[-empyre-] FW: from Jim Andrews in reply to Greg Ulmer on propositional logic

[Brett Stalbaum]

Great conversation this month, I wish I had time to be-in-deep with  
you right now, the various demands of my good fortune to be employed  
standing in the way of pushing forward with you, data Aguirre's that  
you are. But I would like to offer a few ideas and precedent or two.

The question of "how artists have contributed and can contribute to  
dealing with inter-relations between very different datasets, as well  
as unexplored intra- relations within single large datasets of  
considerable complexity" was a central enterprise of C5 in the late  
90s early 2Ks, "the entire field of data mining and knowledge  
management, as we know it today, is predicated on the pre-existence of  
semantic models that allow data to be algorithmically mined for  
meaning. This is the basic philosophy and approach to data and  
information and is, of course, profoundly successful, but its  
application reaches severe limitations in dealing with contemporary  
data and the new kinds of problems it presents... standard information- 
processing techniques are not useful for all problems, especially when  
you do not have a question, when you have a poorly formed question, or  
when the dataset itself is not entirely understood or contains  
information potentials that were unplanned at the time it was  
collected... artists, who do not always participate in the scientific  
method, may well make discoveries or observations in their aesthetic  
and conceptual pursuits with such data that lead to such questions,  
even if the artists are participating as blind probe-heads in data  
space."

Semantic models, or we might today say Ontologies which are a kind of  
semantic model, were for us a problem to be avoided. C5's approach to  
the problem was to look for situations (in data in which)  
"composibility indicates the potential for autopoietic membranes  
existing as data relations via third-order structural coupling in a  
coded environment. This allows for the analysis of data sets where the  
semantic relationships are uncertain. In a sense, this idea can be  
described as the search for algorithms in which superficially  
different data sets might be shown to couple based on their subject- 
less form through inherent sans-semantic or pre- semantic models, and  
to seek these relations specifically to flag the potential for the  
presence of immanent, unplanned or otherwise unrecognized semantics  
flowing from mingled relations, thus revealing something about the  
ontology of the sets that produces new knowledge about them." (This  
stuff from an essay here: http://leoalmanac.org/journal/Vol_11/lea_v11_n05.txt)

Many works were produced out of this line of thinking both inside C5  
and outside by various members... in particular relation to Anna's  
project mentioned below I would refer readers to Jevbratt's Mapping  
the Web Infome project, in which various artists including myself were  
invited to use her web-crawler building software to accomplish aims  
that I think are very related to Anna's below. Currently, my research  
in the area is more closely related to applications of function  
approximations, AI techniques that are useful when you don't  
understand or only partially understand what the proposition (or  
cluster of propositions) that are uncertain and already immanent as  
system are autonomously proposing. This area too has many walls that I  
hit all the time! (Sometimes literally, as in the walls of a box  
canyon...) In that, I really have little so far to show from my  
investigations of these techniques over GIS data sets, although some  
other AI techniques (from gaming in particular) have been useful in  
the goal of finding ways that allow data about landscapes to produce  
experiences that return to the sites of their representation. Still,  
there is something much bigger out there, I feel.

In any case, I nevertheless continue to assume that there is a  
productive space for perhaps limitless research for artists like Anna  
who are mingling data sets with data sets, sifting through them for  
relations, and uncovering their behavior by activating them in ways  
that they were not originally intended to be activated. Bravo! it is  
in a way a very romantic enterprise. I once wrote elsewhere,  
paraphrasing, for every data Norgay/Hillary there will be many more  
data George Mallorys and data Donner parties. To jump into this field,  
you have to be open to a certain amount of resignation to this, no?

Cheers all,
Brett Stalbaum

On Oct 28, 2009, at 4:32 PM, Anna Munster wrote:

