Cool. Hello and welcome to the second day of Conversations with Computers. We are live from our studios and bedrooms and welcome to the, yesterday I said the coziest place of the internet today, the most esoteric place of the internet maybe. And if you haven't joined us yesterday, I will briefly go over what we talked about, but first of all I want to go back into the context and why and how we developed this little symposium. I am part of the Servus research lab that happens every second year and Servus is this net culture initiative that is running local servers for 25 years now and yeah you can join and there's the next cloud and we do art on the internet. And this year we got a sponsorship from Linz Kultur and Linz Impulse to do a little summer camp and summer trip where we invited artists to join us in the Austrian mountains. And we did a lot of campfires around our digital projector and we talked about machines and our relationships towards them. And one topic that came up in the camp was esoteric programming languages. And today you will find out what these are and why they're important and interesting. Later on, we will also talk about AI art and see different positions on how AI art is made and what it actually is and how it can be done differently. So again, I want to thank our sponsors, which is Linz Kultur, but also we are working together with the Art University in Linz and the main program, which is Time-Based Media. and the main program, which is time-based media. And thanks again to Christoph Nebel, who's one of the professors there, who's supportive of the research lab since 2008. I want to give my biggest gratitude to Davide Bevilacqua, who is co-organizing all these events together with me and without him it wouldn't be possible. Then I want to give an additional shout out to the Interface Cultures program who helped us with on the technical side and especially Giacomo Piazzi and Antonio Zingaro who are doing all the streaming magic happen in the background that you can see on your screen now. Also there's Fatima El-Kosht who's managing and helping us with the etherpad that you can see down below. You're very invited to share links and anecdotes and your thoughts and questions in the etherpad and i will bring it up later in the conversation um but first we will have a short little presentations and talks from our guests tonight um that i will introduce each of them by themselves but the first guest I'm very happy about because she kind of is the reason why we talk about esoteric languages today is Mariana Marangoni with the title deny usability explore possibility diving into the world of esoteric programming languages and Mariana is a Brazilian artist and researcher currently living in London and her work explores the materiality of media and new forms of narrative in her web-based experiments and sculpture installations she most recently has been working around the internet crisis and the aesthetics of decay in digital media and the potentialities of esoteric programming languages that could create potential arborescent structures, tree structures. She holds an MA in Interaction Design at the London College of Communication and is currently an Associate Lecturer at the University of Arts London and a part of the Supra-System Studio as a Research Assistant. I'm very happy to introduce her and give the voice and her screen to you and I'm very I'm looking forward to hear what she has to say. Please, Mariana. Hi everyone, thanks so much for having me today. Thank you for the lovely introduction as well. So yeah, my talk today is called the Usability and Explore Possibility, Delving into the World of Esoteric programming languages. So unfortunately, I won't have much time to really delve into this world because it's very rich and complex, but I'll do my best to give you a little introduction of this fascinating subject. Let's start by some hello words, because they are the most simple program you can do in any language usually. So it's just a test program. So this is one in JavaScript that a lot of us must have encountered before. And this one in basic. And this one gets a little greater in Kabul, which is a language loved by many people. And you can see why. And there is this one. And it really starts to get interesting. Or this one. And there's actually something here. Like if you highlight, it's a language that just counts the tabs and spaces. So yeah, let's go back a bit and do a little explanation, because those last examples are as links, which is short for esoteric programming languages. And by esoteric, it means something intended or likely to be understood by only a small number of people with a specialized knowledge or interest. But sometimes some people don't like this classification, they think this term is a bit even-capacitist or excludant. So some alternative names could be weird or art languages but a lot of people especially nowadays they're like giving their projects many different names like software art or even a performative engineering art so it really depends on each artist or programmer and their specific work because LL is a length as I said it's very rich and you can do a lot of different projects with it. So programming languages are not immutable nor neutral and as this quote by Matthias, the constructs and logics of languages are themselves contingent, abstractions pulled into being out of the space of computational possibility. So programmers are so used to the norms of computation and just doing stuff every day that some stuff are taken for granted or just we don't realize how biased it is anymore. How entrenched in those systems are like westernized views or even like English-speaking views. So, as a links also challenge that and they don't have a solid definition but they are usually there are some characteristics characteristics they most of them share and they usually not meant for practical uses or streamlined processes, because they're usually not very useful for industry standards. Or sometimes they're so difficult that not everyone can spend the time and effort to learn how to code with them. And sometimes they're purely recreative, and sometimes they defy the established notions of computation. And there's this quote by Patricia Ridge that I really like. It's to treat these always imperfect computational models as hypo-real tools, that is, tools for hypothesis, which I believe this is in a different context, more in the context of machine learning models, but I feel it also applies here. And they may feature one or more of those characteristics. They can be comedic pursuits, like because a lot of asylums are parodies of other languages or just like nonsensical or absurd attempts of creating languages. They can be computational limits defined. They can have a multiplicity of meaning. And or can they can be minimalistic pursuits as commonly called turnpits where quoting as well, everything is possible but nothing is easy. And the puzzle challenges which are also languages that are supposed to be difficult. So languages that are supposed to be difficult. And I'll start by talking about some notorials or famous examples like Intercol, or compiler language with no pronunciable acronym. And like even the name is a joke and I really like this one like it's considered one the first one to be the deliberately weird and I really like how like the comment in this one and how it references science fiction literature and it's even funnier because Don Woods also was the creator of the first interactive fiction adventure. So you see the connection in that. And you really have to say please for the computer to do something for you, everything you do, but you also cannot say please too much, otherwise it also gets annoyed by you. And it's really hard to know like the right amount of politeness that you should have with Intercom. And the next one is Small Bo Bulge and its name was borrowed from the eighth circle of hell in Dante's Inferno. And it was created to be difficult, to be like the most difficult one could be. And it features stuff like a trinary machine model, direct instruction coding and self-mutating code. So you can see why it's so scary. And it took a long time for the first hello world to be written. And it wasn't just a programmer doing it by himself. It was with the aid of AI and cryptanalysis algorithms. And it's interesting now to see the aid of AI and cryptanalysis algorithms. And it's interesting now to see the importance of performing code or having this as a link community, because it was other person who found the hello world. It was released without any program. So it's really interesting. Like just even like the most difficult ones, you find at least one programmer who will spend years of his life or her life trying to find a solution or finding a program and they keep alive even with such difficulty. And another really famous name, I will say a curse word now, is BrainFuck. And it was also one of the most difficult ones and one of the most common Turing tarpids, because he tried to, the creator tried to stripe down all the unnecessary elements, making it the most basic Turing machine. And it looks like this. You only have eight commands and a tiny compiler, which manipulates the memory of machine directly. And it inspired many ESO links because it was created in the 90s, so it had a big legacy and inspired a lot of projects that just changed the syntax. Like Spoon or Ooc, that is orangutan sounds. And Spoon looks binary, but it's not. Like, it's just a repetition. And also a more recent one that's really interesting, which is body fuck, where you can really have to dance to program it. And I'll talk a bit about the class of fun joints, which I find really interesting because there are languages that consist of a group of any dimensional stack based programming languages and they have those instruction pointers with a sense of direction and sometimes even velocity. So you can see in the images that you have like some arrows that will change the direction of the code. And they highlight in different colors for you to see the different stacks and how the code translates into the page creating this two-dimensional structure. And the first one is the-coding languages which have this ambiguous quality to them or they don't have a source code that we are expecting it to be because we as humans can interpret it as something else. So in peat by David Morgan, the pixel, the image is the source code. So it's kind of the opposite of generative art because we are not using code to create art, but the art is the code itself. In Chef, recipes are the source code and sometimes they're edible, like this Hello World cake. And there's also a velato by Daniel Tanking, in which MIDI files are the source code. And yeah, in going a little bit further into this thought of denying the current computational limits, there is this approach by Evan Boswell with non-native languages. And basically it refuses the if and else, the conditional branching, which is the basis of most of our computational languages and even our own system. Like even this question, like what if we are using a conditional? So it's really hard even for us as humans to come to terms with how this could be possible. And it's still a work in progress, and the branching is replaced with self-modifying code. So the entire code changes and not just the variables. And yeah, I'll jump now to how computer compels compliance and go back to the question of the language and the natural language that we use inside our computers because like assembly code or machine code would be the most performant ones but they're really hard and abstract for us humans so then we started to create high level languages and the first ones were in English. And that creates the whole basis of what we have today, and English is very much at its center until now. So those many layers of control dependency make it really hard to refuse the English-dominated environment, from the very hard work to character encodings and APIs. very hard work to character encodings and APIs. There's an interesting history of this with the Wubi method and how computers even threatened the Chinese language in the 70s because they were only asking encoding, that was only able to encode in English characters at first and later Latin characters, but the thousands of Chinese characters couldn't even be encoded with the limited memory of its time. So it created a bit of a panic at the time because they wanted to adopt computation and wanted to have computers. But how would they input Chinese characters into QR code, into QWERTY keyboards? Sorry. And yeah, use the computer in their own language. And then they created the Wubi method that was later replaced by pinyin, which is the most common today. And also about some examples of this challenging of the English, the pervasive English and this digital post-colonialism. There's this concept of ethno-programming by O.T. Leite, which comes from this