“Caring for Collections of
Software-based Art”
Panel of Three Presentations
November 23, 2025
Software-based art is particularly fragile and as a result, collecting it presents a significant challenge. The actions needed to preserve these works are myriad, resource-intensive in many ways, and difficult to accomplish under the best circumstances. Moreover, the interdependencies between technologies, the rapid pace of technological change, and the barriers erected by commercial interests may make collecting these works in the traditional sense all but impossible. Object-focused conservation is often undertaken at intervals determined by schedules of exhibitions or loans, which can be appropriate for artworks in some media but is problematic for software-based art, which cannot survive long periods of passive storage. What, then, are the efforts that conservators and institutions can undertake at the collection level or even at the community level to sustain these works?
This video of a 3-person panel touches on collection-level strategies for software-based art conservation including:
- Intake processes and priorities
- Collection monitoring strategies
- Management of obsolete equipment
- Preservation of computing environments and supporting software
- Storage and digital preservation
- Networks of care and collaboration
Dragan Espenschied is Preservation Director at Rhizome, stewarding ArtBase, a collection of more than 2200 works of digital art and net art. With a background in net activism, net art, and electronic music, Espenschied’s activities as a conservator are mostly focused on infrastructure and field-wide action concerning web archiving, emulation, and linked open data, rather than singular artworks.
Patricia Falcão is a Portuguese time-based media conservator at Tate, where she researches and develops strategies for the preservation of software-based artworks. More recently, in the context of the Reshaping the Collectible project this has broadened to include the acquisition and preservation of web-based artworks. She has consistently published on the theme of preservation of Time-based media, Digital and Software-based art over the last 8 years, in the Conservation and Digital Preservation Communities.
Claudia Roeck started in environmental engineering and later studied conservation of contemporary art in Berne, Switzerland with focus on media art, graduating in 2016. From 2013 to 2016, Claudia worked at the Tate in London as an assistant time-based media conservator, focusing on the acquisition of video- and film-based artworks. Claudia earned a PhD from the University of Amsterdam in 2024 and works with software-based artworks at the House of Electronic Arts in Basel, Switzerland.
Amy Brost (organizer/moderator) is Associate Media Conservator at the Museum of Modern Art (MoMA) in New York and works with the Media and Performance collection as well as the museum's digital repository for art. She has worked in-depth on sound elements in media installations, documentation of media artworks, and digital preservation and storage of media art, including software-based art.
When coming up with documentation that may one day represent the artwork, do you work with artists and curators on it?
Claudia: For documentation as a presentation strategy, I think it's very important to involve curators because you want to tell a certain story, and very often you have to choose, for instance, if you have a generative work, and then you need to choose somehow what you want to present because it's not generative anymore if you freeze it. So you might even create several versions of it, several documentation versions for different kinds of publics, or exhibitions. I did just tests, but if you really want to do it seriously, you need to involve curators and other people.
Amy: And I'll just say that we actually recently acquired a complex work here at MoMA where during the intake process, the artists came to understand the fragility of it to the point where they created documentation that could one day be presented, properly contextualized. So it's definitely a conversation that's starting to happen even from the artist and studio side.
What do you think is the most important community level action that those of us with software-based art collections could potentially take that would support preservation efforts more holistically at a community level?
Dragan: I think that the exchange of material artifacts can be made to embody a lot of knowledge, for instance. For a project with Claudia, I spent a lot of time configuring a Windows 98 environment that I'm pretty proud of. And I've seen it happening at institutions that they start from scratch. They want to get into emulation and then they start from scratch, and they buy Windows 98 licenses from god knows where, and start doing all this research again. And I think it would be really great if that exchange would be facilitated in an easier way, which we are trying to do with the HEK now. It's not like there is a tool that you can use that does it. So that's a little bit of a challenge. So we share infrastructure, but we also want to have multiple instances of infrastructure. And I think the cataloging of software is very important. It's something that can be shared, but it's also something that needs to be done when the data is not at rest, when the data is executed, when the software is running, because you can read about the software and know the publisher and intended OS and so on, and that information was right at the time, but it might not be anymore. And so collecting information for the preservation context about software is quite a thing that I believe we have a good system for, but sharing that is still not there. So I think that would make a huge difference.
