<< preface

this blog is nina wenhart's collection of resources on the various histories of new media art. it consists mainly of non or very little edited material i found flaneuring on the net, sometimes with my own annotations and comments, sometimes it's also textparts i retyped from books that are out of print.

it is also meant to be an additional resource of information and recommended reading for my students of the prehystories of new media class that i teach at the school of the art institute of chicago in fall 2008.

the focus is on the time period from the beginning of the 20th century up to today.

>> search this blog

Loading...

2011-01-18

>> Glitch. Olga Goriunova and Alexej Shulgin

from: Matthew Fuller, "Software Studies"
text mirrored from: http://dm.ncl.ac.uk/blog/wp-content/uploads/2010/02/softwarestudies.pdf

This term is usually identified as jargon, used in electronic industries and
services, among programmers, circuit- bending practitioners, gamers, media
artists, and designers. In electrical systems, a glitch is a short- lived error in a
system or machine. A glitch appears as a defect (a voltage- change or signal of
the wrong duration—a change of input) in an electrical circuit. Thus, a glitch
is a short- term deviation from a correct value and as such the term can also describe
hardware malfunctions. The outcome of a glitch is not predictable.
When applied to software, the meaning of glitch is slightly altered. A
glitch is an unpredictable change in the system’s behavior, when something
obviously goes wrong.