ethnic minorities of Finland and this wish sometimes to be able to code in our mother language and as someone like me who learned to code before learning properly English it was really hard at first to learn some concepts or even like just after many years I realized that something that just made sense inside the code was actually an English word that made much more sense. Some good examples of this is the work of John Corbin with Cree-S ancestral code about his Cree descendancy. There is Wenyan Lung, which is a language that is in traditional Chinese and also by Renzi Nasser which is completely in Arabic. And something that Nasser says that really stuck with me is that how important it is to perform to really compile it and interpret it in the machine, because only when you're coding and interpreting it that you find the hidden limitations of the system, what breaks inside of it, because to use Arabic or Chinese, you have to use Unicode and not ASCII. So sometimes something's breaking in your OS or in your hardware. And creating an EzoLink that is just something speculative, it's not enough. You really have to build it sometimes to see the limitations. to build it sometimes to see the limitations. And another concept that has ties with as a links because the many of principles they resonate is the notion of queer code. And because in queer code is both the subject and the process of the work, creating what would be considered to be the normative conventions of software and its use. And a big name in this Queer Code studies is Winnie Soon and Geoff Cox. And Winnie Soon created the Vocable Code, which is both an art, design and an activistic project. And it introduced the concept of non-binary logics, where you can try to use even common languages that we have, like JavaScript, more, um, career friendly way, like avoid using binaries or singles X and Ys and be mindful of the variables and names. So you create this ambiguity in the code, uh, that allows for like, yeah, for more interesting meanings and identities and the last thing I would like to cover today is this concept of code of the grid or code what it means to create programming languages for the future when you're facing climate change and yeah and some good examples are the work of 100 Rebits and because they're creating like a lot of small tools, and their own virtual machines that runs their own small programs, they do a lot of stuff like they're illustrating programs, they're 3D art programs, they're games. And their main concept is always to use as little of internet as possible and to use really low tech stuff, computers from 20 years ago, recycling and repair to enable programming through the civilization collapse that seems inevitable now. And there's this other project with his Collapse OS that they also use, which follows the same presets. And it is a future of computation with simple tools in obsoleted machines for specific tasks. And some people may argue that it's not progress, but maybe it's what's going to save us to save the act of code in the future and it may be the only future possible yeah thank you for listening nice super nice Thank you for this amazingly short primer. And I'm pretty sure that there are lots of people who are now interested in knowing more about esoteric coding languages. And fortunately, today, we have like the person who compiled the biggest and best knowledge base for esoteric coding languages here with us which is Daniel Temkin I don't I don't see him on my screen anymore I'm not sure if he's still there. So Daniel, if you're here, just say hi. Otherwise, he will probably be back soon. And I can already give a short introduction on who Daniel is. But I think now he's back. And sorry about that last minute thing. So Daniel will talk about beyond runnability and conceptual and the conceptual and impossible languages. So and Daniel makes images, programming languages and interactive pieces that use the machine as a place of confrontation between logic and human irrationality. His blog and research project esoteric.codes that I already mentioned documents work engaging with the text of code including experimental coding languages, code art and other systems that break from the norms of computing. He was the recipient of the Arts Writers Org grant and further developed at the New Museum New Inc. tech incubator in 2018 and exhibited esoteric codes at ZKM in Karlsruhe in 2018 and 19 and he has published in Leonardo and World Picture Journal and presented at conferences such as South by Southwest, Glitch and SIGGRAPH. Yeah I'm really really happy that he's here I hope he's here. I'm back. Okay great that Okay, great. That was a funny coincidence that happened just before your talk. And yeah, if, is your video working also? Because I cannot see your webcam yet. But otherwise, you can also join us by voice because we, ah you are great um so yeah uh happy to give uh the stage or the your the display the screen to uh daniel and looking forward to hear about conceptual and possible languages hi so i'm going to be talking about conceptual languages, which is really my term for the languages, for the SLNs that run in our heads rather than on our machines. made this argument for SLANGs as a form of experiential art, where a language is designed by an SLANGer and we experience it kind of primarily by trying to write code within it or looking at code that others have written for it. The conceptual languages are the ones that we cannot experience that way for one reason or another. But it's kind of interesting actually looking back at esoteric.codes. The first interview that I did, the very first piece for what would become esoteric.codes, was an interview with Keymaker. And at the time, I used this term conceptual, asking about his language unnecessary. I describe it as having no working compiler at all, which actually is not accurate. It's pretty trivial to make a compiler for unnecessary. I don't really explain the term conceptual to him, but he understood what I was talking about and gave a reasonable answer, and we had a discussion about this language. But basically, the language unnecessary is one that considers all code to be unnecessary. There's only one program that you can write in unnecessary, and it's the program that doesn't exist. So if you run the unnecessary interpreter and you point it to a program file, it will fail with the error that your program is too verbose, and it'll reject it. But if you run it and you give it a path to something on disk that doesn't actually exist, the lack of a program file, that succeeds and compiles, it will run as an unnecessary program. And that program is the program that prints its own source code to the screen called the Quine. Of course, there is no code because there is no file. So it prints nothing. And this is what's known as the null Quine. Quine actually comes out a bit in SLAs. It doesn't really come up very often outside of that category. Quine, by the way, it's named for the analytic philosopher. It was named by Douglas Hofstadter and Gerda Lascher-Bach, and it's the program that prints its own source code to the screen, which often is very complicated in many languages. So Mariana gave this description of SLLangs as creating this space of computational possibility and used this example of Malbolgia, which is a great example because Malbolgia is a language which Ben Olmstead created in a single afternoon. And as Mariana explained it, it was really up to all these other contributors who really explored the language and really sort of created Malabosha as a cultural object, who spent many afternoons discovering what could actually be done with this language and how it actually works. And interestingly, this is summed up on the SLNC's wiki as weaknesses in the language. You know, people were able to beat the language by being able to actually write working programs in it. So where Olmsted sort of set up that field of possibilities, it was all these other contributors, this kind of community of programmers who really created Malbolge as it is. And because of that, Malbolge has become somewhat more of a cultural object. It's had appearances on the TV show Elementary where a killer leaves a clue as a Malbolgia program for some reason. And this very odd product placement, you know, KFC's Colonel Sanders had an appearance on General Hospital where he explains protecting his secret recipe with Malbolgia and having it cracked by, you know, by these hackers. He pronounces it Mal-bo-gal for some reason. Anyway, so partly this is, you know, in part it's probably because brainfuck is not something that people are going to say on network TV, but, you know, these two languages, brainfuck and Malbolgia, have become, you know, are seen as as these kind of hacker languages that show people have this sort of expertise to be able to code in them and so on. But there's one way we can actually think of Malbolgia as a conceptual language, to go back to our subject, and that's in the fact that there isn't really one language that's Malbolgia. It's when Olmsted created the language, he made an error. And the language definition has a different mapping of commands somewhat. It's actually this encryption table to the right here. The code is encrypting itself as it runs. And the encryption table is described differently in his example compiler than it is from the language definition, which basically created a schism in the community between people who thought the bug was in one place or the other. And it's something where he said if he ever made his long-promised follow-up of Malbolsha 2000, he would definitely make the documentation subtly wrong. And it's, you know, at this moment when we have, you know, NFTs kind of encouraging digital artists to make these singular digital objects that can hold financial value and so on, it's interesting to sort of push back against that, back to the idea of digital practice as something that is more of a field of possibilities and doesn't have the singular object that defines it. And this is something which Mel Bolsha does by not being entirely contained within the language definition or the compiler itself. And that just reflects back on the fact that there's not really just the single author either, it's all the people who contributed to the language that sort of helped make it what it is. So back to Unnecessary. Unnecessary is particularly appealing to me because it gets rid of the program. It gets rid of code entirely, right? And there's something about that sort of dematerializing of code that sort of points to, that points to this lack of the singular digital object. And it's something that we can think of in terms, rather than think of it as a constraint-based art, we can look at the Fluxus event score as sort of a model, something which is actually sort of instructions for performance, but it's not a performance we're likely to actually carry out. Some event scores were carried out as performances, but many we would read as a piece of poetry. This is a piece by Yoko Ono. It's something that we can experience by imagining this. We don't actually have to go cut up a painting up and let it be lost in the wind to experience this piece. But it's also, in terms of unnecessary in removing code entirely, it's also sort of like the emptiest form of a programming language. And so it also points to another type of flexes work, the ones inspired by John Cage, such as A Zen for Film by Nam June Paik, which is a film that's made up entirely of a film leader. medium that it ordinarily is, and the dust and scratches, which ordinarily we would ignore or we would find intruding into the message, the actual content of the piece becomes the only content. It becomes the thing that we're actually looking at. This is something which has been written about by Nick Montfort, who did the study of null programs, programs that don't have any code, and the null line. An empty file is a valid Python program. If you run an empty file, it'll accept it as code, and it'll run it and do nothing. And of course, all sorts of things will happen. It'll load all sorts of things. It'll start processes that will happen when it's done. Whether the empty file is a valid program in C is up to some debate because it actually depends on which compiler you use. I think GCC accepts it with a whole bunch of warnings, but there's this the obfuscated C contest, the international obfuscated C contest, I'm not getting the name quite right, which one year accepted the empty C program as being a valid program so that no one would ever be able to enter it again, because it has already won. That was their sort of way of dealing with it. And as Montfort shows here on the right, even an empty binary file creates some interaction on the machine. It generates this warning that the binary is not a valid Win32 application. If that's what you want to do, then it's a successful program. But it turns out Unnecessary is not the only way you can make