Claudia: I would like to support that. It can be a lot of work to set up software environments because they are usually old environments and you need to find this old software. And so it can be so useful to share this information, for instance, through this emulation as a service (EaaS), but it could also be something else. I just want to mention Jack McConchie and Tom Ensom's project about preservation of immersive art [see links below].
Patricia: I think emulation and the sharing of environments would bring us so much further along. Having something like EAASI, which is this platform that helps you use and build emulated environments to make it easier for lay people who don't do it regularly, is key. There are limitations - technical, legal, a fairly high learning curve, etc. So maybe that's the other point we could, as a community, try and address. I don't know how easy it is and there are many jurisdictions. But I think that's the sensible thing to do next.
Amy: And I also think that the cataloging aspect, sharing a data set about our computer-based and software-based art collections that enables people to find one another, right? Because then if I'm looking for somebody who has a work that they've treated that requires a particular environment, then I can find that with a published data set. All of our information right now is currently in narrative form. So I know that this will take a lot of labor, but it's the kind of labor that we want to do that will have the greatest impact on our collections. So I think that's very valuable.
Patricia, you mentioned there's a certain amount you can accomplish in X number of days and if you have a little more time you can maybe accomplish a little bit more. And I was wondering if you could speak to how you allocate that capacity and perhaps each of the panelists, how do you do that? I guess it's kind of a triage approach. How do you approach that?
Patricia: I mean it's not systematic. I would say in our case it's a little bit how well do we know this type of technology? And very often we don't know much. So we have very few environments that are the same. We've got two pieces by the same artist that are very similar. And then we are seeing more. We have a few works acquired on Unity, the game engine, where I can see we could start thinking of it as an environment, but otherwise it is sort of looking at what the description of the artwork is and trying to understand exactly what we're going to receive.
And then there are some basic steps that we would do. We would want to have the work running, create a disk image where we know the work is running correctly, store those things in our digital storage and just have the instructions on how to show it would be the minimum. And then depending on the work, so now we sometimes use this as an opportunity to do a little bit of research as well. I really make strong cases at the point of acquisition that we do as much as we can then, because it's much harder to get time and budget afterwards unless something is going on display. But then usually we are also short of time to do things. So sooner is better. Always sooner.
Amy: And Claudia, how about you? When you're working on acquisitions, how do you sort of determine how much to put into the intake phase and scale that effort?
Claudia: I mean now it's easier because I have much more time, but before that I sometimes had only, I think, two days for an artwork maximum. And then I just checked the basics. Are there any external dependencies, does it work without the internet? Do I have all the artwork components, can I run it? And then I just, yeah, archive what I got. But I totally agree the acquisition phase is the most important one. But in the beginning some artworks were also a bit simpler, so I did need a little bit less time and then they became more complex and then I did disk image as well. And that takes more time than just a day or something to figure out how to do it because every hardware has, again, a different computer, different storage media, et cetera. And now it's easier, I must say, now I have the time to analyze the artwork more deeply. But in the past, when I had very few hours per artwork, I wasn't able to do a thorough analysis of the work. I just tried to do a quick one.
What do you think about the term ‘best practices’? What would you say to someone who's working in a collection who has a lot of worries about best practices and being able to achieve best practices with their collections?
Amy: Personally, I think that ‘good enough’ practices are the goal because then you can thoughtfully bring the greatest number of software-based artworks into a collection.
Patricia: I agree with you. Recently I've started thinking, actually if there's an entry on TMS and it tells me that there's a computer and disk image and I have no idea what's on it, it's so much better than if it's not there. And I know as a conservator it's like, oh, this is so bad. I don't know what it is and I don't know what to do with it. But it's much better than not having it there. So maybe we need to let go of this idea of best practices. And if it was just me doing software-based artworks and we didn't have the whole of the Tate team working on this, I would just put things on TMS and a good catalog entry on the collection website. That’s probably going to do more for an artwork than having only one piece that you can play in 10 years (and then you never truly know if you can either).
