The digital world has become increasingly involved in collections, archives and libraries: first with the digitisation of huge amounts of analogue material, then more recently with the arrival of born-digital material. Archives of architectural firms which began using computers to create designs in the 1990s, present new challenges for Het Nieuwe Instituut’s State Archive for Dutch Architecture and Urban Planning. Software eventually goes out of date, so how do you keep digital documents accessible? What should be done with the increasing volume of data that accompanies this digitisation of architectural practice? Has this created new research issues? These and other questions were discussed at a meeting of experts on 20 March 2018 to mark the visit of Tim Walsh and Stefana Breitwieser of the Canadian Centre for Architecture to Het Nieuwe Instituut.
Het Nieuwe Instituut invited CCA’s Walsh and Breitwieser to an exchange of knowledge and experience about archiving born-digital architecture. Among the subjects discussed at a meeting of experts with representatives of Dutch archives such as EYE Film Museum, International Institute of Social History, Royal Library and National Archive, and firms like MVRDV and OMA, was how to open and view poorly accessible digital documents of the 1990s; how to preserve this material, and how to save and categorise it to ensure it remains accessible for future reference.
Behrang Mousavi, Heritage manager at Het Nieuwe Instituut, opened the afternoon session by introducing the State Archive for Dutch Architecture and Urban Planning and discussing the advent of born-digital material. Architectural firms began to use computers to create designs in the 1990s. Het Nieuwe Instituut first confronted born-digital material when it received Carel Weeber’s archive in 2009. In 2015, came the MVRDV archive, a firm which had started to experiment with computers relatively early, and indeed profusely. In 2015, the State Archive began looking into ways to preserve digital material and make it accessible.
Marcel Ras, Digital Archive Cataloguing project leader at Het Nieuwe Instituut, then presented an extensive list of subjects programmed in blocks for discussion that afternoon including collection policy, implementation, users and the importance of cooperation. Digitisation of design practice has resulted in increasing complexity and a growing quantity of data to be catalogued and preserved. What criteria are used to select material and how best to ensure that this material remains accessible in the future? Archives have to keep informed about current design practice to be able to anticipate future developments.
At present, Het Nieuwe Instituut is examining these issues in relation to the archive of international architectural firm MVRDV. This constantly expanding archive is a good test case. It is big: it encompasses 400 projects dating from 1993 to 2015, comprising 5 TB of data. It incorporates major developments in the digitisation of architectural practice: MVRDV adopted early, the firm experimented eagerly with all kinds of possibilities that the new technology offered and even developed its own software. So this is not just about archiving digital designs and 3D presentations; this is also about websites, games, software and extensive e-mail exchanges regarding projects.
Curator Suzanne Mulder of Het Nieuwe Instituut talked about the State Archive’s collection policy. At its core is the need to achieve insights into the design process. In addition, collectors look at the themes which were prominent in a given period, such as sustainability or social development. Het Nieuwe Instituut’s interest in digital culture as well as architecture and design also plays a role in the approach to digital archives. Choices are inevitable when forming an archive. Usually this comes down to selecting from a series of representative projects. Projects with a similar design processes can be discarded.
MVRDV’s archive contains material in many different mediums: different software programmes were used in the design processes, while digital material is closely connected with analogue material. Paper sketches are converted into digital designs; photos of models are included in 3D presentations which result in physical models: everything is mixed. To understand and to preserve the connections, Het Nieuwe Instituut staffers interview architects and staff at MVRDV who worked on specific projects. MVRDV’s archive is still in use by the firm. Although for quite a different purpose: for MVRDV the database is a source of solutions to design problems which they regularly use when working on new projects.
CCA digital archive
CCA first dealt with an archive containing digital material in 2004, Tim Walsh explained. In 2015, CCA launched a structural programme to build a digital archive. Besides architectural archives, CCA’s collection includes archives of architectural historians and artists, for example. This archive material is also diverse and hybrid: both digital and analogue. The Canadian architectural centre selected 25 key projects with which to examine issues involved when making digital documents accessible and when preserving these. As a privately financed institution CCA has raised funds to appoint three fellows for this research project.
Research into digital archives reveals a highly practical history of digital practice, according to Walsh. When architects began using computers in the 1980s, there was hardly any standard software available so the first users designed their own programmes. This is the most challenging material: it is difficult to open and view. Other software may be inaccessible because new updates no longer support older versions. This makes it essential to collect, preserve and archive software. Software companies can also play a role here, which in Walsh’s experience they are often willing to do.
Another question is who is going to use these digital archives and what these users want. It is difficult to predict. Very few users actually access the archives at present: CCA’s archive currently attracts around 800 users annually, of whom 5 or 10 are interested in the digital archive. These vary from researchers to designers, and not always from the architectural world either, but for example the film industry or science history. Walsh emphasised a couple of times that the archive’s digital material is quite heterogeneous; besides architectural designs, it includes websites, games, art projects and so forth.
Walsh is also concerned with the subject of copyright. That digitisation overcomes the physical limitations of traditional archives, and that they are sustainably preserved and accessible does not mean that everybody should have access. Architects and artists, not to mention software producers, all wish to minimise the risk of information being copied and used for other purposes than research. Users therefore have to come to CCA in person to consult the digital archive, where all kinds of measures are in place to prevent people copying documents.