Glitch is often used as a synonym for bug; but not for error. An error might
produce a glitch but might not lead to a perceivable malfunction of a system.
Errors in software are usually structured as: syntax errors (grammatical errors
in a program), logic errors (error in an algorithm), and exception errors (arising
from unexpected conditions and events).
Glitches have become an integral part of computer culture and some phenomena
are perceived as glitches although they are not glitches in technical
terms. Artifacts that look like glitches do not always result from an error. What
users might perceive as “glitchy” can arise from a normally working function of
a program. Sometimes these might originate from technical limitations, such as
low image- processing speed or low bandwidth when displaying video. For example,
the codecs of some video- conferencing software, such as CU- Seeme,1 visibly
“pixelize” the image, allowing the compression of parts of the images that
remain static over different frames when, for instance, the transfer speed drops.
To comply with the customary usage of “glitch” we propose to think of
glitches as resulting from error, though in reality it might be diffi cult or impossible
to distinguish whether the particular glitch is planned or results from
a problem. To understand the roles glitches play in culture, knowing their origin
is not of primary importance. Understanding glitches as erroneous brings
more to a comprehension of their role than trying to give a clear defi nition that
would include or subordinate encoded glitches and glitches as malfunctions.
Glitches are usually regarded as marginal. In reality, glitches can be
claimed to be a manifestation of genuine software aesthetics. Let us look at
machine aesthetics as formed by functionality and dysfunctionality, and then
proceed to the concept of glitches as computing’s aesthetic core, as marks
of (dys)functions, (re)actions and (e)motions that are worked out in humancomputer
assemblages.
Computers do not have a recognizable or signifi cant aesthetic that possesses
some kind of authenticity and completeness. It is commonplace that the aesthetics
of software are largely adopted from other spheres, media, and conventions.
Thus, the desktop is a metaphor for a writing table, icons descend from
labels or images of objects, while the command line interface is inherited from
telegraph, teletype, and typewriter.
The aesthetics of computers that developed over a few decades from the
early 1950s to the early 1980s, when they were fi rst introduced to the public
and on to the current time (consisting of dynamic menus, mouse, pointer,
direct manipulation of objects on the screen, buttons, system sounds, human computer interaction models) are, in our opinion, not rich and self- suffi cient
enough to be called the aesthetic of the computer.
On top of that the current aesthetic of software is not complete; it does
not work very well as it does not contribute enough to the computer’s userfriendliness.
Besides, it is a widely acknowledged problem that the customary
information design principles of arranging computer data, derived from earlier
conventions (such as the treelike folder structure), result in users having
problem, with data archiving and the memorization of document names and
locations.
Historically, the shape, style, and decoration of every new technology has
been introduced in a manner owing much to the aesthetics and thinking customary
of the time. Thus, when mechanism had not yet replaced naturalism as
means of framing reality, Lewis Mumford argues, mechanisms were introduced
with organic symbols. For instance, a typical eighteenth century automaton, “the
clockwork Venus,” consisted of a female mannequin resting on top of a clockwork
mechanism.2 As technology developed further, some genuine machine
aesthetics were born, primarily derived from machine functionality. And it was
their functionality that some avant- garde movements of the twentieth century
admired in the machine. For instance, among the Russian avant- garde movements
of the beginning of the twentieth century (e.g., Cubo- Futurism, Abstractionism,
Rayonism, Suprematism) artists such as Mayakovsky, Gontcharova,
Kandinsky, Larionov, and Malevich poeticized new machines for their speed,
energy, and dynamics. The methods they used to depict movement, light,
power, and speed could be regarded aesthetically as grandparents of some of
today’s glitches (certain correlation of color mass; unlimited diversity of colors,
lines and forms; repeating geometrical structures, fi gures, lines, dots, etc.).
Rationalism and the precision of technical creation inspired many. Thus,
Meyerhold writes: “Arts should be based on scientifi c grounds.”3 Russian constructivists
such as Tatlin established a compositional organization based on
the kinetics of simple objects and complex ideas of movement—rotating inner
mechanisms and open structure, using “real” materials—all intended to
function for utilitarian use. Punin writes of Tatlin’s Tower: “Beneath our eyes
there is being solved the most complex problem of culture: utilitarian form
becomes pure creative form.”4
Functional machines, primarily built by engineers, established strong aesthetic
principles that have defi ned technological design for years. Functional
elements are later used as nonfunctional design elements that are appreciated as “beautiful” by users not least due to the cultural memory of their origin. For
instance, the curved part of the wing over the tire of some car models reproduces
the guards used in horse- driven vehicles and early automobiles to protect
users and vehicle from dust and to affi x lights onto. It does not carry any
advance in function, but is used in automobile design as a recognizable and
nostalgic element.
Today, the functionality of the computer is concealed inside the gray / white /
beige box that covers the cards, slots, motherboard, and wires. In modding5
these parts are reimagined as elements of visual richness that convey a symbolism.
Hardware elements are aestheticized: Users might install neon lights,
weird jumbo fans and colorful wires into a transparent computer case or even
build an entirely new one from scratch. Electronic boards jutting out at 90
degree angles and architectures of twisted wire are widely used, as in cinema
and design, to represent technical substances.
By contrast, the way data is presented on a hard drive is not human- readable.
It is stored in different segments of the disk and reassembled each time the
documents are retrieved according to a plan kept as a separate fi le. Software
functionality here is invisible and an interface is needed to use the machine.
Modern software almost always conceals its functionality behind the window.
It provides us instead with images such as a page fl ying from one folder to another,
an hourglass, or that of a gray line gradually being fi lled with color.
There are moments in the history of computer technology that are rich in
computer functionality producing distinct aesthetics. At such times, computer
functionality reveals itself through technological limitations. Bottlenecks, such
as processor speed, screen resolution, color depth, or network bandwidth—
4- bit, 8- bit music, 16- color pixelized visuals, slow rendering, compressed image
and video with artifacts—create an authentic computer aesthetics, that is,
the aesthetics of low- tech today.
There are vast contemporary 8- bit music communities (such as Micromusic
.net), based entirely on producing music on emulators or surviving models of
the early home computers of the 1980s, such as Atari or Commodore. Alongside
producing sine waves, the sound chips of such computers attempted to
simulate preexisting musical reality: guitar, percussion, piano. Imperfect and
restricted, the chips could only produce idiosyncratic, funny and easy to recognize
sounds which were far from the originals. Scarcity of means encouraged
a special aesthetics of musical low- tech: of coolness, romanticism and imperfection.
People making 8- bit music nowadays relate back to their childhoods’ favorite toys, memories that are shared by many people. Returning to a genuine
computer aesthetics of obsolete technology is not a question of individual
choice, but has the quality of a communal, social decision.
Functionality, as a characteristic of established machine aesthetics is always
chased by dysfunctionality (if not preceded by it). Functional machines,
robots, mechanized people (from Judaism’s Golem,6 Frankenstein’s monster7)
to the rebellious computers of the twentieth century) are interpreted as alien
to human nature, sooner or later becoming “evil” as they stop functioning
correctly. Thus, the dysfunctional mind, conduct, and vision become human,
compelling, sincere, meaningful, revelatory. As aesthetic principles, chance,
unplanned action, and uncommon behaviors were already central to European
and Russian literature of the nineteenth century in the work of writers such as
Balzac, Flaubert and Dostoyevsky.
In the technological era, society became organized according to the logic of
machines, conveyor belt principles, “rationally” based discrimination theories,
and war technology, with an increase in fear, frustration, refusal, and protest.
As a response, errors, inconsistencies of vision, of method, and of behavior become
popular modernist artistic methods used in Dadaism, Surrealism, and
other art movements. One of Surrealism’s declared predecessors, the Comte
de Lautréamont, provided us with the lasting phrase that something could be
as “beautiful as the chance encounter of a sewing machine and an umbrella on
a dissection table.”8 The introduction of chance, “hasard,” (fr.), subconsciousness,
and irrationality into art and life was seen as being both opposed to and
deeply embedded in rationality and functionality.
Dysfunctional machines are not only those that are broken (images and
fi gures of crashed cars and other mass produced imperfections fi gure in the
aesthetics of Fluxus and Pop Art); they are also those that do not comply with
the general logic of machines, by acting irrationally and sometimes even turning
into humans. Thus, at the end of the Soviet movie Adventures of Electronic
Boy (1977), a robotic boy starts crying and this emotion symbolizes that he has
become human.
A glitch is a singular dysfunctional event that allows insight beyond the
customary, omnipresent, and alien computer aesthetics. A glitch is a mess that
is a moment, a possibility to glance at software’s inner structure, whether it is
a mechanism of data compression or HTML code. Although a glitch does not
reveal the true functionality of the computer, it shows the ghostly conventionality
of the forms by which digital spaces are organized.

