Artists Running Data Centers

Edited by Davide Bevilacqua

0. Introduction #

1. AMRO - Art Meets Radical Openness

2. Within Re:ARDC [Rethinking the Artist Run Data Center] we continued researching the broader context of servus activities, investigating the topics of autonomous infrastructure and self-hosting, as well as the series of server projects that have been set up over the last decade in the data center, the machine called ARDC. Over the process, we exchanged with several actors of the media art and self-hosting scene that have surrounded servus since its beginning. |9

1. What can be an art server today?#

from Manu Luksch

3. For a contemporary art server it would be very important to consider both, like an urban lab where you consider which values are important for the urban and public space, but at the same time connect that to something like an art server. |13

2. Media art reimagining the information society#

from Franz Xaver

4. There are different types of randomness, and it's important to have real noise generators to produce random numbers. It's very important in cryptography. Real randomness is simple the key to getting out of this reality or being able to cope with it all, because the boundaries of reality have to be somewhere... You can also see true randomness in atomic decay, or - I also have been operating radio telescopes for a very long time - in starts and in the past traces of the Big Bang, or in semiconductor transitions.

5. Q: It's not just about the number that is generated. The system must ensure that the values or the events that are generated with it will never repeat.

A: Yes, true randomness is important because otherwise there would be no free will. Free will is also an issue, it is also related to true randomness. Because only if there is true randomness can there be free will. If something is deterministic or the values come from software – these are not genuine coincidences – then there can be no free will. |17

6. Stadwerkstatt had a big sign with the slogan "Free opinion needs free networks, free opinion needs free infrastructure". It was also logical for Stadtwerkstatt to work on its own infrastructure, in-house, so as not to become dependent on a commercial company or system and allow itself to be enslaved there. |19

7. Like the printing press, it's simply a development of evolution. The whole evolution is just there to develop information. All the genes, the transmission, the consciousness and the printing press develop information. And I believe that man is only a tool. Humans are not the end of the line. There is much more to it. |20

3. We never wanted to do "clean" art#

from Tanja Brandmayr $ Claus Harringer

8. Quite a lot of people who know Versogerin are surprised that we are also doing art projects, running a club and a house, and so on. And it's nice to hear! Because each area seems to work successfully. We don't have as many people as it may sound like, so we must use what we have as a strength. And this kind of working together transfers somehow to the outside also.

The spheres are constantly intermingling. Stadtwerkstatt never wanted to be a “usual art institution”, we never wanted to do the “clean” art. This is strongly rooted in the history of this space. That means for example that in art contexts now we declare ourselves as an Anti-White-Cube. We focus more on an unpolished perspective of it, and for me, it must remain open and connected to society and technology. With the question “How do we want to live?” we address the importance of critique and how we can keep all of this in the art scene, and not play the art market or any market game. |23

9. Q: In a broader sense, servus, as a net culture initiative, shares a similar role with Stadwerkstatt in balancing art and infrastructure, while also supporting a wide range or artistic projects.

 A: I also see it as a metaphor for the role of servus' infrastructure. It's about artists gathered around a center or hub - serving both as a space and a concept. There was definition that Fino said that Stadtwerkstatt is a music space that spans a role that goes from hosting concerts and events till just before it becomes a professional club. And it covers a spectrum that stops just before becoming the institution. |27

4. It all started with normal computer#

from Didi Kressnig

10. Q: How did the data center change over the years?

A: It all started with normal computers that you just put in a cupboard or rack, servers were simply too expensive back then and there was not enough money. As we received more and more funding and the servers became a lot cheaper, we were able to take the next step and purchase real server infrastructure. In the last few years, for example, we have acquired seven new nodes. You can think of nodes as one or two height unit servers on which all the services such as mail, cloud, etc. run, so basically hardware on which a bunch of VM’s (virtual machines) run. Otherwise, we simply made sure that as soon as the server structure was in place, we kept it up to date and looked to increase storage and computing capacities. |29

11. What I love is that I have almost complete freedom here in terms of what I do and when I do it. I can also experiment a lot, which is great for me. For example, if members and their websites are not so well secured or updated, I experiment a lot with how I can separate the websites on our servers. So if one website is compromised, I make sure that not several are affected and that the pages are separated from each other.

