I’m taking the challenge by @vgr to offer up 100 takes on “Glitches and bugs in digital art”:
1. Firstly, all technology is created by humans with finite resources.
2. Most of what we think of as “glitch” artistically is the exploitation of an edge case in technology, not a failure mode
3. Actual bugs and failures of technology usually result in little more than an error message, a black screen, an empty file
4. The Fantasy Glitch, as it operates in the cultural realm, is a stylistic flourish that offers a peek at the edges, calling the structural support of the image to the foreground.
5. Every image is constructed through a complex system of translations between different representational spaces before it reaches your eyes.
6. The opposite of the glitch is the perfectly rendered naturalistic CG image, complete with raytracing, caustics, etc, perfectly shown full screen with absolute fidelity. This is the ultimate drive of the technology developed for images
7. If you work backwards from the naturalistic render, you will find several layers (a stack!) of technical intervention that had to operate as designed. The tools should have been operated by someone sympathetic to the intended function, with some craft.
8. So the way you get there is a whole bunch of people all following the rules and doing their part to create an image and distribute it to eyes
9. As an artist, you can choose any layer in that stack to operate. You determine what level of abstraction (conceptually and visually) you want to operate within. The further down you go, the more likely you will called a Glitch Artist.
10. One time I came upon a New Media Art work in a gallery that was turned off. When asked, the gallerist said nobody knew how to turn it on and the artist was in Germany. A rare(?) case of Glitch Art.
11. As I mentioned before, software is made by people with finite resources. Some of those people are artists
12. The unspoken agreement of software developers and users is that they will try to meet in the middle. The developer works in good faith to address the "use case" and the user does their best to avoid "user errors" and "RTFM"
13. There comes a point in every digital artist's life when they find themselves at the edges of that agreement. Image compression gets everyone. Do you:
A. Back away from the edge
B. Explore the edge
C. Brute force your way past that edge
D. Write your own software
A. Back away from the edge
B. Explore the edge
C. Brute force your way past that edge
D. Write your own software
14. I know professionals at every level who have chosen each of these options. All of them are traps.
15. When you are building tree forts at the edge of the map, someone will tell you that you are "reinventing the wheel"
16. When you look at the complexity of the stack that brings you a technological image, and you look at all the unsatsfying images, it's hard not scream through the stack and patch in other options
17. Now more than ever, people need to believe that there is an edge to software. Maybe software didn't eat all of the world yet. Maybe we can still throw it off course.
18. The Glitch speaks to that need to believe that the image world we live isn't fate. If it can sparkle and trip over itself, it can become more human
19. If you can create ways of producing an image that follows an orthogonal logic, that operates along a different axis, it follows that there is something human that we can extract from these tools
20. Glitch is a kind of optimistic nihilism
21. Most digital 2D image making is just a more refined version of things that were happening in MacPaint. A lot of the tricks have been around a long time now.
22. I like to think about "clumsy technology" as a frame for what I make. I think of this as similar to Deleuze's "stutter" but more slapstick 
23. I have always been pretty clumsy, maybe because I daydream a lot, despite many attempts to get good at dancing and climbing and stuff like that. I think it's on some level charming and human.
24. Being clumsy reminds me not to take myself too seriously, it offers a break in the expected order of things. Bugs are like that too
25. A bug is also an expression of your finite abilities to think and perceive and act as a programmer. It's a pause. I'm a clumsy programmer too.
26. Software is supposed to work, but what if it's clumsy? What if something funny happns instead? It's funny to me that when software is doing the expected that is "working"
27. Sometimes software needs a cigarette and a coffee, it can't just work.
28. Sometimes software is working when you went to bed, but it doesn't work the next day
29. sometimes you want to just acknowledge that images are constructed in the first place. It softens the image to see those artifacts.
30. David OReilly (@davidoreilly) wrote a great essay called "Basic Animation Aesthetics" years ago that I have recommended hundreds of times (google it!)
31. The essay is a very specific outline of choices he made in creating a film, but it points out how much of computer generated images are handicapped by decisions made before you got there.
32. The important part is choosing what you really want. Sometimes that's about an affordance, and sometimes it's a struggle against everything the tool was designed for.
