title = "Attention, study and pattern recognition"
glassblowers = ["tomislavmedak.md"]
+++
There's a drive in modern culture to collect culture, create definitive collections of specific historical artefacts, assemble a complete historical record of past events. That's the drive of a museum or an archive. There's an equal drive to create idiosyncratic collections reflecting networks of relata, readers, readings, resemblances, reuses and recontextualizations of expressions, forms and symbols. That's a cabinet of curiosities, a Mnemosyne Atlas, a UbuWeb of one sort or another.
These procedures of collection ground, situate and sometimes blind our understanding of history. However, the collections they create cannot be comprehended by attention we muster in our media diet and do not fall under the strictures of attention economy. They rather require methods, even if idiosyncratic ones. They require the capacity to study, to draw out what lies in hiding between the artefacts.
Machine learning algorithms are trained to recognize patterns in large sets of data. They are engineered to draw out what lies in hiding. Thus, they've been trained in cultural collections. They are extracting potentialities that lie in the historical records and converting them into generated material, with their developers and owners seeking to find applications – and sometimes generate value under the strictures of the attention economy.
Kenneth, in his , was retelling the anecdote just before I made the first partial copy of the media files of UbuWeb to be served from the Memory of the World server. While he remembers his fears and doubts, I remember clearly the technical reasons: how hard disks dying somewhere in Mexico and how the death of a hard disks immedieatly led to the conceptualization of the death of UbuWeb by Kenneth. That part about conceptual death didn’t end up in a book and fears and doubts are described this way: “I wanted to throw in the towel, [..] that perhaps it had come time to wave the white flag. I was getting burnt out and couldn’t see why I should keep on doing this.”. I am not sure if Kenneth remembers his concept impro on how slowness of that hard disk death is, maybe, the right way for UbuWeb to pass away. How people will start to discover that parts of the UbuWeb are not anymore available and by time UbuWeb will lead itself to the slow excintiction from the internet. The slow death as it dies of old age.
Kenneth, in his , was retelling the anecdote just before I made the first partial copy of the media files of UbuWeb to be served from the Memory of the World server. While he remembers his fears and doubts, I remember clearly the technical reasons: how hard disks dying somewhere in Mexico and how the death of a hard disks immedieatly led to the conceptualization of the death of UbuWeb by Kenneth. That part about conceptual death didn’t end up in a book and fears and doubts are described this way:
> I wanted to throw in the towel, [..] that perhaps it had come time to wave the white flag. I was getting burnt out and couldn’t see why I should keep on doing this.”.
I am not sure if Kenneth remembers his concept impro on how slowness of that hard disk death is, maybe, the right way for UbuWeb to pass away. How people will start to discover that parts of the UbuWeb are not anymore available and by time UbuWeb will lead itself to the slow excintiction from the internet. The slow death as it dies of old age.
Kenneth’s narration/conceptualization of the UbuWeb death strongly reminded me on my favourite science fiction novel “Bicentennial man” by Isaac Asimov (in the Robot series). In the novel a character named Andrew Martin starts his living as a robot in a household. He was from the very first series of robots from househould. After some time they found out that he had some software bugs and that was fixed for all newer series of robots. It could be that Andrew was different because of those bugs. He was creative. He lived much longer than his hosting family and as they would die he felt he is not a robot without feelings. At certain point of his life he started to work on his own recognition as human. He felt he thinks freely, feels and contribute to the society so he should be recognized as one of us, humans. Humans of his time didn’t make his mission easy. He had to lie to another robot-surgeon that he is robot so the surgeon would do an operation which irreversibly made Andrew age and die. When humans recognized that Andrew decided to die in order to get recognized as human they recognized him as a human. Eventually Andrew died.
Generative artificial intelligence produces overabundant, individualized, non-generalizable content. Generated text, music and film served to you will not be the same as the generated text, music and film served to me. Frames of reference at the point of composition and at the point of reception will necessarily be different. Generated from individualized prompts and profiles, such forms of content we will struggle to have a water-tank conversation about. Details will not match and be commensurable between our accounts. Who knows, maybe the mismatches might give a new lease of life to water-tank conversations, just as recommendations of our differing media diet do in the era of streamed content.
Historical record in training datasets can be chronologically and contextually vast. However, once processed into collocational frequencies, stripped of the specificity of their provenance, the singularity of historical events and their readings might get lost.
Changes brought about by the era of machine generated text, music and film will strain our capacity to make and share meaning from culture.