Glitches are produced by error and are usually not intended by humans.
As a not- entirely human- produced reality, its elements are not one- hundred
percent compatible with customary human logic, visual, sound, or behavioral
conventions of organizing and acting in space. Aesthetically some glitches
might inherit from avant- garde currents, but are not directly a product of the
latter (fi gure 8). Avant- garde artists inspired or disgusted by technology and
its societal infl uence have created a range of artistic responses, the aesthetics of
which today’s glitches strangely seem to comply with. A glitch reminds us of
our cultural experience at the same time as developing it by suggesting new
aesthetic forms.
A glitch is stunning. It appears as a temporary replacement of some boring
conventional surface; as a crazy and dangerous momentum (Will the computer
come back to “normal”? Will data be lost?) that breaks the expected fl ow. A
glitch is the loss of control. When the computer does the unexpected and goes
beyond the borders of the commonplace, changes the context, acts as if it is not
logical but profoundly irrational, behaves not in the way technology should, it releases the tension and hatred of the user toward an ever- functional but
uncomfortable machine.
Error sets free the irrational potential and works out the fundamental concepts and
forces that bind people and machines. An error [is] a sign of the absence of an ideal
functionality, whether it be understood in the technical, social or economic sense.9
As with every new aesthetic form, glitches are compelling for artists and
designers as well as regular users. Glitches are an important realm in electronic
and digital arts. Some artists focus on fi nding, saving, developing, and
conceptualizing glitches, and glitches form entire currents in sonic arts and creative
music making. For example, the Dutch- Belgian group Jodi are known
for their attention to all kinds of computer visual manifestations that go beyond
well- known interfaces. It’s enough only to look at their web- page http: //
wwwwwwwww.jodi.org to get a sense of their style (fi gure 9). On http: // text
.jodi.org a user browses through an endless sequence of pages that are obviously
of computer origin, and appear to be both meaningless and fascinatingly
beautiful.
Video gamers practice glitching (exploiting bugs in games).10 Game modifi
cations by Jodi, such as Untitled Game,11 as well as by other artists, such as
Joan Leandre’s (Retroyou) R / C and NostalG12 are achieved by altering parts
of the code of existing games (fi gure 10). The resulting games range from absurd
environments in which cars can be driven, but with a distinct tendency
to sometimes fl y into outer space, to messy visual environments one can hardly
navigate, but which reveal dazzling digital aesthetic qualities.
In his aPpRoPiRaTe! (fi gure 11) Sven Koenig exploits a bug found in a video
player that makes a video compression algorithm display itself.13 By deleting
or modifying key frames (an encoded movie does not contain all full frames but
a few key frames, the rest of the frames are saved as differences between key
frames) he manages to modify the entire fi lm without much effort. As a result we get excitingly distorted yet recognizable variants of videos popular in fi le
exchange networks, where such algorithms are widely used. And, of course,
with this much work already done for them in advance, we’ll see the power of
the new aesthetics of the glitch used in commercial products very soon.
Notes
1. Traces of CU- SeeMe can be found through http: // archive.org by searching for http: //
cu- seeme.com.
2. Lewis Mumford, Technics and Civilization, 52–55.
3. Vsevolod Meyerhold, “Artist of the Future,” in Hermitage, no. 6, 10.
4. Nikolay Punin, The Memorial to the Third International, 5.
5. See “case modifi cation” in Wikipedia: http:// en.wikipedia.org/ wiki/ Case_modifi cation/ .
6. For an excellent account of Golem, see: http: // en.wikipedia.org / wiki / Golem / .
7. Mary Shelley, Frankenstein.
8. Lautréamont, Les chants de Maldoror, Russian edition, 55.
9. Pit Schultz, “Jodi as a Software Culture.” in Tilman Baumgarten, ed. Install.exe,
Christoph Merian Verlag.
10. See “glitch” in Wikipedia: http: // en.wikipedia.org / wiki / Glitch / .
11. JODI, http: // wwwwwwwww.jodi.org / .
12. Joan Leandre (Retroyou), R / C and NostalG, http: // www.retroyou.org / and http: //
runme.org / project / +SOFTSFRAGILE / .
13. Sven Koenig, aPpRoPiRaTe!, http: // popmodernism.org / appropirate / .

No comments:

>> labels

>> cloudy with a chance of tags


Powered By:Blogger Widgets

>> ffwd

Loading...

followers

.........

My Photo
... is a Media Art historian and independent researcher. She is currently writing on "speculative archiving && experimental preservation of Media Art" and graduated from Prof. Oliver Grau's Media Art Histories program at the Danube University in Krems, Austria with a Master Thesis on Descriptive Metadata for Media Arts. For many years, she has been working in the field of archiving/documenting Media Art, recently at the Ludwig Boltzmann Institute for Media.Art.Research and before as the head of the Ars Electronica Futurelab's videostudio, where she created their archives and primarily worked with the archival material. She was teaching the Prehystories of New Media Class at the School of the Art Institute of Chicago (SAIC) and in the Media Art Histories program at the Danube University Krems.