Collecting and Conserving Code: Challenges and Strategies
Melanie Swalwell and Denise de Vries
The collection and conservation of code is still in its infancy in Australia. Even where coded items do exist, they are almost completely invisible within local cultural institutions and archives. Born-digital heritage faces unique risks - the degradation of hardware and software, obsolete operating systems, and intellectual property laws that restrict digital preservation activities. Too often, governments and cultural institutions either fail to recognise the precarious situation of historic code-based media, or are not able to respond in an appropriate fashion, due to a lack of resources, know-how, or sometimes, will.
After outlining some of the challenges - for institutions and researchers - of developing collections of games and other software, this article will detail two current research initiatives. The Play It Again project is conducting research into the largely unknown histories of 1980s game development in Australia and New Zealand, ensuring that local titles are documented, preserved and make it into national collections. The Australasian Heritage Software Database seeks to: draw together existing knowledge about locally-developed software, marshal a network of supporters, and develop an enabling discourse that supports research into histories of software and digital preservation. Whilst these projects do not provide complete solutions by any means, a local discourse about the importance of collecting and conserving code is emerging.
In this article, we discuss the importance of software as a form of coded cultural heritage. We first outline the challenges that a researcher seeking to use such resources currently faces in Australia. We then turn to software as a form of digital heritage, considering some of the value that can be derived from studying the code itself, apart from its function. In the second section, we detail two current research initiatives we are undertaking, which are developing strategies for, and contributing to a local discourse about the importance of, collecting and conserving code, to begin to remedy the current invisibility of coded works. The Play It Again project is conducting research into the largely unknown histories of 1980s game development in Australia and New Zealand, ensuring that local titles make it into national collections and are documented and preserved, enabling the public to once again play these games from the 1980s. The Australasian Heritage Software Database seeks to draw together existing knowledge about all forms of locally-developed software in order to document it, to marshal a network of supporters, and to develop an enabling discourse that supports research into histories of software and digital preservation.
Software is largely invisible in Australia’s cultural collections. From our perspective as researchers of local software history, the lack of a framework to capture software (that is, to collect software and to make it discoverable) is remarkable, particularly given the amount of unique local content our research is discovering. A handful of states and territories include digital items in their legal deposit legislation - Tasmania, South Australia, and Northern Territory (see Coates et al. 2009) - but currently there is no national framework. Whilst there are some titles in some collections, it seems that currently no one has a good idea of what exists, where it is, and what is missing. Consider a hypothetical researcher writing a PhD on educational software since its introduction in the 1980s. Their project requires them to consult the educational software titles that were used in Australian classrooms; how would they go about locating these? Let us assume they try some topic and keyword searches. There are no appropriate ‘format’ categories to search within on either their home institution’s library catalogue, nor on Trove (a union catalogue), the National Library of Australia’s “national discovery service”. Software is not a ‘dataset’, nor is it an ‘object’, a ‘book’, a ‘journal or magazine article’, a ‘video’ or a ‘map’. Searches for ‘software’ and ‘disc’ bring up very little in the way of useful references, and ‘education AND software’ brings up only scant references, usually to books.
Whilst our researcher has not been able to find any catalogue entries to lead them to software in collections, they have, however, lucked upon the name of a local educational software title, via a list on a collector’s website: Raft-Away River, published by Jacaranda Software in Milton, Queensland, in 1984. With this information, they manage to track down some of the authors - people who worked for Jacaranda Wiley writing educational software - and, having interviewed them, they have now compiled a list of known software titles which they have been able to search for. Despite being catalogued as a book, according to Trove, Raft-Away River is available in the libraries of the Australian Catholic University, the University of Melbourne, and Hurstville City Council Library Museum Gallery. Of the other 24 known titles Jacaranda produced, our researcher was able to find all except for five in the catalogues of Australian libraries. This apparent ‘success’ is hardly cause for celebration. That one has to know the title of what one is searching for before it can be found is a significant problem. Research is by definition about creating new knowledge - finding out that which is not known at the outset. The problem is not solely with the local ‘union’ catalogue, Trove. Searching for Raft-Away River on WorldCat similarly returns an item whose format is ‘book’. Searching WorldCat for the title of another item our researcher has serendipitously discovered, Funky Punky: A musical punctuation game, yields 38 records (the game is the 12th record). Trove yieldstwo ‘books’. The resource is only found on both catalogues if the whole title is entered. Searching, for instance, for ‘Australia funky punky punctuation’ returns no records. At present, it is not possible to find what you don’t already know about.
