https://pad.constantvzw.org/p/apass.compression
Compression and encoding

"Codecs perform encoding and decoding on a data stream or signal, usually in the interest of compressing video, speech, or music. [...] Software such as codecs poses several analyrical problems. Firstly they are monstrously complicated. [...] Second, at a phenomenological level, they deeply influence the very texture, flow and materiality of sounds and images. [...] Third from the perspective of political economy, codecs structure contemporary media economies and culture in important ways. [...] Despite or prehaps because of their convoluted obscurity, codecs catalyze new relations between people, things, spaces and times in events and forms." Adrian Mckenzie, "Codecs" in: Matthew Fuller (eds), Software Studies, a lexicon (2008)

This session starts with poking holes in different image and video-files. From there we will try put the vocabulary of encoding and compression (Codec, Container, Compression, Interpolation, Interlacing, Artifact, Bitstream, Sampling, Conversion, ...) together with some of the key terms in the work of the philosopher Gilbert Simondon (Allagmatics, Transduction, Analogy, Individuation, Cybernetics, ...). How do structures make operations appear, and vice versa?


Netflix 30% of internet traffic
from files to streaming

PR: "to be less alienated by powerful technologies"
a.pass between research and school. Learning to look at structures. Anti-pedagogy turned to technology.

SS: "coding as a second nature" "a consequential material practice" "fused layers of organic and digital infrastructures"

LM: "what do forms of coming together produce" "interest in the politics of things"

Shanti: "Public administration, brain science. Bodies that share weight. Consent, borders. Naivite in laborotaries"

Difference between binary and 'plain' text files. Machine vs. human readable.
Needs a specific tool - hex-editor. Binary is a numerical system -- 0-F

hexadecimal: binary re-writing, by using letters: goes from 0 to F.

From Binary to Hexadecimal
https://en.wikipedia.org/wiki/Hexadecimal
8 bits vs 4 slots -- same info, differently written = more concise, like stenography -> https://en.wikipedia.org/wiki/Shorthand
16*16=256 possibilities

0001 = 1
0010 = 2
1000 = 8
1111 = f

beware that each 4-digits block is not representing/rewriting a pixel
relation between pixels and bytes becomes more and more complex over time

bmp is a relatively direct -- pixels split in RGB channels
trying to understand this relation by overwriting some bytes

imageviewers, software, browsers interpret binary/hex formats to show the picture.

Mac: http://ridiculousfish.com/hexfiend/
Windows: http://www.wxhexeditor.org/
Linux: 
    sudo apt-get install hexcurse ncurses-hexedit ghex

JPEG is divided in blocks of 8 x 8 pixels, there are operations on each block. Revealing the Discrete Cosine Transform!!
64 passes per 8 pixel blocks

from changing frames
to changes between frames

video needs to be more flexible about poking because of the environment it lives in
Moving Pictures Experts Group = mpeg = ISO

"I am for downloading"

change in temporality
from frame jump ...

fractal views

machines are talking to machines. Where is the frog!?
efficiency -- calculation

reassured by glitches
efforts in making things smooth
glitch and its relation to the thing

going in a straight line from A to B is possible -- so we do.
different storytellers are involved, and emerge, and show themselves.
"quality" and accuracy of vision/showing

"excess quality"
constant bettering of the thing

more resolution, somebody is doing the looking -- labour
hi-res, hi-definition, dimensions to be added.
4k - 5k - ...

Nicolas Maleve
http://sicv.activearchives.org/
https://machineresearch.wordpress.com/2016/09/26/nicolas-maleve/



Morning: poking files
https://cloud.constantvzw.org/s/eGmApCmrlAWSA2l

Codec
Container
Compression
Interpolation
Interlacing
Timeframes
Blocks
Artifact
Bitstream
Sampling
Conversion

"jpeg is a moving image"
jpeg is an oil painting re-made everytime it's opened.

"lossy compression"
https://en.wikipedia.org/wiki/Discrete_cosine_transform


https://en.wikipedia.org/wiki/Chroma_subsampling

Fileformats 

.bmp vs .jpeg

Moving images

.gif vs .mpeg

glitching a movie: poking -- when does it break
revealing timeframes and what happens in-between

Certain files are only understandable (viewable?) through more or less computationally intensive? operations
accelerations of the amount of operations -- making more efficient, less reversable?

Afternoon: The Allagmatic


--

https://pzwiki.wdka.nl/mediadesign/Jpegloop
https://pzwiki.wdka.nl/mediadesign/Video


"The assumption that the structure of something is the same on all scales"
https://pbs.twimg.com/media/DWZJI8TXkAAI1OT.jpg

G'MIC filters on parade
http://opensource.graphics/christmas-is-already-here-for-image-processing-folks/

resynthesizer for gimp;
https://discuss.pixls.us/t/resynthesizer-for-gimp-v-2-10/2714/14
https://pixls-discuss.s3.amazonaws.com/optimized/2X/2/26f69c512ec1ab28d79531adeec53d886d1b6f53_1_690x355.jpg




sina/  question of "quality": the more accurate and sensory the evocation of things, the more we have the sense of someone there doing the looking <-- USA spectacle knows this too well
the question of the labour in creating resolution in the experience of image
[a way of consciousness that literature very well equiped to describe, poetry constantly plays with and redefines "resolution"  in its descriptive acts]
storytelling industry and scientific knowledge objectivity depend on certain rationalization and naturalization of ideas of "accuracy," "quality," an idea of photo that works with an accurate rendering of sensory appearances. industries such as netflix depend on sustaining the "perceptual identity" of its objects [in this case the images of TV series].
-do we need to reinvent "accuracy" every time? maybe yes. --> which informational infrastructures we permit remaining uncriticized for us to be able to do the work we want to do in the limited amount of time that we have?