Het Nieuwe Instituut digital archive
Ania Molenda, researcher at Het Nieuwe Instituut, is currently researching digital practice at six architectural firms. She chose these because they all have a relatively long history (15 to 20 years) and vary in size and in the degree of experimental practice. One of her aims is to examine the problems which occur when preserving and making older digital documents accessible. Firms’ archives grow organically; the structure is adapted as they grow, so that essential links between design documents may be severed and information lost. This may be retrieved by consulting staffers who worked or helped on projects. Although designers regularly change jobs, which can make finding them complicated.
Molenda also looks at where automatic search methods and archiving machines can be used to accelerate and facilitate the process. Other questions include the extent to which archives should be sanitised: what do you keep and what do you delete, and do you maintain the existing categorisation? Categorising based on software – which is MVRDV’s system for example – is not a logical method; the State Archive focuses more on design processes. But what does changing the categories do to an archive, is there a risk of losing information? These are questions which also impact on the future use of these archives.
Cooperation with design firms
Another major issue for archives is how to ensure that design firms are aware of the need to preserve for the future. If architects were to take account of future archive requirements while they work this could save a lot of research hours at a later date. A suggestion from the audience proposed that architectural firms be asked to store their material in a standard format. EYE Film Museum already asks film makers to do this, although that is for the final product and the requirement is linked to the last instalment of a subsidy. Other archives are also more likely to be dealing with uniform archive material, which makes it easier to implement a standard format.
This is different in architecture, as both CCA and Het Nieuwe Instituut agree. Many firms experiment in different mediums, digital and analogue, so that the project material is heterogeneous. Moreover, architects are under greater pressure from deadlines, while in larger firms many different people are working on often long-term projects. Where standard procedures have been established, they are not always followed. As an OMA architect present in the audience confirmed. It can be difficult enough to keep information and material well ordered during the design process. Once the product has been delivered, the focus shifts to producing publicity for the project using recent images and basic information. No attention is paid to the rest of the project archive after this.
Molenda of Het Nieuwe Instituut did however note that she felt that firms who have been working with software longer than most are more aware of the importance of archiving carefully. The OMA architect could also imagine that firms would be interested in receiving best practice guidelines from the State Archive for archiving completed projects. Whether it made sense to offer an archiving service was doubtful, since firms would not have budgeted to pay for this. It would be a good idea to contact educational institutions, another suggested, so that architects would be taught the importance of archiving during their training. Henriette Bier of TU Delft noted that only the final version of a student’s work is ever preserved, the rest is not archived. She also commented that students are used to customising the software they use. Architecture and standard solutions are never a good match.
Automation and software licences
As Stefana Breitwieser of CCA pointed out in her presentation of her research group’s archaeological approach to making out-of-date documents, software and stored material accessible, much of it can be opened. She emphasised that 80 percent can be accessed without much difficulty, however the remaining 20 percent often involves (new) problems which often take a long time to resolve. Each problem solved supplies valuable information for processing other documents. The more digital archives the research group processes, the better they are able to chart which problems occur and where and how best to resolve them. The more data, the more archives can be digitised.
Walsh noted that CCA uses automation software such as Archivematica. This programme is more a toolbox than a prescribed work-flow, enabling it to be used flexibly. For the present, CCA still has a subsidy to finance research, but soon their archive work will be taken over by museum staff. The software itself is also archived, partly to be able to open old documents and partly because researchers are interested in the development of software, the various ways the software is used and the products it produces. Using old software also involves licences. In general software companies are willing to help when it comes to programmes or older versions which are only for archive use. Separate requests are made for exhibitions. Walsh noted that some companies donate software to CCA. Research groups in the US are currently investigating user rights relating to outdated software. It would be useful if regulations were issued regarding this. It is permissible these days to decode old video games to preserve them. Yale University is working on a software library of education and heritage.
A questioner in the audience asked whether CCA had to assist digital archive users intensively. Walsh explained that digital files are stored in the system in which CCA’s physical collection is made accessible (photos, drawings, models, publications). He noted that users tend to ask for information about how the archive is managed and formed and what is protected and why. Users are able to work with the documents themselves. However, it is noticeable that users of analogue material are mainly interested in a specific project or oeuvre, while users of digital archives often do comparative research and are interested in big data – they find individual documents less interesting. Should this require a different method of archiving? According to Frans Neggers of Het Nieuwe Instituut it is important that an archive be as open and complete as possible, including both original material and information about how and when this material was used and employed in an architect’s design process, with information about how it was archived.
Many questions were not immediately answered that afternoon. Yet it was clear that archivists are working hard to chart the issues relating to the archiving of digital material and that with a bit of effort, many of the problems will be solved.
Report by Lotte Haagsma
This meeting of experts was organised in connection with two projects at Het Nieuwe Instituut’s Heritage department: Cataloguing and Implementation of Digital Archives and Making Choices 2.0, a further development of current collection policy. Tim Walsh and Stefana Breitwieser’s visit was part of the International Visitor Programme.