Given that I found said flyer in a comic shop and that the exhibition was in the gallery space of the same shop, I can only assume the exhibition was degrading of Lichtenstein's work, which is part of why I didn't look any further- personally, I don't think what Lichtenstein did was wrong or right- sure, art theft and tracing/eyeballing is wrong, but Lichtenstein also injected his own sense of style, and I'd be very surprised if there was any evidence of this sort of practice beyond his early work. It's more the idea of setting up the exhibition, "IMAGE DUPLICATOR", to degrade or question the integrity of Lichtenstein's work leaves a sour taste in the mouth. Of course, it's understandable why the comics community would be sour about this, even beyond the ethics of Lichtenstein himself, given that Lichtenstein got a retrospective in the Tate this year, while comics fails to this day to be taken seriously as an art form. It's more the idea of an exhibition used to demean art- something that to me will always be reminiscent of the Nazi exhibition of "entartet" work in 1937 as part of their regime.
So, just as the Orbital exhibition was a reaction against the Tate exhibition, so is this blog post against the Orbital exhibition, sort of. I just wanted to look at using other people's work as a "copycat"; after all, in a limited sense, this is how one improves at art: by copying lots of different styles and picking and choosing elements from each into a sort of Franken-style.
I discovered Ross Sutherland on Radio 4's "Shortcuts" by Josie Long, which is a programme I really love, it makes for good listening whilst drawing. Sutherland writes poetry, but in a very unique way, using only words from another person- in this case, John Humphries on one episode of "The Today Programme"- in order to appropriate another's vocabulary and thoughts into a second voice. Sutherland mentions this is helpful as he is depressed, and therefore takes the outside world in a bite-size chunk for someone who has problem connecting with that world otherwise. I would also suggest that stealing someone else's vocabulary is a limited way of stealing their identity, and therefore would work like a fantasy, distracting Sutherland from his own sense of self and mental illness by immersing himself in the life of another (in arguably a more intense way that regular books or films etc, as Sutherland worked with a single episode, a single character and a single sense). The chopped-up poem in its finality also says some interesting things about the linguistics of news, in that the poem, "No More Questions", features a lot of rhetorical, or unanswered, questions; a comment on both the style of the BBC's radio broadcasting in that it asks a lot of questions of the listener as way of encouraging social media comment, as well as in an existential sense adding depth to a simple news report.
The next thing I discovered is one of the many many things that cross science and engineering (in this case, computer science) with art, even though they are considered opposites (eg, if you take the IB at sixth form, there's an arts stream and a science stream, as if ergonomics and design and engineering aren't really close).
Google Research released the code for some visualisation software that helps Google Images work out what is, for example, a dog, and display these as part of a search result (normally helped by the keyword on the webpage, though ultimately there are a lot of images normally on a webpage with said keyword) as well as automating "Similar Images". The software is called DeepDream, and is the result of Google feeding 1.2 million pictures of individual objects through a computer and creating this software that can recognise images by mapping like the above image, using images it already recongises (tending to be eyes and dogs, for some reason). I can't explain how this works but I can sort of visualise it, and also this guy can explain it pretty well!
Now the code is free to use, it has been taken on by artists as a way of creating psychedelic work that seems to strip back the layers of a computers mind. Specifically Johan Nordburg, who instead of creating a single image with DeepDream or applying it to a video, applied it to some random noise to produce an image, and then reinputed that image, so DeepDream keeps on finding images within its own mapped images. Nordburg then creates a film with the resulting images (around 8400 images, given the video is 5:36 and Nordburg states there are 100 frames every 4 seconds).
The resulting video is incredible. DeepDream was meant as a research tool, not the final software for recognising images, given that it has a long way to go, not only in terms of the internet but also for use in robotics, so it's key for researchers to be able to map how computers "dream". I think it's interesting because of the obsession with computers to be like humans, they have to have imagination and creativity- I think this is incredibly creative and far beyond what any human could conceive of, which is why it's so mind-blowing- and yet, this isn't even half of what computers can do.
Science aside though, the free use of the code and the consequent creation of something remarkable with this code, attributed to a person even though created by a computer, seems like something of the future. I don't think things like this are taken seriously enough as art (not least because it's licensed under Creative Commons under Non-Commercial Share Alike, or not-for-profit; and available on the internet: two things which often lead to art being downgraded by the established art world). It's also strange how weirdly simple these strange images seem in a sense- I thought of it like a complicated version of layering up CMYK, in that each layer, instead of a colour, is an animal or object. I wonder how different this neural network is to our own biological network in this sense- are we really thinking or just recognising images in noise? And if so, where is the difference person and computer, and what attributes Nordburg rather than DeepDream as the creator of this work?
No comments:
Post a Comment