33. I know graphics people that say "everyone should write their own engine." That is so hard! You have to find compromises
34. Sometimes I'm the clumsy one, sometimes it's the software, sometimes it's both of us. That's collaboration!
35. I have gotten to a point with graphics where I regret not knowing more calculus and linear algebra, and I also regret not being a better animator, and also not just doing really obvious things.
36. The best glitches are the ones that happen in my mind when I try to understand something difficult. Some of the most interesting things I've come up with had to do with a poor or mistaken understanding. Clumsy brain!
37. I think often about the oral history of the Chairlift and Kanye music videos (the ones that made datamoshing a mainstream thing), how much that threw everybody, how many artists were disrupted by that moment.
38. There's about 10 different versions of the "datamosh" timeline, so I don't want to get into it, but what interests me is how it caught fire and captivated everyone (myself included) and got a whole bunch of people fooling around with video files.
39. Maybe this disqualifies me to speak about Glitch, but I never liked fooling with files like that, for the same reason I don't like GANs and all that now.
40. For me the whole appeal with working at the edges of image technology is manipulating the process to chase after looks and creating a unique interaction between my clumsiness and the computer's. All that stuff is just slot machine art.
41. Glitch is a convenient way to discuss image creation that's informed by approaches often associated with synthesizers - generativity, modulation, signal flow, feedback, sequencing, filtering and distortion - but it avoids talking about the most interesting parts.
42. One of the interesting things I discovered talking to people at Moog was how much they struggled to introduce synthesis as a concept to the public. Visual synthesis is completely underground
43. If you've ever read a SIGGRAPH paper, you can appreciate what a tedious process we have gone through to get Frozen 2 in 2019. How intimidating it can be to even imagine some other future sometimes.
44. I believe we can find language to talk about visual synthesis that isn't framed by failure ('glitch') and centers the actual investigation.
45. The hard part with talking about a synthesis of imagery is that we've got 3D vertices, vectors, raster, framebuffers, shaders, Canvas-style paint stuff, actual photos, etc. Where's the signal live?
46. Where do you locate the material of visual synthesis? We'll figure out how eventually.
47. You can get extra-deep into this conceptual vortex with me, but we all know the outcome is some cool images, some shiny FX, a little texture on the poster.
48. I like those fake analog tape glitch FX people make. I appreciate the desire to bring that texture into the future.
49. Part of the reason I like the fake glitches is that every technology has artifacts that are invisible to us because we just want to watch the picture, and then they disappear when the next technology comes along. There's a kind of macabre in dredging up the old crust.
50. Have you ever met someone who maintains a 15 year old computer because it runs a buggy version of some application that's important to their process? That's a thing.
51. Nurturing a legacy bug is a fascinating side-effect of computers in art and music over decades.
52. I've heard rumors of a much loved legacy After Effects a lot of experimental video people use to get feedback effects. Can't remember the name.
53. A lot of digital art software is poorly programmed. That's fine! It's probably a waste of time to write good art software.
by "art software" here I mean software written as art by artists
54. The longer art software sticks around the less likely it is to work. Longevity is a real problem. I can barely keep websites working for more than a year without updates.
55. My friend @_menkman has written way more coherently about lots of this than I will manage in a thread beyondresolution.info/PUBLICATIONS
@_menkman 56. Also @NBriz writes about Glitch from a more ethical/political angle and has made some cool tools nickbriz.com/thoughtsonglit…
58. A lot of good ideas about digital art were written about music first. It's hard to form a complete picture without knowing both video and sound history
59. I keep coming back to the idea of software as the product of humans. It's easy to feel like it's elemental, that it came to us fully formed, because of the interface, but it's all made by people who are making hundreds of assumptions
60. As an artist, you are trained to metabolize art history up to the present, because you need to operate within that history materially. The minute you engage with software, you are operating within another historical framework.
61. I used to like reading Cennino Cennini's quattrocento writings about painting. I don't know if that helps.
62. It's a real detriment to our sense of aesthetics and history that we have no idea who decided that the stamp tool in photoshop works like that.
63. Of course we have Blinn and Lambert, and all those legends of rendering, but we don't talk about the tool designers and their aesthetic choices. Their challenges and cleverness.
64. The fact that we don't talk about those choices makes it seem like it was always self-evident because you're obviously trying to make the thing everyone makes. That's ridiculous!
