This ARDC series aims to drive recognition of research software and its authors. Each month, we talk to leading actors in the research software engineering (RSE) space and share their experience creating, sustaining and improving software for research.
Melanie Swalwell is Professor of Digital Media Heritage at Swinburne University of Technology. Her research focuses on the creation, use, preservation and legacy of complex digital artefacts such as video games and media artworks. Melanie has recently led 2 digital heritage ARC Linkage Projects and a LIEF project:
She is currently leading a second LIEF project titled “The Australian Emulation Network Phase 2 – Extending the Reach”, funded by the Australian Research Council.
I fell into the game history field by chance when I went to New Zealand for my first full-time academic position. Not long after I started in 2004, I was approached by a curator from a New Zealand museum to undertake background research for an exhibition they were considering mounting on digital games. They were particularly keen to know the local history of games in New Zealand. I discovered a network of very knowledgeable collectors who shared some extraordinary stories of local production and global anomalies in distribution. That led me to wonder who was collecting this unique cultural heritage. In 2005, I brought out the German advocate for game preservation, Andreas Lange, and put together a multi-disciplinary team at my university to pilot game software preservation in 2007.
I returned to Australia and teamed up with Helen Stuckey, then the inaugural Games Curator at ACMI, who had curated an exhibition on Australia’s first game studio, Beam Software, in 2006. Together with Angela Ndalianis, a theorist of fan cultures, and Denise de Vries, my Flinders colleague in Computer Science, we were awarded an ARC Linkage Project grant in 2011 for the first Play It Again project, which focused on games written for microcomputers in 1980s Australia and New Zealand. We’ve recently completed the second, 1990s iteration of Play It Again.
It really does take a village! No one person or discipline can pull this off on their own, or at least not across all the dimensions that game and software preservation touches on. A large part of our success can be attributed to the fact that we’ve had a mix of people bringing complementary skills and expertise to the problem: technical, media historical, legal, and curatorial, working in conjunction with professionals in cultural institutions.
It’s important to start – particularly stabilising content on deteriorating media carriers – even when full end-to-end solutions might not yet exist. Digital and especially game preservationists are a tight knit international community who are generous in sharing solutions, so it’s likely someone will come up with at least part of the solution to a problem while you’re taking the first steps. Reframing the historically quite negative discourse about software preservation – from ‘this is too hard’ to ‘these are challenging problems but we can progress solutions by working together’ – has also been important in giving people and organisations a way to get onboard and see positive results.
Reframing the historically quite negative discourse about software preservation – from ‘this is too hard’ to ‘these are challenging problems but we can progress solutions by working together’ – has also been important in giving people and organisations a way to get onboard and see positive results.
Game preservation has always held the promise of transferability: getting a legacy game to run means you’ve learnt a lot which you can apply to other types of software. The challenges exist in lots of domains and are pretty similar: deteriorating external storage media and dependencies on old computers, operating systems and legacy software. We applied the lessons from the games projects to the media arts domain, and have been expanding into ever more diverse types of software-dependent artefacts under the auspices of The Australian Emulation Network.
While emulation is wonderful, it helps to have working vintage hardware around. I’ve watched software preservationist Cynde Moya troubleshoot an artist’s Amiga disk that comes with no documentation, piecing together clues from the error messages she receives on inserting the disk into an original Amiga computer. Denise de Vries was a visionary in this respect, seeing the need and gathering vintage computers (and software) from her Computer Science colleagues. We’ve run a number of Side by Side events, where we invite people to interact with a legacy game or CD-ROM artwork running in emulation (on the AusEAASI network) and on original hardware. It allows people to compare performance as well as to experience legacy hardware and software – the OS startup sounds, CRT screens, colourful iMacs, etc.
Much of the knowledge about vintage games and game preservation (and also other types of legacy computers and software) isn’t held within collecting institutions: it is held within the enthusiast community. Cultural institutions have been quite slow to come around to preserving born-digital cultural heritage. Now that they’ve realised it’s possible, we’re in a transitional moment where the knowledge is being acquired by the people who are formally charged with looking after cultural heritage – archivists, conservators and so on.
Institutionally, ACMI has been a partner on our research since the beginning of preparations for the first Play It Again project around 2009. Our technical collaborator, Denise de Vries, spent sabbaticals at the University of Freiburg from 2013 to 2014 and again in 2017 with the team of researchers who were developing the Emulation as a Service (EaaS) platform (now EAASI, pronounced ‘easy’). She realised it was a very valuable tool for displaying complex digital artefacts in GLAM institutions. ACMI were keen to try it out and suggested AARNet might also be interested, so we wrote an evaluation of the EAASI platform into the second Play It Again Linkage Project we pitched in 2017. Visitors to ACMI can now play many Australian games from the 1980s and 1990s preserved in the 2 Play It Again projects on their own device using EAASI within ACMI’s wifi footprint.
We also used EAASI with great results in another Linkage Project run simultaneously, “Archiving Australian Media Arts”. This project saw us scaling up the research-GLAM sector collaboration, building a consortium of collecting institutions: ACMI, the Art Gallery of New South Wales, the State Library of South Australia, and Griffith University Art Museum. The 3 large cultural institutions each accepted the archive of a media arts organisation – Experimental Media Arts, dLux media arts, and ANAT (the Australian Network for Art and Technology), respectively. Not only did we manage to preserve and make legacy media artworks accessible again, but there were significant gains from the collaboration. The collecting institutions now have not only preserved digital artworks available for researchers but also the know-how to preserve and emulate other complex digital artefacts. EAASI has been a real game changer.