> i expect that you don't mean 'propositional logic', which is more or  
> less
> synonymous with 'logic', the entire field of logic. perhaps it is sql
> (standard query language) that you are thinking of? sql is the  
> language we
> typically use to query databases. like when we do a google search.  
> we can
> use 'and', 'or', 'not', and perhaps other binary and unary operators  
> to form
> queries that fetch records with the properties specified in the  
> query. a
> query can be viewed as a proposition that evaluates to a record-set  
> from the
> queried relational database. so it is indeed a propositional logic.  
> "a" is
> the key, as in 'a particular one' (sql, that is). it's interesting  
> to try to
> think of expanded query languages that do useful things that sql can't
> accomplish. sitting here thinking about it for a few minutes, i find  
> that my
> ideas are still doable within sql if only the records have the proper
> structure (i.e., the relevant fields) to form the right sql  
> proposition and
> thereby the correct record-set returned. but some sort of similar  
> structure
> would have had to have been there to get the same results even were a
> different query language used, i.e., the info has to be in the  
> records if
> the database is going to be capable of returning the desired record
> set--regardless of what query language is used. so it may be that  
> sql is not
> the issue, i.e., propositional logic is not the key to breakthrough  
> but,
> instead, the manner in which the records are structured and,  
> therefore,
> queriable.
>
> the subject of ontologies in computer science looks very  
> interesting. how do
> you structure a world view so that interesting, programmatic  
> thinking can be
> done with it? tasty. i imagine it's pretty much up for grabs at the  
> mo.
> there isn't likely going to be significant standardization on this  
> sort of
> issue but, instead, multiplication of structures designed for  
> particular
> purposes, ie, ontologies *of* this or that. though, no doubt, things  
> such as
> category theory, which would plainly be useful in structuring *any*
> ontology, will inform the deeper efforts.
>
> i'm creating a project at the mo that uses datamining, to some  
> extent. it's
> at http://vispo.com/dbcinema . i'm writing a graphic synthesizer and
> langu(im)age processor. it's a langu(im)age processor in that you  
> create
> brushes and assign each brush a concept. a string you type in. the  
> brush
> then does a google image search and retrieves images somehow related  
> to the
> concept. dbcinema then uses those images as 'paint'. one of the  
> problems in
> this sort of work is basically the matter of collage. like, ok, you  
> get a
> bizillion related images, very cool, now what r yu goin to do--make a
> same-old out of em? more trash piled higher deeper? so the main  
> pressing
> issues here are not really network-related because that's happnin  
> big time
> ao friggin k. the questions concern what you can do with the record- 
> set
> (sets of images, in this case) once you got em. i mean something  
> that really
> flips the lid. at this point, you look to generative art to see what  
> can be
> done with pixels on the fly. and the answer to that is fill yer  
> boots with
> knowledge of masking, opacity, dynamic motion, exotic curves, color  
> sets,
> dynamic gradients, graphical transformations, and the whole  
> reportoire of
> abstract generative visual art. this provides plenty of gaaaaaaaaaa  
> and lid
> flipping dynamite that just hasn't been seen before, or can, if you  
> really
> go to town on it. that's a mix of tech issues but also artistic  
> issues. the
> artistic issues are not only 'how do you make it look good' but  
> also, just
> as importantly, what sort of narrativity can emerge in this  
> situation that
> could easily degenerate into mindless remix? that's a subtle issue  
> which is
> extrordinarily tasty and part of the deeper artistic achievement that
> beckons.
>
> ja
> http://vispo.com
>
>
>
>>> Also, how is propositional logic "doing for contemporary datamining
>>> efforts
>>> what Aristotle's animism did for physics leading up to the  
>>> scientific
>>> revolution (preventing the breakthrough)".
>>>
>> I would love for there to be a wiki devoted to this question (perhaps
>> sponsored by Turbulence.org?).  I worked this past year with some
>> computer scientists of various specializations on a textmining  
>> project,
>> which motivated me to look further into Semantic Web Ontologies, with
>> "ontology" being a boundary object (we were using the term, but with
>> somewhat different meanings and referents).  I read up on RDF  
>> (Resource
>> Description Framework) triples, and recognized them immediately as  
>> right
>> out of Aristotle's topical logic:  subject predicate object.  Of  
>> course
>> this structure is native to Indo-European language, and as Nietzsche
>> said, God (or metaphysics) is in the Grammar. Maybe this can be  
>> fixed at
>> the software level? (hardware switches, like truth tables, are on/ 
>> off,
>> but Wittgenstein abandoned truth tables for good reason). The  
>> textmining
>> project responded to a grant opportunity calling for new departures  
>> in
>> datamining.  The motivation is that current approaches, as powerful  
>> as
>> they are, are not adequate to the scale of information overload
>> confronting us. We have the cyberspace problem as Gibson dramatized  
>> it
>> for us:  If our databases could give us total information in an  
>> instant,
>> we would not be able to receive it (hence Gibson's metaphor:  Tell My
>> Horse). The point of apparatus theory is that the solution to this
>> problem will have been not only technical, but also philosophical,
>> aesthetic and so forth.  There are multiple invention streams pouring
>> into the vortex of our mobile moment.
>
>
> _______________________________________________
> empyre forum
> empyre at lists.cofa.unsw.edu.au
> http://www.subtle.net/empyre


--
Brett Stalbaum, Lecturer, LSOE
Coordinator, Interdisciplinary Computing and the Arts Major (ICAM)
UNIVERSITY OF CALIFORNIA, SAN DIEGO
Department of Visual Arts
9500 GILMAN DR. # 0084
La Jolla CA 92093-0084
http://www.walkingtools.net

General ICAM advising and ICAM40/VIS 40 office hours for Fall 2009  
(note: they change each quarter...)
Wednesdays, 1-3PM, Mandeville 221 (Near Vis Arts Advising)