Whilst the above example of a search for 1980s educational titles reveals that there are some historic software titles held in public libraries, our judgement is that this is often not the result of purposeful collecting, but incidental acquisitions over time. The collection of code - and discussion of its importance - is in its infancy in Australia. Collections policies sometimes indicate that code is collected (National Library of Australia 2008; State Library of South Australia 2006). If code is being collected, it is not transparent, and works are not readily discoverable. Whilst policies and practices are in place for collecting the traces of some aspects of digital society - the archiving of the internet was, for instance, something that was embarked upon quite early, with the Pandora project - and strategy documents exist (Anon n.d.), in our view, Australia is not collecting or adequately preserving the born-digital products of our highly-computerised society. The previous Labor Government, in May 2013, signalled that a national digital legal deposit scheme and a digital audio visual archive are on the horizon (Australian Government 2013; Anon 2013). These are long overdue provisions. It is, however, unclear whether and when they will be implemented by the new Coalition Government, elected in late 2013, and whether they will extend to software.
Those software titles that are held are not only difficult to find due to anomalies in their cataloguing, they are also difficult to access. Even if a researcher finds a catalogue entry, it is highly unlikely that the actual resource will be available online, due to the contents of the discs not having the compatible hardware, operating system, support platform software or the provider not having the necessary rights clearances. Finding software resources therefore raises access issues. Access, in turn, raises preservation issues.
Institutions with born-digital holdings are currently not set up to allow them to be accessed. There are no vintage computers within Australian cultural institutions on which out-dated software might be accessed. Accessing born-digital material is intimately bound up with preservation issues; in this, there are parallels to accessing other fragile cultural material for research, such as works on paper, or textiles. Where coded works differ is that in many cases, there will not have been any migration of software or datasets to current systems, or disc imaging projects. For example, consultations with the State Library of South Australia reveal that the almost 30 year old Funky Punky title (1984) still sits on its 5.25” disk and has received only conservative preservation management (housing in a climate controlled environment). Common sense - as well as conservation guidelines - dictates that no one would attempt to open what might be the only copy of a title that is held in the country (Erway 2012). Therefore, at the moment, if access to a rare disc is required by a researcher, the content first needs to be copied. The Australian Copyright Act allows “key cultural institutions” to make up to three copies of material of historical or cultural significance for the purposes of preserving the material against loss or deterioration (Commonwealth of Australia 1968: s51B). Making a copy sounds a simple enough operation but given this material is old, it brings up a range of challenges to do with obsolete hardware, old media carriers and their deterioration such as bit rot, and disc and file formats. Only once the staff of an institution have copied (imaged) the software would the researcher be allowed to try and access it. Whilst the ways in which institutions might offer researchers access to heritage software in their collections once it has been preserved (e.g. on emulators) needs to be considered (Lowood 2013), a thorough consideration of this is outside the scope of this paper.
Some have questioned whether cultural institutions ought to be responsible for collecting and preserving software, given that much of it is proprietary. We argue that coded works including software belong within the collections of cultural institutions and that their mandates and collections policies need to include these. The proprietary argument is not particularly persuasive: books are also published and sold by commercial entities, yet they find a place in national and state collections. Furthermore, software and other coded works constitute a form of heritage - digital cultural heritage - and it is proper that those institutions that are charged with collecting and preserving the nation’s heritage include software and other coded works within their holdings.
UNESCO recognised digital heritage in its ‘Charter On the Preservation of Digital Heritage’, adopted in 2003. The Charter specifically mentions software as one of the range of digital materials:
The digital heritage consists of unique resources of human knowledge and expression. It embraces cultural, educational, scientific and administrative resources, as well as technical, legal, medical and other kinds of information created digitally, or converted into digital form from existing analogue resources. Where resources are ‘born digital’, there is no other format but the digital object.
Digital materials…are frequently ephemeral, and require purposeful production, maintenance and management to be retained.
Many of these resources have lasting value and significance, and therefore constitute a heritage that should be protected and preserved for current and future generations. This ever-growing heritage may exist in any language, in any part of the world, and in any area of human knowledge or expression. (UNESCO 2003)
The ‘UNESCO Guidelines for the Preservation of Digital Heritage’, authored by the National Library of Australia, were also published in 2003 (National Library of Australia 2003; Webb 2003). Since then, the Memory of the World program has held two international conferences (Moscow in 2011 and Vancouver in 2012), which have produced Declarations affirming the importance of timely action. Amongst other things, the Moscow Declaration urged national governments to “support research in the various aspects of the preservation of digital information” (UNESCO 2011). Whilst it has a similarly broad scope, the Vancouver Declaration urges:
professional organizations in the cultural heritage sector to:
a. cooperate with other professional associations, international and regional organizations and commercial enterprises to ensure that significant born-digital materials are preserved.
e. identify and evaluate the specific threats to which their digital information is vulnerable, and implement appropriate processes and policies to mitigate these threats. (UNESCO 2012)
Despite this recognition, change has been slow to come. As the Memory of the World Programme states, “Despite the adoption of the ‘UNESCO Charter on the Preservation of Digital Heritage’ in 2003, there is still insufficient awareness of the risks of loss of digital heritage, even though knowledge is today primarily created and accessed through digital media” (UNESCO n.d.).