12. Q: A question regarding servus. Do you see the need for a generational change? Is it even possible to do a handover?

A: I’m over 60 and won’t be doing this forever. I’m not thinking too much about it at the moment, but of course, it will become an issue at some point. The main problem, however, is finding someone who won’t just do it for three or four years but will really stick with it for the long term. The transfer of knowledge itself is possible, especially if the person deals with the existing documentation and continues their own training. But of course, there has to be a personal exchange. |31

5. We'll call the network Eliot#

from Aileen Derieg

13. Q: Did [finding linux and free software concept] it fascinate you?

A: Yes. Translating the introduction [self-employed translator] about what Linux is, my first impression was that this is a feminist strategy and I got really excited.

Q: Where did you see this?

A: THe idea of collaborating and sharing knowledge, emphasizing that the work is a collective effort rather than just one person doing it.

14. Too many people got too interested in how to make lots of money with the internet. In addition, the idea of communication and connection got lost because everything was just focused on how can you turn this into a profit. I think in free software and open source, a lot of people just couldn't keep up with constant volunteer work and then the big companies like Google and Yahoo became a monster. |35

6. From Intermedia Research to Art Meets Radical Openness#

from Christoph Nebel

15. This conversation with Christoph Nebel opens up the early times of media art in Austria and focuses on how a scene working with critical networks developed over the years. The first part focuses on collaboration, experimentation and the use of new digital tools.

Christoph tells the underside of projects such as HILUS, Open Circuit and UNITn, which have been fertile ground for cooperation approaches within art and technology. We move then to the edoption of free software in arts and discuss LiWoLi and its evolution into AMRO. |37

16. Q: What it a space where you gathered?

A: No, it was just an idea. We always met in our apartments because we had no space at the time, I think that was around 1991.

At that time, we were just four or five people, we were aware that everyone knew so many different kinds of works and had different interests. We started to work on concepts and we thought to set up an infrastructure in Vienna with discussion about analogue and digital communication tools, and space for seminars and workshops. As you know, around 1994, the internet became commercial, but we have been already working with Gopher and other basic internet communication. It wasn’t about integrating existing digital art networks or exchanging data and information – it was about fostering dialogue. Our main objective was to provide experience and advice to independent media artists and connect them to the digital network.

We did a lot even though we still didn't have space or funding - nothing. We started to communicate, to do research about artists, institutions, organizers, galleries, databases and groups, and started to think about what to do with it. It was also a critique at the time that we thought some institutions already should do this, like museums, but they were very conservative and didn’t want to show anything with electricity because “if the power is out, the artwork doesn’t exist”. There- fore, we had to do it on our own.

17. The first event that HILUS realized was Open Circuit4. It was based on a concept called the Roundtable Talk for Synchronous Information Exchange. The idea was a three-day retreat for 40 participants around a huge round table in Pöllauberg. It was there because we thought it shouldn’t be in the city of Vienna, where everyone can go out at night. We brought together art groups, artists and initiatives, which sat at the table, with the second row of theorists, journalists and scientists. It was forbidden for the latter to talk and we said also we didn’t want any article or information to go out. Of course, some of them were quite annoyed but still accepted the invitation. So, we drove to Pöllauberg by bus and every invited initiative had half an hour to present their project. Afterwards, everyone knew about what was happening at the time. Probably the same would be happening in a messaging app today. |38

18. This was an incredible event and of course, everyone visited the others to have a look. After this beautiful event, there were strong strategies to institutionalise the project, secure funding and sustain NetzNetz as a network. However, conflicts over money and responsibilities led to its collapse. There is always a common risk: transitioning from a dynamic event to an ongoing project often brings administrative challenges and decision-making struggles. There was a participant (Stefan Lutschinger), who brought in an important text: “The Tyranny of Structurelessness” by the American feminist Jo Freeman.