65. My point in bringing up the tool designers is that part of how we think about Glitch conceptually involves "breaking the system" and taking power back from the given software, but I like to think of it as more of a social relationship. You're challenging the author.
66. I'm lucky to have met a lot of the people who've invented, designed, and programmed all the tools we take for granted. Maybe we are the last generation that gets to have that. It totally changed my view of all this stuff.
67. There are always people that will make you feel like you aren't doing digital art the right way, because people fall in love with solutions, but the enduring fact is that technology is hard, software is all equally imaginary, and you pick what makes sense
68. The relationship between humans and computers is really imperfect! It's clumsy. We're all figuring out a way to circumvent that enough to make something we care about, or at least to feel like what we're doing is fun or cool.
69. sometimes I imagine an extra Lefebvre chapter called "in the computer"
70. Slot machine glitches will only sustain your practice for 2 years tops, then you gotta start exploring synthesis concepts (coding, etc), cycle through meaningful source materials, or find another exploit.
71. It'll take a generation or more to reconcile all the digital artists making work under someone else's name
72. When you say "glitch" nobody knows if you're talking about an effect, a cohort of artists, an ethical approach, a software fact, etc, and it starts feeling like you're trying to say "GIF" right.
73. few of the great artists in this vein will ever wind up in a museum collection because most operate outside of the art market or only occasionally engage or stop making work of their own. It’s up to curators and critics to get curious and gather stories
74. Once you discover the edge of software, you are bound to end up at the edges of culture
75. When you have realized there is little difference between audio data and image data, you start seeing all sorts of social divisions as being meaningless and construct for the sake of a use case
76. Generational changes in software dev and art have changed all the terms of New Media. We now have mature software tools that sit at every outpost on the map, hundreds of entry points.
77. The power relationship between industry and artists is really different from 10 years ago, so it’s hard to use the same ethical lens.
78. When OS X was first introduced, it didn’t handle stable timing very well, so for years electronic musicians ran old laptops with OS 9. Long past the point where it was fixed.
79. Some bugs develop into urban legends for artists, and people develop bizarre rituals to accomodate them
80. Software shapes us in unexpected ways
81. The reason to consider software design as an axis of art history is that taste is always factored in to how those tools operate.
82. from an artist’s perspective, blindly accepting that taste as a given is fine, but it means operating in a default mode.
83. default modes in technology don’t stay invisible forever, and eventually become part of a particular time period
85. Software used to be funnier
86. The stakes have gotten too high for most software. I’m for a low stakes, small scale of technology.
I feel like VC-backed social media has become a dark cloud over software. It occupies so much of the thought space that nobody has time to talk about the other stuff
88. (forgot to number the last one) Facebook is the photoshop of interpersonal communication. Everyone hates it, but its scale prevents other solutions from really replacing it (jk i like PS)
89. I care more about pro-grade computer graphics because I benefit so much from its technical side effects
90. The debate about practical vs digital fx is largely due to how certain things are cheaper in the real world than in CG. (floppy hair and fabrics, dust, convincing dynamics, shadows) All the messy stuff.
91. the analog vs digital debate in music is largely about restraint. Inside the computer, it never quite feels like you have enough. I always used to tell students to look at their images on a projector for the same reason.
92. Some of the most interesting writing about image synthesis and its potential happened in the 70s. You can find a lot at vasulka.org
93. The electric guitar is the most salient glitch instrument. Modular synthesis may catch up
94. The idea of a visual instrument is nowhere near mainstream. The only reference points are “visualizers” and hippie light shows.
95. For most people the moment you move away from strict representation in video and animation, you are making drug art
96. One difficult part of working on experimental realtime video processes is that our brains become desensitized very quickly to visual stimulus. It can be really jarring sometimes when you look at something tomorrow. fwiw painting is like that too
97. There’s a fine line between trance induction from staring at your generative video experiment and other spiritual forms of ritual or meditation. It takes a lot of discipline to not go there.
98. I simultaneously tread a thin line about not being too trippy, but also practice at the very edge of sensory experience and technology, and I’m probably permanently tripped out as a result.
99. I can’t believe how long 100 feels. I’ve been operating at the edges of both technology and art since around 2002, but I rarely take time to think like this
100. People joke about the “intersection of art and technology” but I like to devote my time to making beautiful jokes at the clumsy edges of art and software. /fin