Recognition of software as a form of digital heritage is still quite a new idea for some people, so it is helpful to think about the heritage significance of software alongside other forms of heritage. ‘The Burra Charter’ is used in assessing the heritage significance of place in Australia. It recognises that places have historical, aesthetic, scientific and social significance. Many of its statements can be extrapolated quite sensibly to software and other coded works to articulate their significance. A reworking of this Charter might read:
Items of digital cultural significance enrich people’s lives, often providing a deep and inspirational sense of connection to self, others and community, to the past and to lived experience. They are historical records that are important as tangible expressions of Australian identity and experience. Digital heritage items reflect the diversity of our communities, telling us about who we are and the past that has formed us and the Australian society and cultures. They are irreplaceable and precious. These digital heritage items must be conserved for present and future generations (reworked from Australian ICOMOS 1979: 1).
Early digital games - which we talk about later - are an example of coded works with such heritage value.
Despite agreement that preservation action needs to happen forthwith, one probable reason for the trickle of software entering some cultural institutions is that collecting it entails some responsibility to preserve it, and this is perceived as being difficult. There seems to be little point to collecting software if it cannot be run. It is not that the source code is hard to preserve, as it is usually in text files. Executables are more difficult because many software products are hardware- and operating system-dependent. To date, digitisation initiatives have taken precedence over the born-digital in cultural heritage institutions. These are seen as “low hanging fruit” (National Library of Australia 2003: 48). The threats to the longevity of the born-digital are, however, arguably greater than those facing material artefacts and, as the UNESCO Charter notes, “there is no other format but the digital object”.
The heritage importance of software extends beyond its function. There is heritage value within the code itself. Programming is a creative process. There is an art to programming (Knuth 1974; 1998). Studying software code promises insights into different styles of programming, as well as into the history and development of programming. To facilitate such study, the code itself - source code - needs to be available for inspection.
The authors of much of the early software applications for personal computers were typically hobbyists and were often self-taught for the platform on which they developed. These pioneers of personal computing carried out much research on the capabilities and possibilities of the machine and developed new structures and paradigms. Through magazines, user groups and early bulletin boards they communicated with each other and exchanged ideas and experiences, which led to programs being written that extended known use and prompted new features in new models of hardware.
It is difficult today to understand the need to use memory efficiently. Having only 16K of RAM available to work within made the programmers of yesteryear very inventive when it came to writing optimised data handling in their programs. Both complex logic and graphics need RAM. Pioneers of the 1980s games and demo scenes created methods to fit as much as possible into the available resources and from these standards, code libraries and new instruction sets were developed. Complex data structures for personal computers had not yet been developed. Tips and techniques were written up in magazines especially targeted at owners of particular computers. It is possible that if one were able to arrange these code examples in chronological order we would be able to see the evolution of new paradigms and approaches to data modelling. Only by viewing the source code in its original format can we truly understand the thought processes that went into its generation.
There are also local styles of programming. Stephen Straley, author of many programming books in the early 1990s, remarked at Technicon 93 on the Gold Coast that Australian males had their own style when programming: they wanted to fit as much logic as possible on a single line of code. Thus, the programs written locally had an Australasian style, reflected the local cultures and filled local needs. This approach to coding is only visible when one actually sees the original format of the code. Decompiling and re-engineering is not capable of reproducing a facsimile of the original document, only an approximation of the logic.
To argue that software belongs in cultural institutions is not always a popular argument. Some ask whether cultural institutions are up to the task, given their slow start. We note the success of fan preservationists but cannot endorse the idea of leaving software preservation entirely to the DIY community, given that it is not resourced and draws almost entirely on private (volunteer) labour to do this work (Stuckey & Swalwell n.d.). Others are also doing important work in the area of software preservation, such as the Internet Archive, which is supported by philanthropy. Whilst some cultural institutions would probably be happy to be relieved of the challenges of collecting and preserving born-digital coded works, these institutions persist and enjoy a cultural influence and legitimacy which is the reason they are tasked with keeping nations’ cultural heritage. They need to be involved in keeping coded works, just as they keep physical objects. We therefore advocate a collaborative model which draws upon the knowledge, expertise and goodwill that exists amongst the range of stakeholders in this field. There is considerable potential for sharing and collaboration between traditional cultural heritage and information professionals in Galleries, Library, Archives and Museums (the GLAM sector), commercial enterprise (such as software companies), the hobbyist preservation community, non-state actors (such as the Internet Archive), and research academics. We believe the path forward must be collaborative, networked, and involve a distributed effort.