Written around the ‘60s, it explores power dynamics within radical feminist collectives. For me, it was a crucial text as it illustrates how failing to consciously structure a movement can lead to dysfunction and collapse when trying to institutionalise it. |41

18. I realised that all the servers hosting Linux distrbutions were at technical universities. And I got the information from supporting websites and Linux user groups that people working on software need more artists because otherwise they would not be able to imagine which new problems to solve - artists or cultural workers often need to go beyond the limits of existing systems. Such contact should happen at the University of Arts. |42

19. It would be very important that there's no proprietary tool which tells you what you can do and what not. It would be worth focusing more on the concept and using open-source software. And if you come to a point where something is not possible, get in contact with the community and give this important input to them. |43

7. Beyond the battle for operating systems, we have to continue on the political level#

from THomas Warwaris

20. For me, it was important to have a kind of handover and to do something new, and get new people involved. That should happen without any vision or planning from my side. It was basically to stop doing it and trying to see what came up next. I think the original energy moved into /dev/lol[5], that's what my impression was, it got more into Linux, DIY, and maker-scene. There's a lot of professionalism in the Linux scene right now, but I wouldn't count that as community. I think the idea of community as a big block wasn't something that was happening. |47.

21. AMRO meets two main ideas. one is the idea of free software, mastering your tools, which is a nice idea but has to be taken with a grain of salt. The other is art, and to me, it’s where these two passions meet together, artists seeking openness in their work. This changed from the earlier days when it was more IT enthusiasts driving this forward.

Going back to your previous questions, I wouldn’t say that Linux has won the war on operating systems. I think that the challenges for free software are now more political and less something that can be solved by expanding the user base or lecturing about home user topics. I think there will be a harder disrupt production processes and money flows, which affects art funding. |48

22. Q: You will have for sure more young people coming to you to ask questions about AI and copyright. At least I will. Maybe to close this conversation, do you want to share a final thought on artist-run data centers? Where do you see their future?

A: It’s a mixture of feelings. I’m a bit sceptical of running the old outdated tools like email. I think there is still a need for running your own collaboration tools. I don’t think that there is a possibility for do-it-yourself to compete with the big social media companies. I don’t think Facebook or TikTok can be replaced, because they are much more than just tools. The realisation that we can’t control this by DIY approaches should lead to the idea that this has to be resolved on the political level. |49

8. Running data centers and researching radically open art practices#

from Us(c)hi Reiter

23. The Linux user group rented some rooms in other spaces for LiWoLi, and servus and Stadtwerkstatt were the locations for the cultural program. Then around 2008, Christoph Nebel, who had always been interested in this community, suggested hosting LiWoLi at the Art University, which hosted the editions from 2008 onward. They were still curated by the Linux User Group community, however servus got more and more involved. It was good working together with nerds, but there was a point where the Linux community and servus couldn’t work so well together anymore. There was too much art for them in the program. They received less and less interest from the outside… It felt like something developing naturally.

24. I started looking for artists who were already using F/LOSS tools, because LiWoLi was also very tool focused. Political issues were addressed in the program too because using free and open-source soft- ware has a political aspect as well. At a certain point, it was not only about tools anymore, but about the dynamics of the software industry and what it meant using software. The question of what’s behind such software was coming more and more to the foreground and started being the central question of the festival, which was renamed Art Meets Radical Openness in 2012. |52

25. While I was looking for artists using free and open-source software tools to produce art, I found many that generated their income over sharing their knowledge and didn’t participate so much in the circuit of white cube galleries. I kept being interested in such artists who had political opinions and lived and produced with a principle of openness. |53

9. An Artist Run Data Center in the context of Free Software and Autonomous Structure#

from Peter Wagenhuber

26. I was fascinated by the idea of free software, because you’re doing something and not only giving away the product for free, you’re giving away all your knowledge, so it’s also an inherent knowledge transfer. You don’t try to control what happens with your thing, so it’s free. I can use it, change it, improve it and I can give this improved version back.

There is also a bridge between technical and social, usually, in open-source projects you have direct contacts. I can get in contact with the people who do the code. I can say, “This doesn’t work for me,” and “Could you please help me with this bug?” or I can volunteer and translate the software, even if I don’t know how to code. So, there are ways to interact and to be a member of the community that writes it. Everything is possible there.

27. Q: At servus, system admins don't code so much, but they keep up the machines and are also at the center of a techno-social construct of data center, machines, people and the community around. How did you experience being part of this network? How did this network and the work manifest? What was happening at servus when you were there?