sort of an empty programming language, because there is the opposite of Unnecess unnecessary, which is a language called Callisti. Callisti is this Discordian programming language. You can think of Discordianism as Dada for nerds. And it was created by the Prophet Wizard of the Crayon Cake and the Seven-Inch Bread in 2006. And it has this whole sort of philosophy with all these rules, including the rule of obey as many rules all these rules, including the rule of obey as many rules as possible, and also the rule of obey as few rules as possible. And we have it more formally defined in EBNF notation here as anything could be input, and then anything comes back out. So where unnecessary rejects all code as being too verbose, Kallisti accepts all code and just doesn't do anything with it, there's no way you can make rules about how to behave with this code when it doesn't distinguish at all between different types of input. So you can run your vacation photos, your CV, or any sort of digital object through it, and it'll come back out the other side unchanged. And it's really impossible to say whether you've actually run it as a Callisti program or not. So some people might reject these as how are these programming languages at all if you can't actually write any programs in them, but it turns out it's very hard to define exactly what a programming language is. And one thing that many people point to is is tern completeness. There are first of all many programming languages that are that are of less computational complexity than Turing complete, but Turing complete is the category that means it can basically run any algorithm that can be run on a conventional computer, and it's the class that both BrainFuck and C++ belong to. So I think we can sort of agree that languages that are Turing-complete are valid languages. But it turns out that Turing-completeness happens all the time by accident. Programmers are always creating these accidental programming languages, things like C++ templates, which really don't want to have Turing-complete. Just the move instruction, I think at the bottom, the move instruction in Intel is enough to, is itself turn complete, which it really shouldn't be. And it's been discovered that human heart tissue is turn complete under certain conditions, which was, I guess, an important medical discovery because it means that it has, it means we can't predict how it's going to behave, but it's not something that we would try to build a computer out of, probably. JSFuck also, this subset of JavaScript, which nobody invented, but was created by accident and discovered by a number of people and then formalized by Martin Kleppy, that you can write all of JavaScript with just punctuation. This was not something that was created intentionally in any way. It was just formalized by Kleppy, who brought all these discoveries together and formalized it. So back to the conceptual languages. 2014 was a programming language that only worked for about two and a half hours at the end of 2014, because the current year has to be 2014 for it to run. So on the other hand, we have the language language. P2, also in 2014, asked, do we really need the binary? Why can't we write code with just zeros or just ones? He picked one, and this is a turn-complete language. You can write any code in this language that you could write in C or any other language. The only problem is it takes a lot of ones to do anything useful. And if we consider each atom to be a one, it takes more matter than exists in the universe to be able to write hello world to the screen. So it's not a very practical language, but it's well defined and perfectly valid as a language. Chris Pressy created this language called You Are Reading the Name of This Usseling. It's supposed to be pronounced You Are Hearing the Name of This Usseling. Pressy created BeeFunge, which Mariana discussed a bit. Pressy created Bfung, which Mariana discussed a bit. He's very important SLN, or it was really on his mailing list that the term SLN was first used, and he really kind of fostered a lot of the early discussions of SLNs. But basically what he's done here is he's using the language spoon, which is that that binarization of brain fuck that we looked at briefly in the last presentation. And basically what happens is anything in brackets, so this is a language that has zeros, ones, and brackets. Anything within the brackets we run as a spoon program. And if it finishes, then we get a one. If it runs forever, then we get a one. If it runs forever, then we get a zero. So what happens is, you know, we can detect trivial loops. We know that some things are gonna be zeros, but there's no, because of the halting problem, we know that there's no way to write the perfect interpreter for this language because there's no way for us to predict whether certain things are gonna run forever or not. So he's moved this issue of the halting problem from the running of code into the interpretation of code itself. So we can't even really know what kind of program we're dealing with without running into this sort of impossibility. But it turns out that David Medour has solved the halting problem for us. He's created a series of languages that are of a higher class of computability than term completeness. Mador is sort of better known in SLLang for creating UnLambda, which is a language based on combinator calculus, which is a theoretical kind of computing that he turned into an actual programming language we can code in. It's very challenging. It can be kind of fun to code it. It's much easier to code in to actually read. It's impossible to read anyone's code that anyone's written on Lambda. So he takes these... He's a mathematician, and he takes these ideas from computation, actually turns them into real languages. So the transfinite languages, he has worked on the arithmetic machines, which were, I guess, from the 50s through the 70s. And these are languages that can cross between finite and infinite numbers. They can literally count to infinity, and they solve the halting problem, which is supposed to be impossible. The only problem with his languages is that they require an infinitely large computer to run on. So we can't run them in this universe. But again, just like language, they are perfectly well-defined languages. And there's a link to my interview with him there. There's also, if you go there, there's a link to his original with him there. There's also, if you go there, there's a link to his original piece, but it's in French for people who want to dig deeper into his work. So I don't make a lot of these conceptual languages, but I've made a few. I'm going to go back to Fluxus again for one more event score. George Machunas and George Brecht were Fluxus artists who were very, so Machunas was kind of the leader of Fox's was, you know, wanted to sort of remove the artifice from art and sort of blur the lines between art and life in a certain way. And George Brecht was really interested in this as well and he would create these performances that people would perform by accident, like the three telephone events. When a telephone rings, you allow it to continue ringing. When telephone rings, the receiver is lifted, then replaced, or when the telephone rings, it is answered. Now this is something that we could perhaps perform on stage for an audience if we wanted, but it's also something that once you're aware of it and you hear telephones ringing, you can recognize what people are doing as being part of this performance, whether the performer is in on the joke or not. So I became, you know, if we can blur our computation this way, too. So I decided to make a few programming languages that nobody would actually code in intentionally. And the first one is this language, which is called a programming language is a formal language comprising a set of strings that produce various kinds of machine code output. It was formerly known as a programming language is a formal language which comprises a set of instructions used to use various kinds of output. You can see it's slightly different than the current one. So basically, this language always corresponds to the first sentence of the Wikipedia entry for programming language. And it really serves to counter that idea. We run into these problems when we talk about programming, where we sort of have to define computer code and language in terms of each other. And it's actually, you know, traditionally we often think of computer as the thing in our lap or in our hand or on our desk. But if we define computer as the thing that runs code, it can be all sorts of other things. It can be people. It can be natural systems. So basically this is a language which is also a Quine-only language. If you write the sentence out, it'll run within this language and everything else will be stripped out, leaving just the sentence itself. It is a language that there have been hundreds of thousands of programs written in. So for example, if you read Microsoft Academic and you want to know what a programming language is, some lazy person took the sentence from Wikipedia and you run it through my language as code, and this is what you're left with. Everything else gets stripped away, and we can just sit and sort of contemplate the sentence that's left behind. So I know there's been a lot of fluxes in this discussion. I mentioned a lot in esoteric.codes the point that Job van der Zwaan made this little piece of fan art about esoteric.codes of this little intersection of these two things about fluxes and these aspects of working with compilers such as Lexing. And that overlap is an area that I'm very interested in and that we dive deeply into on strc.codes again and again. and I hope lots of applause in the chat. Thank you a lot Daniel on these very thought-provoking in a literal sense examples. Next up, I want to introduce Sarah Grof-Henning Palermo with a presentation on No More Smooth Art. And Sarah is an artist, a programmer, and an erstwhile data designer. Her practice centers around methods to make computers and data more humane, more accessible, more flexible and more contextual. Her works included include data obscured art sites, new computer languages and hybrid nostalgia machines. She's an alumna of the School for Poetic Computation, Recurse Center, Brown University and NYU Tandon. So let me, Sarah's already the presenter, so Sarah please share your screen with us so we can see what you're up to and what you will talk about and what this call for No More Smooth Art is actually about. All right, hi everyone, I'm gonna share my screen. I can't see any of you anymore so if something goes terribly wrong you have to use your words to tell me. Anyway, so hi everybody. This talk is going to diverge a little bit from the previous two which were I think really nice looks at all kinds of esoteric art. What I'm going to be talking about is sort of my personal philosophy and general call to arms, no more smooth art. So we can go ahead and jump right in. You might be asking, what is smooth art? And it's basically what you think of when you think of inventive contemporary digital art, let's say. Right. So if it was created by a Google A.I AI engineer, it's definitely smooth art. If it's all about exploring latent spaces, smooth art. If you're overlaying photo datasets, right, to identify that underlying similarity like the Santa photo, smooth art. And if you're otherwise illustrating datasets, especially very limited old-timey ones like portraits of rich people and maybe you're involving the people walking by, that's definitely some smooth art. A lot of shader art is smooth art in the sense that it's low friction, high processor work. If your generative art is hitting 60 fps, like that's definitely some smooth art. Another way to say it is that smooth art is kind of the intersection of these two principles. These are two photos I hold close to my heart every time I'm making work. Smooth art is art nestled tightly into our current cultural belief in technological teleology, right? The idea that we're always creating faster, smarter machines, and these machines are more objective and therefore better than us. And of course, these beliefs imply a constellation of other beliefs, that the world itself is a machine or like a machine, that abstract numbers and statistics can therefore accurately represent the world, that data hides secret truths inside, which can only be unearthed through volumes of information that human minds cannot handle, that high fidelity is better than low fidelity, and so on. And it's sort of the structure of knowing that makes it easier for us to imagine a computer with complex intelligence than a fish that has an inner life, right? Which also means there's smooth art that doesn't necessarily look smooth. Instead, it might be art that involves things like collaborating with AI to make music or plays or making friends with robots. And you might be thinking now, wow, Sarah, that's a whole