Wondrous Nature

Imagine the wilderness ecosystem.

Big predators eat the weak of grazing herbivores. Scavengers chew those bones. Hiding rodents gather seeds until snatched by hawks. Worms blissfully migrate through the mulch; shrubs burst from their excrement.

Consider the ecosystem of which humanity is a part.

Some people build towers. Others farm and sell their produce to lawyers who battle over the definition of justice so that soldiers will act in such and such a way. All this is to ensure that the organism that is the state will have regular bowel movements and breathe easily and be fed despite its own complex environment: other states. States guarding their burrows with claws. States gathering into herds and packs. States hearing music played by prophets that make them rabid or docile. States who sell their their voices for food like minstrels.

There are men like ants who build hives to dig for iron. Men like spiders to weave webs to catch wanderers. Men like beavers to dam rivers. Men like proteins that spin strands of chemicals into new strands of chemicals. Chemicals made into laws or played as music in the courts of lions or passed to new enzymes to form the membrane of a spore. Spores that launch into the breeze, latch onto a dead tree, and grow from that richness of decay into a pale crust of life, inert until harvested and cooked in a broth by a historian or a pharmaceutical company or an oil rig.

When we conceive of ourselves apart from all of this, no wonder we are so often disappointed. We are blessed to be part of a wondrous Nature.

dangerous kind

You are
the most dangerous kind
don’t know what you want
and wanting it blind

You may
be the end of me
But when it began
you set fire to me

And I
protect myself by
keeping my distance
but tugging the line

And so
what we will soon get
are four empty hands
two sighs of regret

Storm

Please
don’t saddle me
with your polite company and chit chat ripples on the surface

I
would rather die
than live a life as weakly as the insulated chorus

What will it take to gather
your bursting hearts and souls prismatic?
we’ll grapple with the dreams that matter
pull them back from heaven to the earth

Unrelenting unrelenting
crusades against the automatic
wills on fire and painting spattered
music sleeting wrath and shining mirth

Please
don’t hesitate
we’re fading aging embers in a hearth that’s unprotected from the rain

And as thunder lacks duration
we must become the storm itself
if we are to outlast circumstance and reap the wealth of precious youth again

Heads Heads Tails Heads Tails Tails

heads heads tails heads tails tails

Experimenting with the ideas discussed here.

In class we are discussing the possibility of ‘information’ as a point of reduction across multiple domains. I’m the reductionist visigoth in the room.

This sort of art tries to prove something by breaking something.

Drawing is expressive in a way that complicates the ‘informational’ story. A product of a complex neurocomputational and manual process is going to appear more organic. Working with pure algorithm, stochasticity, and imagery maybe makes a stronger point. (Though there were ‘choices’ involved nevertheless.)

informational art

Just so that I have no direction to go but down, I’m going to go ahead and do the most pretentious thing that it is possible for a human being to do. That is to “define” a kind of art. In doing so, I’m trying to explain some of the thinking that goes into the drawings I’ve been posting here, and inviting others to use this idea to inspire them to try similar things. I would be very interested in seeing any work that is inspired by these ideas, or talking to anyone with similar thoughts on the matter.

Recall that conceptual art is art that is a concept or idea. The execution is perfunctory, the artifact is secondary. (I’m alluding to Joseph Kosuth here)

Concepts are generally not visually appealing in themselves. Though I would maintain that conceptual art can greatly enhance a physical space, some people have criticized conceptual art for this reason. Or, perhaps it is more accurate to say that a lot of people don’t like conceptual art because it is ugly.

With the digitalization of art, we have the option to operate at a different, deeper layer. Art can be defined by its information content.

This is different from saying that art can be data. In fact, most of the drawings on this site are just data. They are rendered for you by your monitors.

But information is something else. The definition of ‘information’ is up for grabs, but I am using it in a sense that is abstracted from Claude Shannon. Information is, in this conception, the revelation of an unknown (probabilistic) event. The page is loaded, the pixels revealed–what are they? Here they are. Information.

Information is a choice between options. It is a selection among differences. When the variable is repeated, the information can itself present a series of differences. Heads Heads Tails Heads Tails Tails.

However, since perception is bounded by context and attention, information can be reflected at multiple levels of abstraction for the same data. So, for example, there is compositional difference in the amount of brightness contrast in one side of an image or another. There is difference in texture at the collision of two brushes. There is a difference in color. There is a representation of something. There is abstraction. There is the destabilization of a representation into an abstraction. All of these events are possible within a single (static) image. (Even more is possible with animation, of course; it adds a whole new dimension of events.)