A: On the one hand, there is this sys admin thing, which is – or at least should be – well-structured work with fixed schedules, and on the other hand we have this artistic work, which is more experimental and is a bit more chaotic. Bringing that together is quite interesting because I’m not the person who structures their day so tightly, so it fits very well – this mixture of a spontaneous, chaotic work situation with structured work as a sys admin. It was always a bit of a translation between the tech and the art world, more like an interface in-between. There are art people and, I mean, they know tech somehow, but they don’t know how we use it. There’s always someone who needs to be this interface or translation. |59

28. Q: When you were at servus, I think you were also in touch with similar projects or communities that were running their own infra- structure as well. What were the characteristics of these people in general? Who were they? Are they still active? What were they doing?

A: I think some people that are still active is Riseup. They worked in a very professional way, but they had a very limited set of services. They set it up in a technically very tight way, so they had their fixed processes. But they worked voluntarily, no one was paid, so I think this is the main difference between servus and a lot of other initiatives in Europe – I met many people over the years,  they all ran their data centers or their free nodes in different locations. They all do it voluntarily, so it makes the projects more volatile.

People changed, so sometimes the projects died, new projects emerged. One very special thing about servus is that we can at least pay a bit, not like a fully blown tech employee, but at least some- thing. The work is therefore more stable, we have been able to run this data center already for almost 30 years. And that is exceptional. There is only one similar project in Austria, which is called mur.at,2 and they do it in a similar way but these are the only two projects I know of that sustained for a long time. |59-60

29. Back then, every server was a physical machine in our server room, so it was the first time that I did experiments with virtualisation back then. It was Xen4 and the machine was called “develop”, because I just tried to develop our infrastructure further. Then some people came and said, “We are trying to make this project,” and I thought about how we could include those in our infrastructure.

You don’t want to let people have root access to your machines. We also didn’t want to have too many machines in our server room because then the space is limited and every physical server needs more electricity. It’s not the best idea, so we came to the decision that maybe they should just get a virtual machine, and we gave them their own networking compartment and so we could have it separated from our infrastructure. This also gave them the most freedom possible. |60

30. Now we have the hyperscale, you can have your Google Compute Cloud, your Amazon Web Services. I think the main point is that servus is not so interesting on the technological side because also all the others use free software, but they are trying to do business and we are trying to make a community, to listen to what people need and to implement that with our knowledge. That’s the main point. It’s not about having technology applied somewhere, it’s always how you apply and govern it. So, this is one problem with the free liberal open-source software. It's always just this technical aspect of it, it doesn't say anything about how the projects are run and governed.

Linus Torvald is the dictator of Linux, he is the one in charge and he decides what is in Linux and what’s not. The recent things about WordPress – it’s another level. Facebook, Amazon, Google and other companies are not thinkable without this free software because it was the software they had at hand and they could adapt it to their needs. So, it’s the foundation for those huge companies now. It’s different if you try to offer a service or develop something within a community. |61

10. LURK#

39. More generally we want to address that there is a certain kind of navety that surrounds specific ideals of community ran network infrastructures about, well,how the rent gets paid and the dishes get done, etc. There's a notion that joy and a sense of purpose will be enough to motivate those doing the labour, but if you ask anyone who has done a lot of volunteer work involving a lot of responsibility, over a long span of time, you risk getting burned out. So we try to make sure that we can pay people sometimes which has the double benefit that we can

1. justify spending the time on bigger meaningful efforts 
2. know that our efforts are valued and supported by our larger community – that does not prevent us to also lament and be aware that taking into account the economic dimension of such work is a painful reminder that the underlying precarity of these practices under capitalism is getting worse and worse.

So, it’s not a career or a big budget operation but it seems to help keeping things a lot more self-sustaining. If someone needs to hibernate for a while or if we need extra help, it’s also nice that, through the support we get from the LURKers, we can invite someone and offer financial compensation for their work, instead of deploying yet another free labour scheme.

This is, of course, still not 100% ideal, our model works because we benefit from the support of hosts, like servus.at, who are giving us free servers. We don’t take this for granted and it’s all a big work-in-progress for the whole network to be fairly compensated, but it’s moving in a good direction. Ten years ago, even within niches like ours, discussing the economic dimension of our practices was a big taboo and there was always a stream of people ready to volunteer for admin burnout. We have to change this culture. |66-67

40. Q: What is your motivation in keeping up the infrastructure?

A: It’s fun. We get to support individuals and groups that we like. We can also use it as a chance to make a small argument for a different kind of IT culture – one that’s a bit more personal, participatory, honest (sometimes things break or go offl ine) and has a (virtual) local flair. It’s a way to make digital spaces where we can do things our own way outside of the more normative commercial regimes we all know.