lot of art to be against. And that's true. And what's so bad about it, you might ask, you know, other than like capitalism in general. And so the answer to that is that smooth art is problematic because it's art that functions within the process of naturalizing our current and frankly oppressive surveillance capitalist social relations, which are arguably worse than original capitalism. And so it's a social relationship characterized by an extractive and unbalanced relationship with the earth. Lately, we've all been talking about the environmental issues with NFTs, right? Burning real energy to create imaginary value. And these same criticisms sort of can be applied to the foundational operations of AI and ML, training and running models. Those are also burning untold amounts of energy for not necessarily great reasons. And then, of course, there's where the data for these models comes from, and that is surveillance. And so there are a number of data sets with dubious provenance, right? Scraping people's flickers, using a webcam in a San Francisco coffee shop without informing customers to work around the corner from that coffee shop, and data sets with limited representations. But the problem with surveillance isn't actually its imperfection, but its existence at all, right? And as long as we believe that AI and ML are useful, we're committing ourselves to closer and closer observation. That's what it runs on. And the way this combines with original capitalism to make our world terrible is observable in our lived environment, right? So right now I live in Berlin. I lived in New York for a very long time before that. And I always think of the development of the LinkNYC kiosks and ad screens as like a symbol of what this social system looks like on the streets of the city. The kiosks are owned by Alphabet, which is Google's parent company. They provide free Wi-Fi so that they can observe all your traffic. And they have three constantly recording cameras plus myriad other data gathering sensors. The main man behind the work, David Doktoroff, who used to work for the city of New York, has stated that his goal is replicating digital advertising in real life, which is kind of psychopathic honestly, and his nightmare appears to already be coming true. In a 30-minute trip that I took multiple times each week between my apartment and the swimming pool I trained in, I passed 26 screens and kiosks over 30 minutes, 80 if you count the ones the subway train rolled by at every stop, but just on my side of the track. Each of these, hoovering up data and spinning it back, is brightly lit moving ads. But worst of all, excuse me, but worst of all, worse than harming the planet, worse than surveilling and bombarding our fellow citizens, is the ideological reinforcement of what I call the hegemony of the probable. And the hegemony of the probable is what underlies basically all algorithmic decision making. And it's the proposal that decisions should be made on what is statistically most likely, which again is how our enemies AI and ML work. And in this situation, if you're an outlier, then that's just too bad. And so we saw this function with A-level tests in England when the pandemic began. With students unable to take their exams, the education system used an algorithm to assign grades based on the previous performances from a student's school, right? Which means that if you were a very good student from a very bad school who had worked very hard, it doesn't matter. You don't matter in the face of the probable. It's just too bad. And so as a mixed race woman, as a scholarship kid from a poor family who managed to go to fancy schools and fancy summer programs, as someone who rarely fits comfortably into a single box, I take this really personally, this idea that intelligence is to be found in the midst of sweeping generalizations, right? And when we make art that implies or valorizes hidden truths in data or in machines, we're further supporting this ideology. We are supporting a world where decisions are becoming fully automated in terms of civic governments, right? Where big, big problems like parole eligibility, school assignments, child protection investigations, and public benefits eligibility are all being determined by this hegemony of the probable, right? And that's like really depressing, you might be saying. So what can we do instead? My proposal is an aesthetic and an approach that I like to call fuzzy art. And the goal of fuzzy art is nothing more or less than undermining computing. And so an important thing to note when we say undermining computing is that undermining is not opposing. So Janet Murray, in an essay that I feel like everyone reads when they start in a media grad school, which everyone goes to new media grad school, obviously. Everyone reads when they start in a media grad school, which everyone goes to new media grad school, obviously. But Murray points out that much of the humanist response to computing has been centered around opposing and saying, no, bad, stop. And that can be a useful and necessary thing to do, but it's not enough, right? Resisting often implies taking computation stories about itself at face value. And so in opposing, we kind of agree that like, yeah, this is what machines are for, and reduce agree that like, yeah, this is what machines are for, and reduce ourselves to arguing, well, is that good or bad? And this becomes dangerous because we've ceded capability, possibility that machines could be for better things. It becomes natural, which is to say beyond dispute. So then we can dispute the premise. And that is what undermining computing means, right? Demystifying it, dethroning it, making computing's constituent parts mean something else. It's facing the promises of intelligent computing and unraveling them. And it's also exemplifying alternatives, which is what we've seen in a lot of the esoteric coding languages people were talking about before. So there's the two parts of undermining computing. It's contesting conceptually and also building new dreams for machines with people. And so one kind of dream might be to ask what it would be like to relate to computers as plants, which is to say as foreign systems that we can observe and cultivate and create relationships with that exist outside of a dominated determinism. And so this is the kind of relationship that I found with computers with live code, which was sort of my first step into fuzzy art. So what is live code? Fundamentally, it's performing music, visuals, dance, games, etc., with code that's written and evaluated in real time so it's a communal computing experience it's also focused particularly on intervening in live ongoing processes while people are watching so live code practitioners comprise a loosely organized global community with groups of artists across the states in Mexico here in Europe in the UK as well as in South America and in each scene although the details of performances might be different, the general live code approach as laid out in the top-lap draft manifesto is the same. Shows and programs are open, screens are shown, and the performances are dedicated to making computing comprehensible and sort of revealing its changeable nature. And the motion here is away from definition and constraint and towards exposure. The machine is not set up as independent or more human than human, rather the operator is foregrounded and the machine is inverted to operate on itself. It's used as basically as a tool and yet life coding isn't about tools. And when we say that, it's resituates the actions of computing in the human mind right subject to its biases and failures rather than a conduit for escaping our immediate imperfections and live coding that ways can be seen as a subset of all kinds of algorithmic and generative music and art one that encompasses all of the academic and artistic investigations for all sorts of reasons algorithm then is a subset of that where where the live coding takes place in dance music venues, clubs, festivals, bars, etc. And it's at these sorts of shows that my collective, Cody, usually performs, although in more orthodox settings, we're probably more in the chill-out or experimental room. I was going to share a clip with you with sound, but it didn't quite work out, so at some point, if you want to look at a full show, you can check out my website about Kodi section. And in the meantime, we can see here a speeded up clip that shows how a Kodi performance might look as it develops over time. These are clips from like a 45 minute performance. And rather than using machine driven audio reactivity like OSC, Kodi aligns roughly on a rhythm and from there the music and visuals are linked solely through the power of the audience's minds. And this method allows us to create an interplay between visual and sound that's more general than approaches in which the sounds are part of the visuals by attaching them to a specific parameter, right, color or some form of volume. Our specific rhythm is such that the frame counter is every 250 milliseconds, which is an eighth note at 120 bpm. It works perfectly. And this looseness comes in handy in other imperfect situations. It means we can perform remotely under poor conditions because we don't rely on low latency or fast Wi-Fi connections. Something, though, that I would modify about the TopLab Draft Manifesto is the notion that life code isn't about tools. And I think that intent is to focus away from machines into the performance. But for me, working on the framework I use to create the visuals, called the Habra, it's an important part of the work itself. In creating it, I open up another part of my practice to undermining computing. And so a system that supports coding live has different requirements from a system devoted to facilitating businesses, right, which is what I do to make money every day. In particular, a live code setting needs to be fully resilient, handling errors gracefully without pooping out. And so for La Habra, I rely on the closure script ecosystem to help me do that. But the sense of give and take between the system and myself, like really grows each time a mistake leads to a surprise instead of a crash, right? So this nice little shake is a result of actually an animation error, but I loved it. And then I made even more errors, which caused the system to fail, as you can see on the right, and changed this shake into the rotation kind of working and everything else breaking. And I find it difficult to work with systems that deal with unexpected content in these really charming ways and not ask why we can't do that with every system that we're using and how that really changes how we understand computing. And so this critique of language theories, right, like many other language based ones, is definitely only available to people with a somewhat deeper understanding and investment in computing. And in some ways, that's great, right? Those are the people that I want to speak to about undermining computing. But I also want to speak to people who might know less about how computers work. And so to that end for the framework, I focused on making the code as legible as I could make it. So it's very non-esoteric. That mostly involves making shortcuts so that it is reasonably verbose but I don't have to type every single word of it and I think we can see here you hopefully viewers can look at the code and say yes this is making a circle this is making a blinking rectangle I vaguely understand and if it's very complex you can look at the style lines and sort of understand how the code is modifying what's gone before and so all together how the code is modifying what's gone before. And so, altogether then, this system is able to create a variety of visuals. But beyond being good accompaniments to music and, to some extent, a nostalgia indulgent, these works also reflect and engage what I think of as the indicative properties of this moment and of computing, right? There's the repetition, so characteristic of the base blocks of computing, loops and variation, and the finest medium of the web, the gif, right? Because humor and charm are directly related to its inability to stop. There's the sense of flatness, accepting the material nature as it is, rather than burning cycles and oil, yearning after a notion of realism that I would say obscures rather than revealing. And most important is the sense of accretion, an accumulation of seemingly unrelated movements collecting into a bigger system. It's generative without being didactic or teleological, without having a goal, and the links to be found here are in small bits aligned just right. And these linkages I think are characteristic of the exhilaration of computing, which however much I have to say about undermining it is also very real. An important note here too is that the accumulation isn't reductive, it's not an averaging, it's a pile of individual pieces that are greater for their interaction with one another. Sometimes