to optimize layouts, the design of navigation buttons etc, as well as to develop the interface technology and style so that moving about a complex and multi-layered site becomes a thing of excited anticipation rather than an experience suffused with the fear of getting lost in some Gothic labyrinth.

The other key finding from our research into user's attitudes is that those with no Net authoring experience are often unrealistic in their expectations. Again there is a paradox. The Web has been (self) promoted as individualistic and potentially anarchic, capable of fitting the needs and desires of individual users at the point of design and production - at the authoring stage. It can also be highly reflexive if the means of production are the right ones: feedback forms, and a team of individuals able and resourced to respond to such demands. But, at the point of consumption, despite the hype, the user is largely stuck with what they are given: and then only in the form their machine can be set up to receive - unless she or he has access to a great deal of memory and storage to save and remanipulate material.

Of course the technology is in its infancy, and needs to be nurtured and disciplined. The screens we look at now will probably seem, in ten or twenty years time, like the flickering and dim shadows Logie Baird first witnessed at the advent of the television age, as compared with the sharp and in your face images of the emergent wide and flat screen technology.

Conclusion

RUDI has shown us that we have to work with the technology that is available, accessible and already familiar: designing such a site requires both an acceptance of the knowledge and skills lag - practitioners and academics are likely for the most part always to be a step behind the technophiles - and a willingness to experiment and take risky steps with as yet untested tools. There is a danger that the effort and money required to produce real content will prove so unsupportable that the Web as used by design professionals and academics will end up looking like a serpent chasing its own tail. However, RUDI has demonstrated that this does not have to be the case. It has shown that it is possible to build, maintain and develop a site with real content of use and interest to a wide range of individuals - from academics to practitioners. This is evident in the fact that its corpus of material is growing week by week, largely because of the enthusiasm of its constituency. We are keenly aware that there is a massive demand for the sort of resources RUDI offers. It is this above all that encourages us to overcome the challenges outlined in this paper and that , in our view, means that we shall continue to pioneer delivery of information through this new and exciting medium.