[-empyre-] glitch device/divide

Fri Dec 9 02:39:02 EST 2011

Hi Curt,

..on Thu, Dec 08, 2011 at 05:34:05AM -0500, Curt Cloninger wrote:
>
> It seems like you are wanting to limit the discussion of glitches to  
> occurrences that happen at the hardware and software level, within  
> machines. But humans can (and frequently do) glitch as well,  
> particularly in response to media which their bodies receive as  
> glitched. In one sense, phenomenological sculpture, op art, and  
> structural film are all about trying to get a human body to have a  
> glitched experience. I don't have to read binary to have a glitched  
> experience. I don't have to be a programmer (although I myself  
> occasionally program) to receive an affective, bodily or linguistic  
> glitch effect from media sent to me via a machine. I don't have to  
> understand compression schemes at the binary level in order to  
> affectively eperience the phenomenological differences between various 
> compression schemes as my body is exposed to them.

I'm not limiting the breadth of the term, I'm leaving it just as it is; 'glitch'
is a term derived from the culture of electronics, circuitry in particular. Only
recently has it come to be applied to software at all, possibly by way of errors
writing and reading from bad physical or temporary memory.

Glitches specifically relate to systems and machines, things we have designed.
They express something we don't yet know about something we've made, a potential
born of failure.

Positioning glitch outside of Engineering seems unproductive and/or
opportunistic; animal kind like us don't experience bodily, cognitive or
behavioural 'glitches', rather lapses of judgement, unintended behaviour due to
nervous stimuli (confusion or stress) and illness due to mutation or failure of
organic parts. 

A glitch is a brief, sometimes recoverable fault in a circuit, system or
machine. We are only any of these things if you yield entirely to Cybernetic
metaphor. Most importantly however, we are not systems of our own design. 

> It's a bit like talking about a Rothko painting in terms of whether he 
> used horsehair or synthetic brushes. Such discussions are always  
> possible, but they  are only obliquely related to the things which are 
> most interesting about art, and they usually dead-end fairly quickly. He 
> either did or he didn't use horsehair brushes. It either is or isn't a 
> true glitch.
>
> Mez Breeze's new book is called "Human Readable Messages." So perhaps a 
> distinction needs to be made between human-readable glitches and  
> machine-readable glitches. Human-readable glitches are media glitches -- 
> they occur when humans are communicating to humans through machines which 
> mediate this communication. Human-readable glitches don't freak machines 
> out. But neither do machine-readable glitches. Machine-readable glitches 
> may crash machines, but that doesn't freak machines out, because machines 
> have no sentient expectation of "normal." Machines lack the ability to 
> have an uncanny experience. So to limit the discussion of glitches to 
> events that only happen within machinic systems, glitches which never run 
> on or involve human bodies, is to talk about something quite limited. 
> Because a machine can't know or experience a glitch. Only a human can.

A phenomenological refactoring of glitch as being only existent within
experience, therefore dependent upon -and expressed through the human- seems
pretty hairy and/or confusing to me. What would be gained by glitching out with
Heidegger or Merleau-Ponty?

> You imply that true, machine/code-level glitch-event ruptures have  
> radical, political, Latourean-entangled impact on human cultures. I am 
> curious how this happens without them also intersecting networks and 
> scales of human bodies, human-readable forms, and human aesthetics. Does 
> this radical impact happen metaphorically (machinic glitch as a mere 
> symbol for actant agency) or directly (glitches in stock market networks 
> crashing third world economies)?

I didn't say that glitches have "Latourean-entangled impact on human cultures"
rather that they create an entry point, a de-punctualisation, within what is
otherwise a black box, an inaccessible prosthesis or system. In this way a
glitch can express potential for further knowledge, increased subjectivity and
freedom of movement.