Collection and conservation of code in Australia
One of the implications of scarce or sporadic holdings of software is that local histories of software creation are not well known. In the domain of media arts education, for instance, Lisa Gye (2011) notes that whether local early media artwork is still held, accessible and will play on contemporary computers is a critical issue. The Media Arts Scoping Study similarly noted the value of media arts history in educational settings (NOMAD 2010: 25). Existing scholarship documents some of the work that Australian artists were doing in the early years of computing. Stephen Jones has written on the period ending with the 1975 Canberra exhibition ‘Computers and Electronics in the Arts’ (Jones 2003; Jones 2011) and others have made notable contributions to a local 1990s media arts history (Tofts 2005; Tofts et al. 2002; Zurbrugg 1994). Currently, the 1980s - the period when micro-computers first appear - represents a major gap in knowledge. Whether computer-based media artworks from the 1980s, for instance - or the 1990s for that matter - survive and can still be accessed, is currently unknown.
Whilst Australians have undoubtedly contributed to digital preservation discourses in their respective professional settings (in librarianship, for instance, see Harvey 2005; 2010; 2011) there is little in the way of a public discourse about preserving coded works in Australia, at present. In France, by contrast, stakeholders held a conference in 2012 asking whether a new Museum of Computing and Digital Society was warranted (Anon 2012). In the US, the Library of Congress runs a prominent blog, The Signal, dedicated to digital preservation (Library of Congress n.d.), whilst the UK’s Digital Preservation Coalition “acts as an advocate and catalyst for digital preservation” amongst its member organisations (Digital Preservation Coalition n.d.). These informal channels are valuable in presenting digital preservation to interested publics; a lack of public discussions limits understanding about digital preservation to specialist professional groups.
There is an urgent need to get the collection and preservation of software and other coded artefacts onto public, political and research agendas, in order to get the requisite attitudinal, policy and funding changes to ensure an enduring record of the born-digital. In the next section, we introduce two of our collaborative research projects, which aim to accelerate the collection and conservation of coded software artefacts in the region.
Play It Again
Play It Again is a digital games history and preservation project focused on 1980s digital games in Australia and New Zealand. It is a collaboration with the Australian Centre for the Moving Image, the New Zealand Film Archive, the Berlin Computerspiele Museum, and colleagues from several universities in Australia and New Zealand.
Digital games make up a significant but little-known chapter in the history of the moving image in Australia and New Zealand. Beginning in the early 1980s, the Australian software industry developed a remarkable record of content creation. Beam Software was Australia’s first games development studio. From their first electronic game in 1982, Beam went on to create over 100 games titles. A number contain distinctive Australian content, while others went on to international success, including such seminal titles as The Hobbit (1982) and the record-selling title in Europe, The Way of the Exploding Fist (1985). Early video and micro-computer games were extremely popular, and Australian software was popular with Australian players: in November 1985, four titles from Australian Commodore Review’s national top 10 were developed by Beam. New Zealand also had a booming local digital games industry in the 1980s, across arcade, console and home computers, with a vibrant homebrew culture (Swalwell & Loyer 2006; Swalwell 2008; 2010; 2005). Despite this, surprisingly little is known about the history of these local digital games industries, the antecedents to an industry which earns substantial sums today; $1.1 billion in 2012 in the case of the Australian industry (IGEA 2013). To date, digital games have also not enjoyed the care accorded other historic screen-based media by national institutions, such as the National Film and Sound Archive. The turbulence of the games industry - where many companies are short lived and firmly future-oriented - partly accounts for why it has not undertaken archiving activities.
The project addresses both the history and preservation of locally-developed titles. We have worked with the wider retro games and computing communities to develop lists of locally-developed game titles, documenting 700+ micro-computer games written in 1980s Australia, and 200+ in New Zealand. The process has affirmed the wisdom of crowdsourcing - that no one knows it all, but everyone knows something - as even the most knowledgeable collectors tend to have very specialised knowledge, which needs to be aggregated.
Recognising that an institutional collection and preservation solution is urgently needed, the project will get 1980s Australian and New Zealand digital games software into the collections of our local partner cultural institutions, together with other items of historic significance related to the production and reception of these titles. The project is effectively serving as a trial to see what is involved for cultural institutions in collecting digital games and what digital games collections might look like. Once target titles have been identified and sourced, they will be preserved in-house. Separately, the team is creating multi-platform versions of selected titles so that these will play in a browser, allowing early games to be legally accessed and played by those without in-depth knowledge of emulators.
The project is also interested in the historical consumption of games during the 1980s. We are keen to collect ephemera from the time as well as players’ memories. People often have strong and vivid recollections of their gaming experiences during this era, as playing early games was one of the great familiarising experiences to have with computers. To capture such material, we have built a Popular Memory Archive, where people can share their memories and items related to 1980s games, in a range of media forms (de Vries et al. 2013; Stuckey et al. 2013).