Based on these 2 Linkage Projects, we’ve continued to scale up the collaborations with 2 consortial ARC LIEF grants funding the roll out of The Australian Emulation Network, known affectionately as AusEAASI. We’ve not only been able to make AusEAASI available to many more academic researchers and GLAM institutions, but also to auspice a thriving Australian community of practice in software preservation.
EAASI allows a user to build a stack, selecting the appropriate emulator for the particular type of computer, then the operating system, and any utility software, browsers, plugins, etc. When such an environment has been built, it can then be shared so others on the network won’t need to repeat that work.
Emulation means that formerly inaccessible content can be accessed again. In disciplines like media studies and games history, people need access to the actual legacy content – up till now they’ve either had to figure out their own solution or rely on memory or written accounts.
Emulation itself is not new, but the EAASI platform significantly lowers the bar by centralising the technical infrastructure via an EAASI Service Provider (in our case, AARNet) so that users can access content via a webpage.
We will soon have more than 40 Australian academic and GLAM sector organisations using the AusEAASI network, and it’s only the beginning! Soon it will be realistic for any researcher to access legacy content in emulation. It means that more disciplines will be able to embrace their digital pasts. It also means that collecting institutions are more likely to consider digital heritage acquisitions because they have confidence they’ll be able to make these accessible.
Currently, visitors to ACMI can access games from the Play It Again project (1 and 2) and some 1990s media artworks. The Art Gallery of New South Wales is in the process of making legacy media artworks accessible for researchers to access in the National Art Archive.
LIEF Chief Investigators are specialists in Media Arts, Games, Design and Architectural History and Museum Studies so they are working to emulate specific archival content in these domains. For example, architectural historians Kirsten Day and Peter Raisbeck at the University of Melbourne have been working with disks from the archives of Australian architects, William J. Mitchell[1] and Greg Burgess, which require a range of different CAD environments to run in.
There are a lot! We work closely with the Software Preservation Network, which is the body that manages the International EAASI Research Alliance. Domestically, our partners encompass many national and state GLAM institutions and universities.
Participating GLAM institutions have a range of content they are emulating, with many unpublished collections or databases of content on CD-ROMs that they’re keen to regain and provide access to. Often they do not have the utility software that an artefact requires, so the ability to share configured software environments over the EAASI network is an enormous benefit. Re-using an environment someone else has created saves everyone time and ensures the best use of scarce legacy software. (We rely on the libraries and archives research exception within the Copyright Act for the preservation and sharing of such configured environments.)
The consortial approach we’ve taken with the 2 LIEF projects emphasises collaboration and GLAM and university staff learning together, because few people had the skills to make use of the EAASI network at the beginning. Participants have found the training our resident software preservationist, Cynde Moya, provides in imaging obsolete media and configuring environments in EAASI invaluable. The support of regular community of practice zoom meetings – to share problems and solutions with other digital preservation professionals – has also proven very beneficial.
I have been seeking to scale up collaborations to include international organisations that are interested in working on the very significant research challenges that remain, particularly as regards contemporary digital cultural heritage. It’s too soon to say anything about that, except to underline that there is a significant appetite for collaborative research amongst organisations in this space, both domestically and internationally.
In some ways, we’re in a brave new world with commercially available software. In the past, we might have said, “Document the system requirements and keep the utility software and operating system”. But these days, that’s difficult to do because software is no longer on external storage media; it’s a download or just a subscription to a service, so people don’t have a physical copy they can keep. The Computer History Museum has suggested that “collecting commercial proprietary software … is becoming practically impossible”[2].
Those who create their own artefacts and software have more control. Research software engineers are in quite a good position to keep their software accessible. Research data repositories have not always collected the software that goes with legacy research data, which puts FAIR data principles and reproducible science in doubt, so it is a conversation that really needed to happen some time ago. But the next best time to have the conversation is now.
The software preservation and emulation infrastructure that we’re rolling out in conjunction with the Software Preservation Network and AARNet will have far reaching benefits for the broader Australian research community. While our LIEF projects focus on art and cultural domains, the shared digital infrastructure is going to be of use to any area that needs to study complex born digital artefacts, which are all dealing with the same issues of long term change in hardware, OS and software. If we are to retain access to such digital content, it’s critical that we have the ability to render both legacy and contemporary digital artefacts.
You can connect with Melanie Swalwell via LinkedIn and Bluesky. You can send inquiries about joining AusEAASI to [email protected].
If you’d like to be part of a growing community of RSEs in Australia, become a member of RSE-AUNZ – it’s free!
This research was supported (partially) by the Australian Government through the Australian Research Council’s Linkage funding schemes (LP180100104, LP180100307, LE220100057, LE250100051). The views expressed herein are those of the author and are not necessarily those of the Australian Government or the Australian Research Council.
The ARDC is funded through the National Collaborative Research Infrastructure Strategy (NCRIS) to support national digital research infrastructure for Australian researchers.