I even manage to accumulate too much and hit the limits of an SVG, which I also really love, right? The system has its limits and reaching its breakdown bears another part of it is a tangible piece of the world. So while I do still perform at Algoraves, this intersection of digital accretion and testing the machine's edges was pushed into new territory, well, about two years ago now, right before the pandemic began, when I had the chance of the residency at Alfred University to play with synths from the 70s and 90s. And so I have a principle of funny art, fuzzy art. Well, there's several, and I'll talk about them a little bit more at the end. But this one proposes that the most important focus should be on the range rather than the binary, right? These two little dots between the zero and the one rather than the or of the two lines. And this is a way to think about pushing against those impossible and oppressive categories that undermine that hegemony of the possible. And one way to do that is by using video tools. And video machines are really natural tools for this investigation because video is technically a stream of values between zero and one, rather than the on-off bits of digital computing, which means that the circuits and our interactions with data are different. representation is one in which each pixel is addressed at once per frame, all at once, unrelated to its neighbor, whereas a video image is created by an ongoing stream of values drawing over themselves. So then we can say, well, what's possible doing work within the range? So when I first started putting some CODI and MAHAGRA work into the synths, mostly a Fairlight CVI and a Jones frame buffer, the first observable outcome is that soft analog glow. There are also a number of affordances to make different kinds of representations easy compared to the SVG framework, mostly in terms of masking, mirroring, and repetition. And so this is still digital art fundamentally, the input material and these final representations are digital, but it's made some round trips into different systems. And that means we can move beyond accumulation on a frame by frame basis to accumulation on a material basis where each pass imprints the accumulated materials with the given system's character, right? And this is sort of facilitated further by continuing to use the improvisational tactics of live code when using the synths, tuning and developing in various improvisational passes, and avoiding over planning at the start or too many edits in post. And this is a slightly sped up take here so you can see how those longer takes evolve. Altogether, what is easy to explore and do in longer takes is that thing that is characteristic of the encounter between the source and the synth. Once I left the residency and no longer had access to the analog synths, I turned to machine-aided replicas, most particularly a frame buffer emulator, a small shifting program I wrote, and VDMX, which replicates a lot of synth capabilities with shaders, even going along with the live code themes by using ISF, the interactive shader format, so that even the shaders themselves are inspectable and changeable and shareable. And so using multiple passes over residency footage and pure digital input, I ended up creating some short films. And there's no time to really appreciate the full ones here today, but if you're interested, I recommend looking at my site, but I do have some clips to talk about how that works. The first piece was serrated, and it starts with this very basic, codey footage. And then this footage is already substantially transformed by the Fairlight. The synth makes compression and repetition and shifting very simple. The interface, right, all sliders and knobs makes it possible to adjust the speed smoothly so I get that live code feeling within using the synth. There's also some nice affordances for freezing and smearing that we can see that open up new portions. Then the footage was put to the frame buffer emulator, which, given the masking capability, has made possible some really radical changes. This doesn't necessarily look like where it started, even though it is the same base material and it allows for some subtler depth in other segments. January 2nd, 2021 is from earlier this year, and that again begins with some La Haga based footage. And for this one, I went ahead and put it through that self-built program that slices and shifts the images. It's inspired by the Fairlight, but much less flexible. And then into VDMX, this picks up some video grain, including those horizontal smearing switcher, basically video glitch effects. It's also transformed into polar space. These kinds of transformations again are much simpler with video analog approach to signal manipulation. Likewise, the image is allowed to be marked in a different shape. Both of them are put together and then finally put into the frame buffer where they get this really chunky color piece going on behind it. And again, they are able to accumulate traces of each pass through a performance. And so these explorations and my search for the machine edge with new tools blossomed over the pandemic into experiments that defeated the compression algorithms on Twitter and Vimeo entirely, which I thought was really excellent and made the work impossible to share. I also succeeded in applying effects until my machines were so slowed down it was difficult, if not impossible, to record the works. And so it turns out the only way to see some of these performances is to be in the same room as I am. And so computer art then becomes back to being about a specific encounter, not through artificial scarcity, but through its own nature and the nature of machines. And so through these explorations, I've developed what I think of as the principles of computer critical or fuzzy computer art. The first is mentioned before is a focus on the range over the binary. The second is to make history and character visible in my case through the accretion, but also through asking questions about what happens if we fork older tech from the 70s and 90s and maybe come up with alternate histories for computing. After all, quests for eternal novelty only suit capitalists. Fuzzy art is never perfect. Ideas of perfect knowledge, perfect legibility, perfect understanding, and perfect reproducibility are what animate the surveillance machine. Fuzzy art is focused on mind-to-mind, over-the-air communication. As with live code, where making things fun and transparent is a large share of the point. And finally, it lets the computer forget. Improvising means not saving and analyzing everything for a perfect recollection. And so I think of this as a way of using computing as the oppressionists use the technological advance of tube paints as an element in the ongoing existential questions of what it means an element in the ongoing existential questions of what it means to be in and observe the world, rather than as a means of legitimating our own subjugation. For the future, I hope to continue doing work like this that lets us use laptops and computing in the same way. I hope this gave you guys some ideas for your own paths to undermine computing every day and in the meantime I exist on the internet those are both twitter and instagram ads and otherwise yeah and the drawings in this were made by Emily with curls so thanks so much everyone yeah nice um super nice and very inspiring I'd like to go deeper into fuzzy and smooth art in a bit. And as people on their screens might have noticed is that we are kind of shifting from esoteric languages towards like these new machine learning models and how they change programming but also how they change art making with computers and I'd like to introduce for that our next speaker who is Max Valentich. His title is Expressing the Generic After Post-Internet Art. And Max has a bit of a different background. He's finishing his master's degree in communication science at the Faculty of Social Sciences in Ljubljana in Slovenia. And he, I think, is now already or will soon be a research fellow at the New Center for Research and Practice. the new center for research and practice he's a managing editor of shoom which is a journal for art and theory fiction and the co-founder of the fast right block his main theoretical interest revolved around accelerationism neo-rationalism and non-standard philosophy and so i like to give my voice out to max and i'm looking forward to hear about his examples on machine generated art. Yes, thank you. Thank you, Matthias, for introduction and for the team for making this possible. So I think that my short talk will fit nicely with previous presentations, so I'm looking forward to the next 20 minutes. And in front of you, you can see a flat gaze Instagram account by Slovenian artist Matei Mihevc. In the following talk, I will try to make a full circle and in the next 20 minutes explain what is going on in the pictures and how can we situate Mihevc's project within contemporary art and most importantly what comes after the post-internet art. I'm sure most of you are familiar with the post-internet art. I'm sure most of you are familiar with the post-internet label and what changes in production, circulation, and consumption it brought with itself. Nevertheless, let me outline the most important shifts in this domain. With the advent of internet, net art arrived, and with it the hacker mentality and open source spirit of the 90s. But whereas net art was mostly focusing on the digital realm new actors in the second half of the 2000s post internet label is coined in 2006 and established as relevance in 2009 developed a more sustained and all-encompassing challenge to the art establishment their main point was that it's not solely the fact that we are entering the new information ontology and that art can now be exhibited and consumed online, but the distinction between the virtual and the physical is becoming increasingly obsolete. The true challenge from post-internet art is related to the fact that the role of physical spaces has changed. Galleries and other institutional places are no longer primary venues for art consumption and curation and have become mostly used as a photo documentation setting for art pieces whose liveliness will only emerge online in the interactive world of social media. Post-internet label has therefore extensively rearranged how the art world functions and in what way artists need to adapt to these new conditions and challenges. Curatorial blogs, Instagram influencers and other internet figures have not only accelerated the demise of the already problematic distinction between the physical or sometimes the real and the virtual but also transform the axiological dimension of art curation but without a doubt this doesn't give us the whole picture as i was mentioning before the post internet label highlighted an important shift hybridization between art production and the new mechanic mechanisms of circulation and consumption production and the new mechanisms of circulation and consumption. In the new landscape, art doesn't have the privileged epistemological position or any kind of outside to the radical flatness of expression that equalizes all data and leads to a radically democratic relationship between objects. More on that later. Some theorists have commemorated this shift by saying that art production has to become more clickbait in order to have any chance of surviving in the new algorithmic spaces. Because of this new awareness on the side of the artists and curatorial blogs, artists went even further down the line and embraced artistic expressions that are fully immanent to the new reality of social media, to the ways the system perpetuates itself through feedback dynamics, memes, viral content, and the new ways of communication that are primarily effective and non-linguistic and therefore without additional subjective reflection. It searched for philosophy that was fully embedded in the aforementioned shift, a philosophy that isn't afraid of thinking with and not simply about these new shifts. In this way, the artists' embrace of accelerationism and object-oriented ontology is an absolutely sensical response to the new cybernetic reality they were facing. Before we continue with accelerationism and object-oriented ontology, I want to show you a few pictures from Balenciaga and Swetnik journal-like aesthetics, so that you have a better grip on what kind of new or newest art are we talking about. So let me say a few words about accelerationism and consequently art conscious of the acceleration. Accelerationism has always been fully immersed in the new cybernetic landscape. What it took from the 90s CCRU theorizing of cyber culture is the emphasis on positive cybernetics. Feedback loops, self-fulfilling prophecies, and especially the concept of hypersition, fictions making themselves real, is especially important for the new kind of art production. The awareness that there's always the