Australasian Heritage Software Database
The era- and games-specific Play It Again project presents in microcosm the challenges around documenting, collecting and preserving software more generally. As we’ve detailed above, software is not often collected or preserved in Australian cultural institutions and the national climate for recognising, discussing, and caring for born-digital heritage is not well developed. Moreover, documentation is extremely poor, with no centralised records of locally developed software. A significant sector of cultural endeavour and production is thus at risk of not only being lost - in that little software preservation work is currently being done to arrest deterioration and issues such as format obsolescence - but also of being forgotten, given that there is no documentation that it even existed.
For these reasons, we built the Australasian Heritage Software Database (AHSD) (Swalwell & de Vries 2011). In launching the AHSD in 2011, we sought to put the collecting of computer-based media, research into early computing, and preservation of the born-digital firmly on the national agenda. The AHSD particularly responds to the media historical knowledge gap around early software. Put simply, no one knows what software has been locally developed, as software historical surveys are usually U.S. or European in focus (see, for example Campbell-Kelly 2003), and usually only profile the ‘winners’, those companies which made software titles that had particular impact. There is a lot more to software history than this, but these other histories cannot be told if they are not known.
The AHSD aims to document locally developed software, as there is a lack of assembled or aggregated knowledge about this. To our knowledge, no one has previously surveyed software production in the region. Knowledge about local software does exist, but it is not in cultural institutions; more often than not, it is in the community. Public expertise can contribute significantly to archival projects, as the efforts of the Computer Conservation Society in the UK and other projects have evidenced (Galloway 2011). As noted, collectors and private enthusiasts have considerable knowledge in the areas of their specialisation (a particular brand of computer, perhaps), but this knowledge tends to be uneven and not joined up.
The AHSD presents an invitation to the public to contribute what they know, so that such knowledge can go towards building a centrally-collated documentation archive. It builds on the method Swalwell prototyped in 2007-8 as part of the NZTronix software preservation pilot. Like her Early New Zealand Software Database, contributors are invited to complete a web form with as much information as they know or can remember (Swalwell 2007a; 2009). This is then entered into a searchable online database. There is the capacity for others to add information if they know more about a title. If people have multiple titles to enter, there is a spreadsheet to facilitate mass entries.
A repository of documentation about software will have several outcomes. Ideally, it will clarify what holdings there are in institutions; as detailed earlier, this information is elusive. It will also help to clarify what collections are held privately. Documentation is not only sought for its own value; it is relevant to collections and preservation efforts (compare Kirschenbaum 2013). Informed judgements are needed regarding what is historically significant and what the targets of preservation ought to be, but this is impossible to do in a knowledge ‘vacuum’, where no one knows the ‘universe’ of software that has been developed. Compiling an evidence base around what software has been locally produced should influence collection policies and acquisitions, though the debates about what to collect and what to keep will no doubt often be had internally within individual institutions.
We are seeking to generate strategic alliances by inviting relevant organisations and groups to support the project, to build a broad coalition in the region which recognises that documenting and preserving early digital history matters, and that there is some urgency to the task. Building an awareness of others with similar aims and current projects underway is an important step in developing a wider discourse about digital preservation generally, and more specifically about the importance of collecting and conserving coded works. We argue that such a discourse is key if larger shifts are to take place in institutional and government policies, and the funding environment. Whilst there are regional coalitions of librarians (National and State Libraries Australasia) and archivists (The Australasian Digital Recordkeeping Initiative), the set of supporters that is forming around the AHSD is more expansive. We are deliberately courting the support of retro computer groups and collectors as well as cultural institutions and industry groups. Digital preservation cuts across a range of institutions and groups (computer societies, libraries, private collectors, and university researchers), some of which don’t talk to each other much at present. Each of these has important contributions to make to digital heritage efforts.
It is our intention that from this broad-based group, new, synergistic collaborations be explored, particularly between different supporters who might not ordinarily consider collaborating yet who share similar goals and perhaps bring complementary expertise. The pioneering preservation techniques and emulators developed by game preservationists are an example here of tools and techniques that are transferable to other forms of software. Collaboration with such groups represents a significant opportunity for cultural institutions (Stuckey & Swalwell n.d.). Another potential collaboration is around the software holdings that individuals in the community possess, which are not always stored in ideal conditions. Cultural institutions have climate controlled storage, and may be able to store discs in their premises short term, potentially slowing the deterioration of storage media (Rothenberg 1995) until collections can be assessed and disc imaging projects conducted.
Whilst the AHSD is still at an early stage, we have had success working with particular collectors who have been willing to compile their knowledge into datasets to share on the AHSD. We have, for instance, very good information from collectors of the Sega SC3000 micro-computer, with some 445 software titles entered, all of which were written or published locally in Australia or New Zealand. In another example, Alan Laughton, administrator of the Microbee Software Preservation Society (MSPP), has created a dataset of the known software for this range of iconic Australian computers. His research compilation indicates that there are in excess of 1,150 known titles for this system, many - but not all - of which the MSPP has in its collection and has preserved, by disc and tape imaging. Some of the titles it does not have are in public collections, but these have hitherto been inaccessible, as software is not lent. Our next step will see us move beyond documentation to: work with the relevant cultural institutions to get access so that MSPP can image these ‘missing’ Microbee software titles, as far as possible completing the collection of software for the Microbee, and house this in a public collection.