Perhaps this is similar to the idea that a single document can contain several narratives, nested and interlinked.

The most famous kind of informational art is the fractal, which is special because the information available at various levels of analysis is self-similar. Fractals are pretty precisely because they contain less information than you might think just by looking at them, and in so doing express recursivity as an event in the possible arrangement of information.

But there are a lot of other options. Exploration of this space of options could be called ‘informational art’.

What do you think?

Procedural drawings

color-procedural

(This is a thumbail. Click for full size.)

I recently got a tablet. Drawing seemed like a natural thing to do with it, so I searched the Android app store for a drawing tool and found Sketch n Draw. So I’ve discovered procedural drawing, which is apparently a thing.

The world needs a wikipedia article on the subject, but I think the term refers to drawing that is algorithmicly enhanced. So, you draw with a pen, and it algorithmically adds shading. It feels amazing. There is an on-line, all open source, client side JavaScript implementation called Harmony. Click that link and play. It’s worth it.

People have been messing around with it, sometimes impressively.

I’m pretty blown away by this. I’ve been trying to draw my whole life. It’s always frustrating when the limitations of a creative medium limit what you’re trying to express or put down. I’ve been experimenting with the Wacom/GIMP combination lately and enjoying the options there. I guess many of the pictures I’ve been posting here have been ‘procedural drawing’ in a sense because I’ve often painted with a brush that involves randomized coloring and/or sizing. Also, I’ve been using computer-generated gradients, etc….

What the hell, I guess I’ve been ‘procedurally drawing’ here for a while. But something feels different about these tools. Somebody’s designed an algorithm that fits brains better, I guess.

bw-procedural

This one is also a thumbnail.

The Case for Defunding NASA

However shall we fill our time?
Exquisite distances
between the lines

Three thousand years to get to outer space
all the time wishing we were
face to face

What makes a difference?
Is it information spat back to the earth?
Why not just use our lives
to celebrate the moment of our birth?

You can bring the olives
and I will bring some wine
he can cook some lentils
and she can bring petroleum and rice

and we will slay the fatted calf
and take a bath within its blood
as we recall that in the end
we’re mulch to mulch and mud to mud

Neoplatonic

I have looked so many years to see through to your soul
and sing to you a melody — the one to make it roll

If you are tired of those who wish to take from you by force
then we can find all that we need by sipping from the source

Just give me a few minutes as I talk to you alone
no friends no fakes no internet no links no telephone

so we can be
communion unafraid of unity
bare all for me
you only hide your greater beauty

beneath the interruptions of
our endless worldly chores
and those whose blind intrusions
make them unintended bores