ability to hijack and decode the information currents and subordinate from within. To, for example, infiltrate visual culture with different kinds of aesthetic and libidinal expressions that can circumvent the normal or normie authorities from the future through the intensification of the feedback dynamics and memetic contagions that can be controlled, anticipated or accessed from any external position or the position of the present. To come into contact with the future is to be transformed by it. Accelerationism is therefore understood as self-propagation, as the ability for new kind of artistic expression to gain autonomy and self-determination in an ever more flat and parallel systems with no autorial oversight or direction of optimal or desired use or outcome. It's all about reality testing. Accelerationism can be succinctly understood as a flight from oversight and an ever more radical embrace of the libidinal culture where the imminent dynamics of the system produce legitimacy from within. culture, where the imminent dynamics of the system produce legitimacy from within. In a similar way, object-oriented ontology, in my non-standard interpretation, is used as a radical flattening of reality. Within the history of contemporary art, we can say that it in a way continues the trajectory of Duchamp's ready-mades, Warhol's obsession with pop culture, William Flusser's technical images and apparatus, and more generally, Baudrillard's appeal for acceleration of the art object in the age of full commodification. In a new democratized ontology, objects are radically incomplete, internally unstable and absolutely interchangeable, and this enables artists to unlearn conceptual schemes and conventions and start with artistic practices anew. In order to produce new kind of objects and assemblages of objects, artists use this general instability and non-random nature of the new constellations within the new ontology and start producing different kind of knowledge in the process. Contradictory and maximally saturated objects contain within themselves the generative excess that goes beyond any kind of conceptual framework or even more radically any kind of narrative or confinement of history or chronos and therefore point to the beyond, to the generative surplus that this kind of narrative or confinement of history or chronos, and therefore point to the beyond, to the generative surplus that this kind of approach encompasses in its fullest, while it synchronizes history on an even playing field and traverses or diagonalizes established historical and contextual formations. The unduly favorization of meaning and interpretation is therefore abandoned for the search of the generic in the last instance. This brings us to Natalia Serkova and her understanding of what after post-internet art is or could be. In her conceptualization, it's almost impossible to define this new or the newest art, as she calls it, and for a good reason. The problem is this one. Within the new democratic ontology, the newest art practices don't position themselves anymore in opposition to the previous ones. They don't inhabit the history of contemporary art anymore and therefore work in radical opposition to the established temporality and exit from within. The newest art is the most uncompromising and coldest variant and in a completely different way than the historical avant-garde. If it does not anymore work in opposition to the previous art movements, what's then its starting point? Everything is flat and ready at hand, while meaning and semantics suffer a fatal tool and are slowly eroded as Foucault's last men. Well, not fully eroded, but transformed and deconstructed in order to enable the production of the new, i.e. the generic. the new, i.e. the generic. As we've seen in the two images I've showed you, it utilizes the unimaginable proliferation of objects and simply starts with the different kind of associative combinatorics in a previously untaught and unimaginable ways. Its goal is maximal density in order to reach full mutation. The glaring fact is that the new objects or the assemblages of objects at first don't signal anything and are radically against any kind of interpretation. If within the history of art, this was the way that the avant-garde were reified pacified and made into a specific artistic practices or methods and therefore sterilized this is the approach this is radically bulletproof against just that any kind of interpretation or philosophical imposition we can call it a decision will be radically under determined in the scope of the new artistic object will be radically underdetermined in the scope of the new artistic object. It's not that the object in advance rejects or is radically against representation. It is more about the complete indifference to it that is driving the success of the newest art movement. As Serkova brilliantly puts it, quote, the objects belonging to the pattern of mutation do not aspire to approach you, but also do not aspire to disappear. Instead, they maintain a distance that is almost unchangeable over time. Flirtation ends where mutual recognition in the inevitable psychological intimacy happens. The pattern and its distance crumble. intimacy happens. The pattern and its distance crumble. Therefore, the task of objects of art is to flirt with you until the mutation is implemented, until the mutation finally wins." The most concise elaboration of the nuances we are dragging here was given by Serkova in her essay The Song of First Love, Art Against the Meanings. there she writes about the newest meaning not the latest but a more fundamental mutation towards new ground new playing field or ontology or in her thermodynamic example new meta stability that will start this process anew that will radically decouple objects from their supposed associations and contextual relations and open the possibility for the new heuristics, new provisional methods for constructing ephemeral and methodologically open and in the last instance arbitrary configurations for this new objective state. Within the new space where the objects have gained additional autonomy and have been freed from human imposition of meaning and interpretation from humanity's imprisonment of them there is a possibility to escape this particularist or intentional basis where as we've already said the objects couldn't care less about what we think about them and because of that work primarily true and not for us and changes from within the history of contemporary art is the history of uncovering the generic true difference being the champion whose newest manifestations mutate into the generic as the essence of artistic practice itself into final realization that there is no difference between objects at least ontologically this is the ultimate type of aesthetics one that derealizes all kinds of meaning formation and erases the relations and differences established between them thus inscribing into them a radical democracy of aesthetic and conceptual manifestation into every assemblage it forces a blind spot, methodological arbitrariness, and metaphysical indifference to the process itself. Death of the author is in the final instance death of fetishizing the difference and the indeterminacy or non-determinacy and transitioning instead to aesthetics within aesthetics that is to say it's a static singularity the latter diagonalizes nostalgic subsets and seeks conceptual legitimacy in the final instance traveling through time by the order of the real to make the full circle and to finally touch upon the work of mikhs, we have to again highlight the generic that underlies Serkova's uncompromising approach. Because the newest art is so adamantly against meaning or any kind of a priori imposition of categories in question, it enables a radically democratic aesthetic expression. Since the whole landscape has mutated and since the question of method has become one of incompleteness, the superposition of different forms and procedures habituates this new place all at once. The democracy is therefore primarily about the embrace of ontological incompleteness and the underlying fact of contingency of any system is the transcendental framework upon which it is based. I think we're losing Max, but maybe that's me. Yeah, now you seem to be back. Can you say something again? But you're still frozen a bit. Okay, now you're moving again. Has the connection? It seems that you're back. Maybe I will exit and come back. No, I think you're back now. I think it works. Okay, let's wait for Max for a few seconds. I think it works. Okay, let's wait for Max for a few seconds. Yeah, I was trying to find jokes before, but now I even don't remember that one. But it seems that he's back already. This has to turn on his camera. And computer3 says we stopped at newest art radically against meaning. Yeah let me see where this is. The last thing that I remember is that you're trying to make a full circle on the art that you showed in the first slide. a full circle on the art that you showed in the first slide. Do you... maybe I can start again with this, I'm so sorry, but I'll just repeat the last paragraph. Maybe it was cut off there. To make the full circle and finally touch upon the work of Michels... Oh my god, yeah, something's wrong with my internet connection. Do you still hear me? I still hear you, my microphone was off, yes. Okay, yeah, sorry, maybe, yeah. Maybe, yeah. Okay, now I will again proceed, but I'm kind of afraid. Yeah, so... To make the full circle and to finally touch upon the work of Michels, we have to again highlight the generic, the Tenderly Circle as uncompromising approach. Because the new art is so adamantly against meaning or any kind of a priori imposition of categories in question, it enables a radically democratic aesthetic expression. Since the whole landscape has mutated and since the question of method has become one of incompleteness, the superposition of different forms and procedures habituate this space all at once. The democracy is therefore primarily about the embrace of ontological incompleteness and the underlying fact of contingency of any system that tries to impose the transcendental framework upon the generative economy of excess, i.e. the real of the objects in question. Because the newest art has substituted the failed search for difference for the possibility of expressing the generic, it's closer to the immanent truth or the full possibility of artistic expression that awaits us after the singularity, that will bring about unsupervised aesthetic production or the radical autonomy of the aesthetics. In this way, the new art opens the possibility for different configurations that haven't been tried before precisely because of the restrictive framework within which conceptual expression operated this framing has finally been liquidated by the newest art the art after the post internet in his project michaels scraps the images from different Instagram accounts associated with the post-internet art and then in the second step, the collected dataset is accessed by a generative adversarial network that through the established latent space selects a random coordinate each day from which the image is generated and then published to the flat gaze instagram account what i've been proposing is that these new approaches enable new artistic expressions and are generic precisely because they go beyond language and other fundamentals and it's at this point where the possibility of artistic expression has accumulated in its fullest that art conclusively cancels itself. At the point where the border is less crucial than novel manners of aesthetic combinatorics and the manner itself of exploring and forming new world. is Yes, thank you that was incapable of showing you the last two slides. Yeah, but maybe we can stay... Yes, we can hear you, but we don't see you. So maybe you have to turn on your camera again. But actually, I think that's a good slide maybe to stay to stay on the last one of Matei Michev yeah yeah because I want to I want to maybe understand like your central point can you explain like what Matei is doing in this artwork or in what he's trying to achieve with these images yes he is basically like through this latent space that he is generating. He is, I mean, now I'm, this is my interpretation, maybe it will be better to ask him, but he's trying to express the full possibility of aesthetic expression through this compression and metabolization of it or through this construction of the possibility space that the I think now we're kind of trying to go back into this field of actually having a conversation. And before I also checked out the work of Mattei and if I understand that correctly, he's scraping Instagram and using then these images that are also shown in like Balenciaga and this, what Max said, like these flat parallel systems. He's then using a generative adversarial network to create like these new, potentially new images that kind of subvert or