Hypermedia Authoring and Contextual Modeling in Architecture and Urban Design : Collaborative Reconstructing Historical Sheffield

Studies of historical architecture and urban contexts in preparation for contemporary design interventions are inherently rich in information, demanding versatile and efficient methods of documentation and retrieval. We report on a developing program to establish a hypermedia authoring approach to collaborative contextual modeling in architecture and urban design. The paper begins with a description of a large-scale urban history study project in which 95 students jointly built a physical model of the city center of Sheffield as it stood in 1900, at a scale of 1:500. Continuing work on the Sheffield urban study project, it appears to us desirable to adopt a digital approach to archiving the material and in making it both indexible and accessible via multiple routes. In our review of digital models of cities, some interesting yet unexplored issues were identified. Given the issues and tasks elicited, we investigated hypermedia authoring in HTML and VRML as a designer-centered modeling methodology. Conceptual clarity of the methodology was considered, intending that an individual or members of design groups with reasonable computing skills could learn to operate it quickly. The methodology shows that it is practicable to build a digital contextual databank by a group of architecture/urban designers rather than by specialized modeling teams. Contextual modeling with or without computers can be a research activity on its own. However, we intend to investigate further how hypermedia-based contextual models can be interrelated to design development and communication. We discuss three aspects that can be explored in a design education setting. The Sheffield 1900 Urban Study Project: A Case of Collaborative Contextual Modeling The urban study project “Sheffield 1900” was recently launched at the Sheffield University School of Architecture. The starting point for our urban study was a consensus among the staff that context is important, and that one should always know the history of a site in order to understand how it was formed and what it means, whatever use is eventually made of that information. Allied with this is the problem of what constitutes a city, how cities might work today, and how buildings might (or might not) be integrated within them. Again, the current situation is only comprehensible with some understanding of the past, with an idea of how a city developed and how its past states contrast with the way it is today. To understand an urban context, more is needed than a peremptory glance at some old maps. On one level this is a question of developing perception and understanding, of knowing what to look for and how to look, but it is also a question of learning how to use maps and archives, and of getting familiar with the kinds of things that can be found in them. Project Objectives. For the academic year 1998/99, the urban study was devoted to Sheffield. We considered it desirable to start some kind of database in which information could be cumulatively recorded, and which could be consulted by anyone planning a project in the city in future years. The urban study project is intended to continue taking a different aspect of the city each year. Pedagogically speaking, the study has four objectives: to demonstrate the importance of understanding how a place has evolved before contributing a new design; to show the changes that have taken place in cities during the last century; to teach students how to undertake historical research; and to build up a database about the history of Sheffield for future reference. Project Organization and Team-working. But how does one study a city systematically? We decided to try to base our work around a physical model of the whole central area as it was in 1900. This date finds Sheffield at its industrial height and the peak of its wealth, before the depredations of war, redevelopment, and the collapse of heavy industry. It is late enough for accurate maps and photography, yet early enough to show many original features missing today. Topically, it marks the passage of a century, so coincides pertinently with the millennium. We decided to divide the 95 students taking part into groups of four or five. A consequent division of the work was needed, so we imposed on the plan of the city a north/south and east/west grid. A one in four subdivision of the current Ordnance Survey grid produced 20 squares with a side-length of 200 meters on the ground. At a scale of 1:500 this gives a piece of model 40-cm square, and a whole model 2.0 meters by 1.6. The scale of 1:500 is just large enough to show the forms of individual buildings and even chimneys, but without in most cases including such details as door and window openings [Figure 1]. The gray model of contours alone was temporarily assembled at an intermediate stage in the model’s development to check for inconsistencies. It was also necessary for groups to collaborate where grid-lines passed through buildings, but the difficulty of accurate matching between one grid-square and the next was mitigated by leaving a gap of 10 mm. Sources of Research and Modeling. The initial sources of information for the model were the Ordnance Survey Maps of 1897 and 1903. The use of two editions immediately highlights the problem of a changing situation, for even if there had been a map published in 1900 it would have been out of date, having taken a year or more to prepare. Inevitably there were places that were cleared sites in 1900, and there were buildings under construction: one of our model squares actually shows one of them scaffolded. Groups with sites in transition soon found out about them, and were usually able to establish from records what was present in 1900. Things that had existed for a number of years before or after the date were usually recorded in some way, but more temporary things proved elusive. Some buildings present in 1900 are of course there still and could be photographed or surveyed in place, but the extent of survival varied greatly from one grid-square to the next. Some parts of the city had remained relatively stable, but others notably sites of industry, slum housing and post-war ring-road development had changed out of recognition. One group found only a single pub still intact on their site. In dealing with such areas, careful research was needed to establish the three-dimensional form of the fabric, and here the collections in the City Archive, Local Studies Centre and the Hawley Collections proved essential. The Ordnance Survey maps give accurate plot outlines and ground heights, but no information on the number of storeys or the roofforms, while courts and light-wells are sometimes also missing. Much valuable information could be gathered from period photographs, but the unexpected major source was fire insurance plans. These not only noted the number of storeys and locations of small internal courts, but even details such as roof lights. Figure 1. Modeling one of the grid squares (200m side-length) at 1:500 in the Sheffield Urban Study Project. Tasks of Data Creation. The physical model was the focus of the research, but the student groups were also asked to provide other kinds of information. They were asked to account for the presence, shape and orientation of every street within their square, and this meant looking at earlier maps to trace the growth pattern. Plans in the archive show not only every detail of the layout, but even the earlier field boundaries. With such information it is possible to explain the development of nearly everything in Sheffield since the mid-18th century. The student groups were also asked systematically to chart the uses of the buildings in 1900, and here contemporary directories were the major source. They allowed a consistent series of coded plans to be drawn showing uses across the city, and it was soon obvious that the different squares had very different characters. Illuminated by knowledge of all these uses, the model conveys the impression of a dense city teeming with activity. Spatially it was organized in a clear hierarchical manner, with major streets giving way to minor, these leading in turn to alleyways and backcourts. The pressure for maximal land-use conflicted with the requirement for daylighting to produce a complex and irregular built pattern of yards and light-wells, a use seemingly being given to every square foot. In addition to the work outlined above, each student within a group was asked to select a building from the grid-square in question and to examine it in detail. In some cases these could be unique buildings of declared architectural significance, such as the Cathedral and the Cutlers’ Hall, but students were equally given the option of looking at an ordinary building as example of a type. A detailed report of more specific findings from the Sheffield Urban Study project will soon appear in a separate publication (reference to be added later). The project was accompanied by a series of lectures and discussions, and one of the topics explored was the nature of building types, and whether they should be classified on a formal, functional or technological basis. We stressed that the language of building types is significant in giving a city’s coherence both in terms of the repetition that establishes a type and the contrast that differentiates it from another. A further task for the students was to identify the equivalent of their chosen building today, and to pinpoint its location in the modern city. With the later expansion and loosening of density occasioned by the motorized transport, many are located in relative isolation and more towards the periphery.