who can really blame them
for seeking out your light

so goes the night

Notes

  • Dr. Specterman explained to me the progression of consciousness in The Phenomenology of Spirit. The first stage is animal, and so there is no contradiction with the empirical result that children (infants even) have empathy for other beings. Meanwhile, recognition that there is one essence does not mean the same thing as, for example, contemplative withdrawal into the Absolute. One can be fully engaged in ones own manifestation of it? I will lend her The Ethics of Ambiguity; I wonder what she will say about it.
  • Speaking of empathy, Dr. Parikh explained his interest in empathy-inducing technology and business. He did so tentatively, as appropriate when switching contexts from technical questions to questions of value or intent. Or…is this a switch? Should the norms of communication or evaluation be different? In what context is it acceptable to rigorously discuss ethics?
  • Technical discussions on an open source mailing list (for example) are passionate. The use of religious rhetoric (e.g., the “Holy War” between emacs and vim) is not accidental. Conversion happens, ideally through persuasion.
  • The ability for groups that initially disagree to come to consensus via internet discussion stands in opposition to David Weinberger’s proposal that disagreement on the internet is fractal and irresolvable. It is possible for one side to, in certain contexts, win over the others. More often, a synthesis of positions will emerge from active discussion by disagreeing parties. A sublation of opinions via dialectical reasoning. Though there is nothing constraining the number of parties to 2. Fractal pluralectic. (You heard it hear first!)
  • The politics of search engines. Two issues. First, ‘filter bubbles’ caused by an individual’s ability to find only what they want. Fragmentation of culture. What is the common ground for communication (lifeworld, lebensvelt) that is necessary for e.g. democracy? I think Anosognosia (wearing Kantian hat) still believes that this will be transcendental reason. I’m worried about “access” to transcendental reason. I.e., if participation in “directly democratic deliberation” or “consensus process” or civil society depends on transcendental ground, then doesn’t that condition civil society on Kantian education? Too weird.
  • Clues in David Graeber. Education as process of deferring authority, or power. Maybe in fractured culture the broadening of the lifeworld is empowering enough to legitimize it strategically despite lack of normative justification internal to the perspective of the subject?
  • The politics of search engines. Second issue. Algorithm transparency. Who is biasing what you are getting? Skepticism about media channels creates problems for credibility of information sources analogous to global skepticism problems for inference from sense data. Hard to see how transcendental rebuttals could apply. How about antifoundationalist ones? Non-parametric Bayesian ones? In general, the problem of who curates the curators.
  • Very hopeful about this coming summer. Dave Kush interested in sentiment analysis. Dwins interested in a non-maps side project. Last semester’s work highlighted some limitations of topic modeling with Latent Dirichlet allocation. I wonder whether something more inspired by Conceptual Role Semantics is possible. Promising application of Neo4j? Need to keep learning Scala. Twitter keeps open sourcing tools; becoming more and more impressed with them every day. Dr. Parikh says, encouragingly, “It’s run by a bunch of hippies.” Despite this, skepticism from without makes sense. Limited perspective. Other options?
  • To what extent is society a macrocosm of personality? Ultimately, convinced by Rachel McKinney’s argument that functionalism implies that it is in principle possible for mental categories to apply to social bodies. Dr. Specterman’s analysis of politics corresponds to this as well: we see revolt, both internally to psychology and in the political landscape, when a voice is unexpressed or unacknowledged. Creative solutions happen when there is an open space for expression. Given that assumption: Is the Buddy Roemer Twitter campaign the cries for attention by an oppressed voice? Dr. Specterman thinks that “we all already know” about the money in politics issue, but who are “we” to believe that our perspective is complete? Isn’t exclusion from debate a sign that this aspect of politics is not sufficiently conscious? Or is there such a thing as overdoing it–real pathology and egoism under the guise of needing to be heard?
  • Eddie Pickle wonders after “the physics of information” that would enable the design of, for example, a business model around open data. In our last conversation, was able to persuade him of the connection between open data marketplace and open code marketplace. Code is data–see LISP for extreme demonstration. I think: Code has logical depth (in Bennett’s sense). Excited for research this semester with Dr. Chuang on “Economics of Data and Computation.” It’s where a lot of the missing pieces are? Chuang has already posed fascinating and difficult (for me) question: how do the bounds on confidence level and computational complexity of machine learning on combinatorial optimization problems vary with the properties of the optimization function (continuous? discontinuous? multiple local maxima?)

animation study 1

animated-study-1-burn-in

Experimenting with animation. Unfortunately, image quality appears to deteriorate significantly with conversion to .gif. I don’t know if this has to do with the lack of alpha channel support in GIF, or just a reduced color range (compared to PNG, which is what most of the art posted here is).

Looking into it, there are apparently multiple animated PNG formats, APNG and MNG, with different browser support. APNG works for Firefox and (with a plugin) Chrome, MNG for Internet Explorer (with plugin). Safari supports neither? Sporadic (vendor-specific) support on mobile.

That’s too bad. I’ve had some hunches about what I guess you might call digital abstract expressionist animation but I didn’t expect to get hung up on browser compatibility issues.

Interestingly, others have been doing similar work, but using time lapse video to capture their painting. For example:

You could see how that method affords different constraints and opportunities.

I’ll probably work with APNG’s in the future, because IE is on its way out and Safari will likely catch up one day. Though I suppose that if I really wanted to be badass about it, I should try to work in HTML5 and Canvas.

But is it possible to do expressionist art in a code medium? My experience of creating these images is, it turns out, tied up in the experience of manipulating the image. At times the strokes themselves have been an emotional outlet.

Maybe a hybrid approach is possible: static PNGs for the raw imagery, and JavaScript for the animation between them. This wouldn’t be space efficient, since it would likely require storing each frame in its entirely, rather than just the visual difference in each from (which GIMP can created for you automatically in an optimization step).

I’d be interested to hear about anybody else’s thoughts or experiences on this.

Oh, and here is the still PNG of the image above, to give you a sense of how much it degraded when converted to GIF.

animation-study-1-still

This site uses a Hackadelic PlugIn, Hackadelic SEO Table Of Contents 1.6.0.