A local focus
As we’ve noted, local software history has been overlooked. Whilst there are other supra-national efforts underway which we welcome, such as the IT History Society’s recently launched Software Database, which aspires to “eventually include basic information [on] all software ever created” (IT History Society 2013), it is hard to see how such a massive undertaking could possibly have any depth to it, without local researchers on the ground to unearth the information. Beyond the cultural need to be able to tell our own histories, there are other reasons why a local focus is desirable and necessary. The focus of cultural heritage institutions almost always requires an artefact’s local significance to be clear. We also recognise it is important to bring local stakeholders along on the journey, if there is to be ‘buy-in’. Project funding often requires a demonstrated national and/or local benefit. Finally, as we’ve argued elsewhere, if we want to see our local digital treasures kept and preserved, it is necessary for local institutions to undertake this work themselves. If they don’t, it probably will not happen. Larger countries might have more resources for software preservation projects, but their allegiances will be to their own heritage material; they simply don’t have the same stake in local heritage as the nation(s) from which it has come (as argued in Swalwell 2007b; 2009).
Whilst these projects are small scale and do not provide complete solutions by any means, software history and preservation research is beginning to happen, and crucially, a discourse about the importance of collecting and conserving code is beginning to develop. Play It Again is - to our knowledge - the first digital preservation project to receive research funding in the region. Meanwhile, the AHSD is advancing a series of interlinked projects on software history, with the intention that documentation should support collecting and preservation initiatives. It will catalyse projects from amongst supporters and interested others, exploring and exploiting collaborations and synergies. We take encouragement from the Vancouver Declaration’s emphasis on the urgency of cooperation (between professional associations, organisations, industry and other commercial enterprises working in digital preservation), and the development of supportive policies by member states (legal deposit, copyright exceptions, access). Our explorations are in keeping with its recommendations to explore “a multi-stakeholder platform for the discussion of…digital preservation” (Anon 2012).
That issues around institutional frameworks, models and collaboration are being highlighted at this forum is not a surprise. Where the preservation of born-digital artefacts is concerned, the old silo models - with their attempts to neatly apportion, delineate and contain responsibility - need to be reworked. Digital preservationists working across different types of institutions - galleries, libraries, archives, and museums - have much in common, both with one another and with skilled community-based collectors and preservationists. We know from our projects that some institutions are aware they do not have all the expertise required to undertake preservation of complex digital artefacts; this has served as a kind of ‘brake’, holding them back from doing more. What needs to be recognised is that the required skill sets that might complement those of existing staff often exist in a neighbouring institution and/or amongst specialist groups in the community. Furthermore, no one institution wants to volunteer to take on the responsibility for collecting and preserving all born-digital content. Whilst new institutions might comprise part of an answer to preserving coded works, new frameworks for collaborating between and beyond institutions would arguably be a more significant innovation. We argue that the principles of distributed responsibility and collaborative effort can be taken quite far. There is significant scope for developing partnerships and distributed networks in which skills and know-how are shared, fluidly and transparently. It is encouraging to see the National Library of Australia commit to “working with others to preserve the nation’s digital information resources…[and] to foster digital preservation” (National Library of Australia 2013). One concrete suggestion we have is the creation of a national software collection, to which a range of public and private institutions and community groups could contribute. The scoping of such distributed models will be the subject of our future research.
We wish to thank the reviewers for their detailed reviews; their guidance has provided us with advice as to future papers that should be written to contribute to this area of research. ‘Play It Again: Creating a Playable History of Australasian Digital Games, for Industry, Community and Research Purposes’ is supported under the Australian Research Council’s Linkage Projects funding scheme (project number LP120100218).
Anon (n.d.) “Strategic Directions 2012-2014”, National Library of Australia, viewed 13 March 2013, http://www.nla.gov.au/corporate-documents/directions
Anon (2012) “Towards a Museum of Computing and Digital Society in France?”, viewed 12 November 2012, http://the.cnam.eu/culture-and-knowledge/towards-a-museum-of-computing-and-digital-society-in-france--486990.kjsp
Anon (2013) “Creative Australia - Capturing Our Digital Culture (Press Release)”, viewed 19 March 2013, http://www.minister.regional.gov.au/sc/releases/2013/March/sc038_2013.aspx
Australian Government (2013) “Creative Australia: National Cultural Policy”, viewed 13 June 2013, http://creativeaustralia.arts.gov.au/
Australian ICOMOS (1979) “The Burra Charter: The Australia ICOMOS Charter for Places of Cultural Significance”, viewed 13 June 2013, http://australia.icomos.org/publications/charters/
Barnes, J.G.P. (1982) “The Development of Cultures in Programming Languages” in Ninth Australian Computer Conference (Sale, A.H.J. & Hawthorne, G., eds), ACS. pp. 9-21.