It's also in part about "doing the actual thing". A lot of people in the field of art, design, and culture production spend an awful lot of time about computers and network culture - negatively or positively - and here we're actually working on th real, material conditions. Both our critique and hopes are deeply rooted in the practice of net and computational culture, and reciprocally, we can express the critique and hopes by experimenting with these machines and software. |67

49. Ultimately, it is not suprising that communities who are working critically on technological matters have shifted their interest from the importance of openness and culture freedom to questions of social justice, labour and ecological matters, as seen for instance in groups discussing permacomputing.

50. Q: Is there a future of artist-run infrastructure?

A: Not only there is a future, but it is also a long history that the cultural sector tends to forget because of its addiction to novelty and complicity in legitimising Big Tech. ... In a general sense, most artists outside the mainstream always have and always will need alternative infrastructures, digital or not, to support their work, and we see this effort as part of that long tradition of off-spaces, independent galleries, indie publishers, studio collectives, equipment pools, etc. that will most definitely continue into the future. This can hopefully help the field of art, design and culture production rethink its capacity to materialise different ways of living and working within and outside of its own field and move away from the same tired recipes of the culture industry.

Lately, a lot of other groups have started to take an interest in running various kinds of infrastructure in varying forms and degrees, to gain a bit of independence, create localised spaces and have more of a say in the form and character of everyday digital life. |67

11. Scatter Chatter: the server as a chatty machine#

from Chae # Kamo

51. One the main focuses f this collective is the idea of combining the notion of a shop with the self-organised structure of a collective that caters togetheness. It is a meeting point where transactions and exchanges of knowledges, skills and resources are circulating. How do we organise this work and how do we manage to keep track of all these different scattered realities that now are living in different parts of the world? I think this is something that really shaped our way of working together. |69

12. The server is out of capacity, we will go elsewhere.

from VO EZN

52. You would think we would use all the technical developments that humans have created to make our lives easier. But no, it's always used against working class people. Always. ... And then comes the union busting practices and all the policies that forbid workers to unionise. All these machines are indeed turned against us, even if we are running the machines, it becomes like a race to keep them up 24/7/

That's why I like this feminist approach to machines, that we're going to take care of it at our own capacity. If it's down, that's fine. We're going to come back to it when we have the time. |74

12. Femfester In-grid#

This femfester is In-grid iteration of the feminist server manifesto. As we rub up against and dance our way into network practices, festering acts as a place to form intentions and imagine how we want to collectively meet here.

- A femfester server ... [is] ferments the materiality of software, hardware and the relations of the bodies gathered around them.
- Does not strive for seamlessness as it creates a bubble. Talk of transparency too often signals that something is being made invisible and inaccessible.
- Avoids efficiency, ease-of-use, scalability, and immediacy because they can be cages we need to abolish.
- Approaches network technology and sociotechnical systems as places of social change.

13.(un)making: the more you avoid production, the more value you create#

from Inari Wishiki

53. My approach is more value-focused and I explore a new kind of value-creation mechanism where the more you avoid production, the more value you create. I'm working on the guidelines for those what I call "(un)made [unmaking approach] objects" and conducting workshops to verify them.

Q: What would an Umaking approach to infrastructure look like?

A: Unmaking approaches to infrastructure are NOT something like building an eco-village from scratch with like-minded people. Rather, situating yourself in an intricately knitted apparatus of modern infrastructure in a way that you might appear opportunistic but cannot be removed. I basically do that all the time, I’m just scavenging for life! |81

14. Halkf the risk of data loss is accident#

from irational.org

54. Q: What do you think is the future of artist-run infrastructure?

A: There is a tendency in this space which is ever increasing concentration. In other words, alternative spaces are becoming scarce and the infrastructure is concentrated, which is a bit ironic, as distributed systems should be kept distributed. We need to collectively think to be kept distributed. Irrational is currently looking for an alternative server farm for our next back up server so that our people can be slightly more resilient. |83

15. From infrastructure for the commons and towards infrastructure for the struggles#

from Mansoux Ayuneric

55. Between 2006 and 2010, we also organised a festival called Make Art.8 It was quite similar to what AMRO would eventually do when growing out of LiWoLi. There were many other little festivals that were popping up in Europe at the time, with a mix of conferences, workshops, performances and a small exhibition. The main difference with the other festivals was that – like for LiWoLi and AMRO – we were very picky about who could get to apply, as we invited only people who were not only using free and open-source software, making art, design and music with it, but also that these people were the ones developing such technologies. Sometimes when selecting the applications we got for the festival public call, that meant looking at the source code of the work and checking if the artist had contributed significantly.