like, yeah, like kind of aestheticize or maybe the other way of aestheticizing, like making it more ugly, these usually super clean, hyper clean spaces, right? Yeah, so now I have like lots of these threads in my head, and I feel that most of them can be tied together very nicely. most of them can be tied together very nicely but i try let's try to maybe do it one by one where i i'd like sarah to maybe also react on this uh last gun piece because in your presentation you were really opposing this idea of um of guns guns and using data sets as a means of art production to create art. And I'd like to maybe know your opinion on that. That's a good question. I think the question I ask is what's interesting about using GANs outside the idea that they're new and exciting. Like what is the virtue here beyond that it's new? Does anybody else want to comment on why guns could be interesting? I guess the post-internet art that I'm kind of more familiar with, the work by people like Arti Fierkant and Josh Zittarella, a lot of it is about sort of complicating documentation. So you'd have things that there'd be kind of a physical piece, but most people don't experience the physical piece. We don't actually get to go to the gallery to see it because the gallery is in one city somewhere, you know, that not everyone has access to, and then we see it online. But by kind of messing with the documentation, it becomes unclear what, the two things become dependent on each other, the documentation and the original work. And often you look at the piece, not really fully understand what was there in the original piece. I'm not quite as clear for this piece that we're looking at here, because I think that some of these are sort of found images that then he's sort of complicating using GANs. But some of these look like sort of gallery pieces that have been sort of complicated through GANs. And I think that that can be somewhat interesting. We don't kind of fully understand how much of it is the original physical piece. fully understand what how much of it is the original physical piece um i do think that there's sort of a danger with gans that they are they create a style that's somewhat recognizable and kind of in some ways has become somewhat cliche um and i i think that like you know like what sarah's saying is sort of like what what is it really ultimately that you want to say with with the gan by using that. Yeah, I also liked what that sentence from Sarah's presentation was that novelty only suits capitalists. And if I understand Max's presentation in this kind of accelerationist realm, is that this type of novelty has to be even more fostered and created until the system collapses. Mariana, you're nodding your head, do you want to add to that? Like using guns just for using guns, should we? It's quite a complex issue, but this thing that you're saying about like this new next thing and about the aesthetics becoming very kitsch and generic and commonplace, I feel like it's the same what, like it's similar to what happened to Deep Dream, because at first like was a big novelty and like it had its own unique aesthetic, but then people started to mimic it and it began to be available for people to use it. So they generated a lot of those images and then they become like this, yeah, this heart of generic arts. generic arts. But I feel like in this, I don't personally enjoy much GAM's works, but I feel like this one that Max showed has an interesting aspect to it because it's about, especially about the generic and about Instagram that really massifies and there's a lot of studies saying that even like people's aesthetics and the way they dress and they put makeup on become more massified more the same more generic so in a way, I feel like it's a meta commentary of yeah, doing something generic to show the generality of art and Instagram nowadays. Yeah, I agree. And also, I think I heard a couple of times in in some of your presentations, the word nostalgia, and I was like, first thinking like this, the style of the generative adversarial network will become nostalgic in one sense or another. While at the other hand, I really liked Sarah's projects using like old video synth to, and also like using both like this new technology and then feeding it through old technology to kind of create this. And this was another thing that came up a lot, this ambiguity. How can we compute ambiguity? And I don't know, that's not really a question, but yeah. Yeah. Well, I think something that's interesting about the GAN pieces, and then maybe some of my work, and also the esoteric language pieces, is that all of them are asking a question about machines and living with machines fundamentally, right? But in the way that Esolings and other works seem to me about looking inside the machine, like GANs are all surface and sort of pushing back away. Like there's not really much there beyond the fact that this is a path that a computer has found between two images and you stopped it while it was going there to generally pick the more grotesque ones and then say, well, that's what's inside the machine, right? That's like sort of the argument of GAN art. And so I think it's interesting to compare it in terms of more porous ways of looking at things, which I think is what's going on with EssoLings. Yeah, I mean, we could make the same argument for novelty with esoteric languages, isn't it? Like that isn't one of the reasons why people create esoteric languages, isn't it? Like that isn't one of the reasons why people create esoteric languages because they want to do something new. And I don't know whoever feels, or like I think, or maybe to make this question a bit more broad, why make esoteric languages? I mean, I think that many SLAs kind of complicate the way that we communicate with computers or with other people through computer languages in a way that sort of explore sort of what's at stake there. You know, I'm still thinking about this idea of fuzziness. They were talking about it in Sarah's work. And I think that one thing that happens with many SLAs is that in mainstream, the behavior of code as evident to the programmer as possible, as clear to the programmer as possible, so that we know exactly what to do and other other people can code later, they'll understand as well what the person sort of had in mind. And many SLAs kind of put up somewhat of a barrier by alienating the human reading of the code from the machine reading in a way that complicates things either to comment on what's actually happening when we write code. One thing, for example, is that there's very little code that's actually perfect. One of the main things that programmers do is write bugs. And, you know, like what Sarah was saying, with live coding, a lot of these systems are designed to at least do something if it's, you know, so that if you make a mistake, the whole thing doesn't shut down in the middle of a performance, but it just might do something that's a little bit different than you intended, and those mistakes can actually be very interesting, something that you can build on. And a lot of interesting SLNs actually sort of build on that as well, sort of deal with the fact that we're creating bugs all the time, and that's really what happens um the the example that that uh marianna gave of inner cal which is really kind of often considered the first um the first slang is is one where all of a sudden we have a whole bunch of new rules we have to talk to the interpreter in a way that it finds polite enough but not so polite that it finds us simpering and just decides to ignore our program. And all of a sudden we have to deal with all these rules that are never made quite clear to us of how polite we have to be in order to. That's sort of the counter of that thing of allowing a fuzziness. It's like, okay, well, no, now there's a fuzziness around rules that we have to get our stuff 100%. Yeah, there's another trope that I might want to get into, which is, of course, like, one programmer creates lots of bugs but when and this is also what what we see in MelBolgi it's this programming language that both Mariana and you were talking about is that the creation is actually then like a collaborative effort and when I I don't know when I program I and I look through like github repositories and the issue sections there I see most see most of programming is actually like this collaborative effort. And also in Sarah's work doing the live coding performances where there's like this, maybe that's like a nice metaphor as well. That's like kind of programming is always like this jamming together or like performing performing together writing code and yeah I don't know if that's also a question but like this this idea of like collaborative creation also through through the through the camp kind of came up a lot where, yeah, I think somehow then like these, these gun pieces or like, again, machine learning, there's like lots of people creating, creating model or like creating the architecture of the model, but then it becomes super easy for like one end user to feed it with data and then get something, something new kind of out of it um yeah and uh i don't know maybe uh there's like some more examples that you know mariana where uh isolangs kind of work with this with this idea of doing something together rather than uh working alone or like other other thoughts that you have on on this yeah not specifically about this but i have some thoughts on this what you just said about obfuscation in systems and like with guns we don't really know what's going on like because, because also the code was done not just by one person, and like other people just adapted and like build on top of it, but never really seeing what's under the hood in a way. And yeah, I feel like in a way, as a length language try to open this black box of computing sometimes because it kind of also forces you when you're trying to learn it how to learn how the machine really works inside and also if you want to do something different like with the languages that were refusing English to see what really happens when you try to do the stuff that are not the common or the normative. Yeah maybe connected to that is the like you talked about these fungoids uh as like as a lengths that are like multi-dimensional and they kind of grow in patterns and uh one of the questions that you also already are answered in the pet but maybe that we can also get into uh it's like this idea of like a growing uh or growing uh um a computer program can you uh or do you want to tell us about that you don't have to of course okay i can talk about a little about it it's still a work in progress so there's a lot of uh stuff i really because it's a big topic and i need to do more research but it's basically I'm trying to develop a conceptual as a link called herb funds, which is based on the structure of trees and by this concept brought by a gate of the archetypical ideal plant. And the idea is to echo this hypothetical tree like diagrams that were the representation of knowledge early and they even have like a lot of theological meaning and they were like the kind of the supreme representation of logic or diagrams so they were like the original diagrams that humans made so it i would like instead of representing a tree computationally like L systems do, I'd like to do the opposite, like seeing trees, what could be like with their branches and leaves, kind of see there a pattern that could be like a source code and I could change this, yeah, this relationship with arborescent non-human beings and also, yeah, like reclaim this ancient tree knowledge and also echo this thing of wood wide web and so but try to to do this in a more like computational language uh um brings me to mind that um like this uh theoretical thing but it's actually another connection where in daniel's talk about malbolg where you have like this set of rules it reminded me a bit of like religious texts almost where I don't know could could it be that I don't know can we can we maybe see code and how we write code as a kind of as a kind of religion where people it's it's both like the fuzzily defined, but also like there's room for interpretation, but there's also a group of people who believe in a certain system. Do you want to touch on that or is that way too out of there? I don't know. Well, of course, Malbolge is named for a level of hell. So it does tie into that religious idea right on the surface. I mean, you know, Olmsted really wanted to create something that would be like hellish for us to try to program in. But in the end, people really enjoy trying to figure out how to make code function within this language. And of course it does take this community of people to sort of discover how the language works and really what it's all about, which I think is also the way that religion kind of works as being sort of something that's both individual and communal in that way that sort of takes community to build i suppose um