Campbell-Kelly, M. (2003) From Airline Reservations to Sonic the Hedgehog: A History of the Software Industry, Cambridge: MIT Press.
Coates, J. et al. (2009) “The Permission Barrier: The Impact of Copyright Law on Digital Preservation in Australia”,Information Online - Australian Library and Information Association conference, viewed 8 November 2012, http://conferences.alia.org.au/online2009/docs/PresentationC18.pdf
Commonwealth of Australia (1968) Copyright Act, viewed 25 July 2013, http://www.austlii.edu.au/au/legis/cth/consol_act/ca1968133/
de Vries, D. et al. (2013) “Popular Memory Archive”, http://www.playitagainproject.org
Digital Preservation Coalition (n.d.) “Digital Preservation Coalition | Welcome | archiving & digitisation information”, viewed 13 June 2013, http://www.dpconline.org/
Erway, R. (2012) You’ve Got to Walk Before You Can Run: First Steps for Managing Born-Digital Content Received on Physical Media, Dublin.
Galloway, P. (2011) “Retrocomputing, Archival Research, and Digital Heritage Preservation: A Computer Museum and iSchool Collaboration” in Library Trends 59: 4 pp. 623-636.
Gye, L. (2011) “Media Arts Education: Sublimation vs Subjugation” in Realtime 203: 14 p. 36.
Harvey, R. (2005) Preserving Digital Materials, Munich: K.G. Saur.
Harvey, R. (2010) Digital Curation: A How-to-do-it Manual, New York: Neal-Schuman.
Harvey, R. (2011) Preserving Digital Materials, 2nd edition, Berlin: De Gruyter-Saur.
IGEA (2013) “Australian Video Games Industry Records $1.161 billion Sales in 2012 (Press Release)”, viewed 13 June 2013, http://www.igea.net/2013/02/australian-video-games-industry-records-1-161-billion-sales-in-2012/
IT History Society (2013) “Software Database”, viewed 13 June 2013, http://ithistory.org/software/software-name.php
Jones, S. (2003) “Synthetics: A History of the Electronically Generated Image in Australia” in Leonardo 36: 3 pp.187-195.
Jones, S. (2011) Synthetics: Aspects of Art and Technology in Australia, Cambridge: MIT Press.
Kirschenbaum, M. (2013) “An Executable Past: The Case for a National Software Registry” in Preserving.exe: Towards a national strategy for software preservation. NDIIP/Library of Congress. pp. 12-22.
Knuth, D.E. (1998) The Art of Computer Programming: Sorting and Searching, Addison-Wesley.
Knuth, D.E. (1974) “Computer Programming as an Art” in Communications of the ACM 17: 12 pp. 667-673.
Library of Congress (n.d) The Signal: Digital Preservation, viewed 7 June 2013, http://blogs.loc.gov/digitalpreservation/
Lowood, H. (2013) “The Lures of Software Preservation” in Preserving.exe: Towards a National Strategy for Software Preservation, NDIIP/Library of Congress. pp. 4-11.
National Library of Australia (2003) “Guidelines for the Preservation of Digital Heritage, UNESCO, http://unesdoc.unesco.org/images/0013/001300/130071e.pdf
National Library of Australia (2008) “Collection Development Policy - Australian collections”, http://www.nla.gov.au/collection-development-policy/australian-collections-general
National Library of Australia (2013) “Digital Preservation Policy”, viewed 25 July 2013, http://www.nla.gov.au/policy-and-planning/digital-preservation-policy
NOMAD (2010) Scoping Study for a National New Media/Electronic Arts Network, http://mass.nomad.net.au/wp-content/uploads/2010/03/NOMADscoping studynet.pdf
Rothenberg, J. (1995) “Ensuring the Longevity of Digital Information” in Scientific American 272: 1 pp. 42-7.
State Library of South Australia (2006) “Collection Development Policy”, viewed 19 March 2013, http://www.slsa.sa.gov.au/webdata/resources/files/colldevtpolicy.pdf
Stuckey, H. et al. (2013) “Remembrance of Games Past: The Popular Memory Archive” in Proceedings of The 9th Australasian Conference on Interactive Entertainment: Matters of Life and Death. New York: ACM.
Stuckey, H. & Swalwell, M. (n.d.) “Retro-Computing Community Sites and the Museum” in The Handbook of Digital Games (Agius, H. & Angelides, M., eds), New York: IEEE/Wiley.
Swalwell, M. (2005) “Early Games Production in New Zealand” in DiGRA Conference: “Changing Views: Worlds in Play”, Vancouver.