In a time of iPads, Sentiment Analysis, internet enabled fridges and SkyHook
Wireless we know less and less as to the inner workings of the machines and
systems we use and/or depend upon. The glitch can seed a vital techno-political
subjectivity, an important break from what I refer to in a current paper as an
Ideology of Seamlessness; of a continuous, hermetic and harmonic interaction
between parts in the conveyance of intent.

> You assert that, "some of the most potent and transformative glitches in 
> technological history are quite boring to behold. To most, they'd  
> probably go unnoticed." It would help me to better understand what you 
> are claiming if you would list some of these historically transformative 
> glitches specifically.

There are many examples out on the internet, here are a few that are quite
famous.

On February 25, 1991 the system clock on a patriot missile glitched, putting it
out of step by 1/3rd of a second. It crashed into a compound in Dhahran, killing
28 Americans.

The 'smart ship' USS Yorktown simply stopped out in the ocean in 1997 due to a
divide by zero error. This glitch itself wasn't seen, no one knew what had
happened, but analysis of the issue triggered deep inspection of seaborne
auto-piloting systems world wide. 

5 people died due to excessive X-Ray exposure in the 80s due to a bug in the
Therac-25 radiation machine.

In 1996 the European Space Agency's Ariane 5 Flight 501 self destructed due to a
glitch in the guidance software.

Again, a glitch itself doesn't have to be seen for it to have been highly
significant. No one saw the glitch at work in Flight 501, buried deep in the
guidance software subsystem. They saw the craft explode as a consequence of the
glitch.  Perhaps no one even saw the explosion, but a statistical representation
of it a second after all their comms went down.

> Also, I'm curious to hear your response to Jon Cates' previous post  
> regarding John Cage's prepared systems. Can prepared aleatoric systems, 
> oulipian (constraint-based) systems, and other human-orchestrated systems 
> still lead to uncanny outcomes? It seems to me they can, for humans. 
> Perhaps the outcomes are not uncanny to the systems themselves; but 
> again, a system can't experience itself as normal or uncanny.

I'm not sure really. I most certainly don't think one can design glitches,
merely encourage them or work with them. Parametric control of a glitch would be
an oxymoron in that one cannot choreograph an accident, only create conditions
such that one is likely to occur. 

For me a great glitch is unanticipated, devastating, wild.

Cheers,

Julian

> Julian:
>
>> Indeed it is what is relevant, as it is with any cultural trope. It's here
>> however that software developers like myself find ourselves cynical  
>> about Glitch
>> Art precisely because we know that what we're often looking at/listening to is
>> not a glitch, rather an event designed to have the appearance of one.
>>
>> A glitch-concert using Max MSP is not glitch, rather the application  
>> of digital
>> synthesis to mimic sounds that sound like what we understand to be glitch,
>> namely electrical sparks, servos breaking under load, etc.  Similarly, someone
>> playing with GTK or Quartz Composer to manipulate a desktop interface 
>> such that
>> it performs unexpectedly isn't glitch, it's UX/UI design.
>>
>> This leads us to the question "Can you design a glitch?". Perhaps you can only
>> design /with/ glitches, not glitches themselves..
>>
>> If glitches are political at all it's in because they represent a possible
>> entry-point within an otherwise closed system, a 'de-punctualisation' (from
>> Latour) of the Black Box. What many call glitches are in fact just the 
>> beginning
>> of what later becomes an exploit (whether that be jailbreaking a device or
>> injecting malicious code into a process running on a server). In this way
>> glitches signal the possibility of further action; an opening, they express
>> freedom of movement.
>>
>> Purely aesthetic fetishising of glitch depreciates this potential, I think.
>> After all, some of the most potent and transformative glitches in  
>> technological
>> history are quite boring to behold. To most, they'd probably go unnoticed.
>>
>> Cheers,
>>
>> Julian
> _______________________________________________
> empyre forum
> empyre at lists.cofa.unsw.edu.au
> http://www.subtle.net/empyre

-- 
Julian Oliver
http://julianoliver.com
http://criticalengineering.org