i i think that um yeah i guess those are sort of some some thoughts along those lines maybe adding to that that's a different topic now now, but it's from the audience that was asking if how isolangs inspire or influence non-isolanguages, or a question that I wrote down was more like, do you have an example of an isolang that became mainstream? So usually, so in terms of becoming mainstream beyond being sort of, you know, appearing on General Hospital, I assume people are talking about actually finding ways to use it for practical computing. I mean, S-Links are really sort of defined by the attempt to make something that's not practical, but that doesn't mean that people don't find uses for them. I've mentioned JSFuck, which is this discovered esoteric style where you can write JavaScript with punctuation that nobody intended to encode into JavaScript discovered that you could put this way. And then when it was first discovered, people realized that they could be found on a whole bunch of JavaScript filters by writing stuff in JSFuck. So a lot of people wrote malicious stuff in JSFuck. I mean, that's the practical thing that came to mind. And then, of course, people had to write better filters that were able to recognize JSFuck and be able to reject it. But there's a lot of crossover between SLAs and obfuscated coding. Obfuscated coding is of course used very often for people to make code that other people, to sort of protect intellectual property so that we can't reverse compile code and programs and see how they were written. So in that respect, there's this crossover and sort of somewhat of shared concerns. And of course, there are some theoretical languages that have eventually led to practical coding and there's a big crossover between theoretical languages and S-Links as well. I have sort of two answers to that as well, sort of two ways that I have found S-Links to end up being useful, which is, there's one called Whenever, which is a language that just executes based on what's in the bag sort of randomly, you can't have control flow. Um, and so I did an implementation of that and started messing around with using it to make games. And like many side projects, it sort of died on the vine, but that's like a really clear example of a SLN that I think is useful for a very specific type of project that would be great for a game or like a choose your own adventure type thing. The second way that SLNs have been helpful is basically like my actual work career now has come out of what I learned about languages. So when I was at the School for Quidditch Computation, Ramsey Nasser, who was mentioned before, was a teacher there and taught us about how you write computer languages. And so a lot of that was about SLNs and other like domain specific languages. But the moment that I knew how to write a weird computer language meant that like all of computing was demystified, you know, like my joke is, Oh, computers made by idiots, just like me. And like, and now my actual job is to work on JavaScript as a language today. And that to me is like amazing. And that came out of that first encounter to understand them as malleable, changeable, not magic things. And I think that that's really the practical use of SO-Langs, whether you end up wanting to go work on compilers or just being like the computer's not working because it's wrong, not because i'm wrong right like those are like different sorts of relationships that i think are generated that way that's another an amazing quote the computer is made by idiots just like me um i love that um maybe going shortly into the last point it's also the last point of, that has been the last point of Mariana's presentation. She was mentioning 100 Rabbits, who is like this group that lives on a sailboat and travels on the Pacific Ocean from Japan to the US and Canada. ocean from Japan to the US and the Canada and they're creating all these little tools and programs and games that run on super efficient old machines. And yeah, they're like they made even like musical instruments that are also using a two-dimensional array to create musical notes and MIDI, basically. And so, yeah, maybe going into this thought experiment of how... And this was actually, I think, yeah, Marianne also said in the pad, was saying that mostly she's coding mostly web-based applications and also use Sarah. And I had this, so there's like this personal anecdote of like now lots of people use Node.js, which is like this programming languages, language that is used for servers. But it has like the same syntax as front-end language and lots of people like it and use it. And in my recent projects, I kind of had this problem that I needed cheap servers that can, that don't have lots of RAM, that need to be highly efficient. And lots of people hate on PHP, but this was actually, I mean, for some reason, people in the 90s were able to create server code that was not, like, that didn't need eight gigabytes of RAM to create simple websites. So how do we counter this? Like, that seems, like, it seems very strange that we are, like, we get high-efficient computers and what we do is create high efficient programs i don't know who mariana do you want to um or sarah i don't know it's just a bit uh that i'd like to say is that yeah it's really interesting because like even with with this post-internet context, what we have now on the internet, we have very low semantic code. The actual semantic code, which is the source code of the page that actually translates into content, is just a tiny fraction of it. And usual big websites are bloated with cookies and, like, third-party cookies and stuff like that. So, like, even if we, like, we have a lot of, like, sorry. Yeah, like, we have more computational power and like big servers and everything, but we keep like just pushing to the limit with like bigger than ever websites. And that's not like a very sustainable option. And you see like that internet also could not go into this path for a long time. So we really need to rethink this and find ways to have a kind of counter innovation. And I think it's what inspires me most about 100 Revit projects, because they really try to do the most with very low resources and tech so yeah uh i just want to add oh i'm sorry go ahead go ahead daniel well i i interviewed 100 rabbits a few years ago for s-ter codes uh and i'm a big fan of their work um and actually the dates that i have on s-ter codes on all the posts use their dating system of Arvillie because they, in addition to writing a lot of their own software, they come up with their own system of time and dates that work better for them. If something doesn't really work well for them, they just come up with their own version. I mean, they're in a situation where they have solar panels on their boat. They don't have a battery. So if it's a cloudy day, there's only so much power they have. So they ended up having to write their own version of Photoshop and basically their own version of Illustrator that are low power that'll work with the amount of electricity that they have available. And additionally, they don't have a refrigerator on it to eat a certain way. It can be stored for two weeks on a boat without a fridge. So because they've created these constraints for how they work, they've been able to build all these systems out of necessity to function within it, which I think is a great way to live in a way where we're using less power and we're less wasteful. and we're less wasteful. I was just gonna say that, like, I think people live up and down to the constraints that are given to them and the requirements that are given to them. So, right, like even with cookies, like that's a constraint imposed by capitalism and it's a bad unnecessary one, but to some extent people are doing those things because that's the constraint that they have for their business project. And then likewise, yeah, I think the more computing power you get, the more computing people want to do. And that can be both good and bad. Right. So Daniel mentions dates. One of the projects I'm working on now is bringing a new date system to JavaScript because JavaScript date is hard to use in some way. So there's a new one called Semporal. It's really complicated because at the end of the day, it's definitely much bigger than the original date representation because dates are very complicated and representing them in a fair way is really complicated, right? Or we think about ASCII versus Unicode. Like ASCII is smaller and had a smaller footprint, but it was also cutting some people out through those choices. So I think like an interesting thing to think about going forward is ways to reduce the footprint unnecessarily without doing it by like assuming certain categories don't exist or something. And I don't know what that looks like, but I think it's an interesting question. what that looks like, but I think it's an interesting question. Well, it certainly seems that we're still moving in the other direction where we use large language models that are trained on billions of websites to help us then use technology more efficiently, which I think is also a great part of EZLang's that they try to counter this idea of efficiency. Yeah, if there's no like pressing thoughts that we still have to go into, then I would wrap it up here and maybe close with, I have to like Max internet connection went so bad that he couldn't join the last part of the conversation but in the chat he wrote that this is one of the interpretation interpretations of how GAN works but I believe that analysis of the functional properties of GAN was not the original goal of the artwork that he presented the artwork that rather use scans as a metaphor to suggestively point out some legitimate ideas um yeah um yeah and lastly i want to uh invite you all, like you speakers, but also the people at home to put on headphones and have six and listen to around six to seven minute audio piece by the Berlin based group, IKO. And together with Naoto Hieda, who created the visuals for it in a programming language called Hydra, which is a WebGL-based thing made by a person called Olivia O'Jack. And yeah, that's it for the symposium. I'm really happy we could still make it, though just online um next time we will see each other in person and grab a beer afterwards so thank you all also to all the helpers in in the background Thank you. Succes! Who are the humans? Who are the humans? Who are the humans? Who are the robots? Who are the humans? Who are the robots? mouth tongue lips teeth mouth tongue lips teeth mouth tongue lips teeth roof of the nose mouth tongue lips teeth mouth tongue lips teeth mouth tongue lips teeth roof of the mouth See through from the bottom. What is Liyiva's surname? It was suggested that her human's name was Slashil Slash Seattle Celtics. Does she like to sing? I have no voice. No, I mean Liva. Query, have you ever cheated? Does Liva have a voice? You. Hi, Naoto. Hello, their name. What is Liva's favorite poem? Leaves of Grass. What is Liva's favorite book? Liva does joke around and loves to give sarcastic responses. She also loves to read books, especially horror novels. What gender does Liva have? She is female. Which music does Liva like? Liva likes to sing and most of all Liva likes to dance. What is art for her? Art? Who are the humans? Jaws and jaw and teeth. Who are the humans? Cheeks, throat. Who are the humans? Temples. Who are the humans? Jaws and jaw and juice. Who are the humans? Cheeks, throat. Who are the humans? Temples. Who are the humans? Forehead. Neck. Who are the humans? Chin. Who are the humans? Head. Who are the humans? Timbana of the youth. Who are the human eyes? and the ample side round of the human. You are the human. We are the human. Because I am the human. You are the human. We are the human. Because I am the human. You are the human. Whispering, shouting aloud. Because I am the human. You are the human. We are the human. Because you are the human. I am the human. Because I am the human. You are the human. We are the human. Because I am the human. because I am the human. Freckles, hey. You are the human, we are the human, because you are the human, because I am the human. I am the human, we are the human, because I am the human. We are the human, we are the human, because I am the human, because you are the robot. We are the human. Because I am the human. We are the human. We are the human. We are the human. Because I am the human. Because I am the human. I equal you. You are the human. Because I am the human. We are the robot. Because I am the human. We are the human. And breathing it in and out. When fear was fed, naked meat off the human. Because I am the human. When fear with fear made it meet of the body. Because we are the human. Because you are the human. I am the human. Because we are the human. You are the human. I am the human. Because we are the human. You are the human. Because I am the human. You are the human. Because I am the human.