Swalwell, M. (2007a) “Early New Zealand Software Database”, viewed 24 October 2012, http://nztronix.org.nz/main.php
Swalwell, M. (2007b) “The Remembering and the Forgetting of Early Digital Games: From Novelty to Detritus and Back Again” in Journal of Visual Culture 6: 2 pp.255-273.
Swalwell, M. (2008) “1980s Home Coding: The Art of Amateur Programming” in Aotearoa Digital Arts Reader (Brennan, S. & Ballard, S. eds), Auckland: Clouds/ADA. pp. 193-201.
Swalwell, M. (2009) “Towards the Preservation of Local Computer Game Software: Challenges, Strategies, Reflections” in Convergence: The International Journal of Research into New Media Technologies 15: 3 pp.263-279.
Swalwell, M. (2010) “Hobbyist Computing in 1980s New Zealand: Games and the Popular Reception of Microcomputers” in Return to Tomorrow: 50 Years of Computing in New Zealand (Toland, J., ed.), Wellington: NZ Computing Society. pp. 157-169.
Swalwell, M. & de Vries, D. (2011) “Australasian Heritage Software Database”, viewed 25 October 2012, http://www.ourdigitalheritage.org/
Swalwell, M. & Loyer, E. (2006) “Cast-offs from the Golden Age” in Vectors: Journal of Culture and Technology in a Dynamic Vernacular 3: http://vectorsjournal.org/projects/index.php?project=66
Tofts, D. (2005) Interzone: Media Arts in Australia, Fisherman’s Bend: Thames & Hudson.
Tofts, D., Jonson, A. & Cavallaro, A., eds. (2002) Prefiguring Cyberculture: An Intellectual History, Cambridge; Sydney: MIT Press; Power Publications.
UNESCO (n.d.) “The Memory of the World in the Digital age: Digitization and Preservation”, viewed 7 June 2013, http://www.unesco.org/new/en/communication-and-information/events/calendar-of-events/events-websites/the-memory-of-the-world-in-the-digital-age-digitization-and-preservation/
UNESCO (2003) “Charter on the Preservation of Digital Heritage”, viewed 18 December 2013, http://portal.unesco.org/en/ev.php-URL_ID=17721&URL_DO=DO_TOPIC&URL_SECTION=201.html
UNESCO (2011) “The Moscow Declaration on Digital Information Preservation”, UNESCO Information for All Programme, viewed 18 December 2013, http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CI/CI/pdf/mow/Moscow%20Declaration%20on%20Digital%20Information%20Preservation.pdf
UNESCO (2012) “UNESCO UBC Vancouver Declaration”, viewed 17 December 2012, http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CI/CI/pdf/mow/unesco_ubc_vancouver_declaration_en.pdf
Webb, C. (2003) “Saving Digital Heritage - A UNESCO Campaign” in RLG Diginews 7: 3 pp.1-6.
Zurbrugg, N., ed. (1994) “Electronic Arts in Australia (Special Issue)” Continuum 8: 1.
 Many early programmers for personal computers worked alone and their revision control was handled by saving multiple versions of the source code, which were then deleted when the final version was completed. While revision control was used in industry in collaborative projects, it was not the norm for hobbyists and sole proprietors.
 Most source code was (and is) written in plain ASCII format. Many of the early programs were written in assembly, BASIC, COBOL, and Fortran. Assembly, COBOL and Fortran programs were compiled to produce an executable file while BASIC programs ran under an interpreter. The difference, for preservation purposes, is that the source code is kept intact when it is interpreted whereas the source code is often lost during compilation. Also during compilation all comments are stripped and formatting is deleted so that any unnecessary information has been removed to save on space requirements. Decompiling or re-engineering machine code does not give the original source code, only the original logic. Ironically, then, whilst BASIC was sometimes maligned as a crude and even “disgusting” language for amateurs, it may be that as a language, it survives the passage of time better than “professional” languages (Barnes 1982).
Melanie Swalwell is an Associate Professor in the Screen and Media Department at Flinders University and an ARC Future Fellow. Melanie’s current research is on the histories of creative computing in the 1980s. She is Project Leader of the multi-disciplinary Linkage project Play It Again. Together with Denise de Vries, she runs the Australasian Heritage Software Database, www.ourdigitalheritage.org
Dr Denise de Vries has, since the early 1980s, developed commercial complex database systems on a variety of platforms from mainframes to a range of personal computers. She is currently a lecturer of computer science in the School of Computer Science, Engineering and Mathematics at Flinders University. Denise’s current research is on techniques to preserve digital history and data semantics including techniques to deal with changes to information in a database such as structural change, semantic change and constraint change. She is a Chief Investigator of the multi-disciplinary Linkage project Play It Again and developed the Australasian Heritage Software Database, which is co-managed with Melanie Swalwell.