Claudia: For me, a good practice is if you think about preservation policy, you need to think about, how do I spend the money and the resources and to be honest about it. To be honest and say, okay, we can't do everything. We have to set priorities, or we have to apply some kind of documentation strategies and reduce dependencies. So I think that's good practice to think about the policy you have and to be realistic. It can be policy to say, okay, you are only focusing deeply on the works that go on exhibition. It's fair enough, but then maybe you have to still think about what you do with the rest. Maybe you have a different treatment for them.
Dragan: Yeah, I would agree with both of you. I think doing something is always better than nothing, and some nasty cell phone video of some installation or so is better than not having that. Also, for instance, my idea has always been that software always has to perform. I don't want to look at documentation of software. That was 10 years ago, and now I think, oh wow, there's a video of that software. Isn't that fantastic? Now I know how it looked, now I know how it behaved. I so much love this Michel Majerus series by Cory Arcangel. There's a very knowledgeable person going through that computer, interpreting the system and explaining it. That's incredibly valuable. And I would just say keep the copies around and at some time in the future you might be able to make it work. So I don't throw things away, but I'm fine if you just look at them as documentation.
Claudia: By the way, a lot of artists make documentation of their artworks too, and they even see them sometimes as artworks. And they do it themselves. So I think it would be really important of course to do it with them, with the artist obviously, but I think they might not even be so opposed to it.
Amy: Exactly. Bringing documentation into the acquisition process in a different way is something I'm definitely taking away from the conversation. Thank you all for this exciting discussion!
In no particular order, below are research papers and projects by the speakers or suggested by them for further information:
Claudia Roeck. 2024. Sustaining Software-based Art (PhD Thesis) [PDF]
https://hdl.handle.net/11245.1/daccbe73-fc09-45d2-8798-c733f009af41
Roeck Claudia. 2021. Web browser characterisation, emulation, and preservation.
https://zenodo.org › records › 4476030 › files › NDE_2020_browser-emulation.pdf
Tate Academia repository, Patricia Falcão
https://tate.academia.edu/PatriciaFalcao
Rechert, Klaus, Patricia Falcão, and Tom Ensom. 2016. Introduction to an Emulation-Based Preservation Strategy for Software-Based Artworks. http://www.tate.org.uk/research/publications/emulation-based-preservation-strategy-for-software-based-artworks.
Dragan Espenschied at Rhizome
https://rhizome.org/profile/despens/
and iPres:
Espenschied, Dragan, K. Rechert, Dirk von Suchodoletz, Isgandar Valizada, and Nick Russler. 2013. ‘Large-Scale Curation and Presentation of CD-ROM Art’. iPRES.
Espenschied, Dragan, and Klaus Rechert. 2018. ‘Fencing Apparently Infinite Objects’. Paper presented at IPRES 2018, Boston, Mass. Proceedings of the 15th International Conference on Digital Preservation (iPRES). https://doi.org/10.17605/OSF.IO/6F2NM.
Espenschied, Dragan, and Klaus Rechert. 2023. ‘Software Preservation after the Internet’. Paper presented at iPRES 2023: The 19th International Conference on Digital Preservation, Champaign-Urbana, IL, US. Proceedings of the 19th International Conference on Digital Preservation.
“A Condition Assessment and Documentation Framework for Multi-variant Works of Net Art”
By Nicole C. Savoy, Dragan Espenshied, and Martina Haidvogl
https://ipres2024.pubpub.org/pub/ruud1i2r/release/2
Software-based Art Preservation: A research project into the care and preservation of software-based artworks in the Tate collection. October 2017 – ongoing
https://www.tate.org.uk/about-us/projects/software-based-art-preservation
"EAASI" platform in the Software Preservation Network and the EAASI Research Alliance: https://www.softwarepreservationnetwork.org/eaasi-research-alliance/
Preserving Immersive Media Knowledge Base
By Jack McConchie and Tom Ensom
https://pimkb.gitbook.io/pimkb
“Preserving Software-Based Art at Tate: From Research to Best Practices”
By Tom Ensom and Patricia Falcão
Electronic Media Review, Volume Seven: 2021-2022
Preserving and sharing born digital and hybrid objects
https://www.vam.ac.uk/research/projects/preserving-and-sharing-born-digital-and-hybrid-objects
Software Heritage
https://www.softwareheritage.org/
Software Preservation Network