Some people were a bit upset about this, but to contextualise, this was never about being elitist but to instead resist the little EU media art old boy network that had polished a formula in which an artist could have a residency in a medialab, where a bunch of invisibilised ghost programmers would turn a vague concept into something tangible that the artist could then exhibit with an open source label setup on top, and with absolutely zero understanding or appreciation of the underlying technological apparatus. This was particularly ridiculous when the work was addressing critical aspects of computation and net culture. |89-90

56. Twitter and Facebook captured so much of the audience from these critical media practitioners that it simply almost entirely destroyed the infrastructural ambition of this scene and its inner energy necessary for its renewal. I saw a lot of people who used to be and are still quite critical of the tech industry, suddenly falling into the trap of being able to reach a larger audience and getting hooked on it. You could see all these politically engaged people from the community realising: “Oh, I could have an actual audience for all my stuff and I’m going to do that on Facebook.” There’s been a lot of migration happening. So, everything that had been built by these communities shrank violently within two years. |90

57. Always work with others, don’t try to solve problems by yourself, try to integrate the questions of organisation and production in your practice to really address how things are made. Try to move away from this cliché of the professional artist as much as possible. We have to go back almost to the ‘60s culture industry critique to address the question of professional art as being already discussed as a dead end.

this is a very old discussion and one that should be given more visibility in art and design education. Today, professional media art practices are part of mortifying circuits linked to politics of linear growth economics, funding pressure, mass consumerism and extraction, and speculation over social, cultural and economic capitals. Artists, designers, cultural workers and cultural practitioners in general are good at connecting the dots and organising collectively while offering other ways of living that can be more generative, joyful, enriching and critical. How can this be applied beyond the production of a bunch of random stuff spread on a gallery floor?

How can such other ways of living, seeing, being counter the seasonal pattern of cultural programmes favouring the ever growing discourse formed of incomprehensible buzzwords? And to be clear, this is not a challenge that should only be left for the cultural practicioner themselves to solve. Everyone is concerned, especially those having more instrumental positions like cultural organisations or those running funding and policy bodies. |90-91

58. Right now, in the context of permacomputing, there are quite a few projects happening. People recycle hardware to host on old machines that can be solar-powered, and old phones or laptops. It’s something that I’m also doing as a teaching method, and it works. You can transfer practical skills and knowledge and use that to open up about different issues happening in the tech industry. But then you end up with the typical “hello world syndrome”. What do you do then? You’ve got your community server running on a phone. But was it really something that is going to make a change? Who is going to benefit from that? What was the purpose of it?

This is where we see this Venn diagram between artists, designers, cultural workers and hackers, there is something happening in the middle that is powerful. But still, a lot of things feel like unsolicited help or unneeded infrastructure. So maybe making this connection, thinking how the community servers are historically self-serving for these communities so that they can explore art and design practices. What would it mean if they stopped being self-serving? If all these skills were employed to support communities that actually need alternative infrastructure and active servers to be able to communicate and chat together, all this knowledge would not end up in showing a Raspberry Pi at Ars Electronica, but in doing work that has an impact.

59. I get the point. The self-hosting scene should focus less on running the self-run infrastructure but finding more, stronger reasons to exist. Otherwise, we all end up having our own Nextcloud hosted on the fancy Raspberry Pi that we just bought, and everything is great because we feel that we are finally free from big tech.

We are trying to have this discourse about communities and collective practices, and yet, we are still obsessed with hosting things that concern a very tiny audience, not realising that we could join forces much more efficiently. And this would be both from an environmental and social perspective much more interesting and relevant.

So that’s maybe the next step: What are the struggles addressed by such forms of organisation, what are the communities and groups concerned, and how helpful are these approaches compared to other ways of organising and working? |91