Steve Land's Brain
Unformed thoughts are much more interesting than hardened opinions
















 

One Thing

If you think of how you'd accomplish something with electronics and computing technology, you begin to think about subcomponents and parts. For instance, if I wanted my computer to "see", I'd go get a Web camera and plug it in to the USB port. The camera itself is a device totally separate from the computer, the USB subsystem is a combination of hardware and generic drivers that work in combination, and the computer "sees" by virtue of software that can interpret incoming information from this camera.

All the pieces are separate, and software makes them work together.

On first blush, you might say that human vision follows this fairly closely. Eyes equal camera. Retina equals USB. Brain equals software and device drivers.

Science has concluded that the boundaries are not so sharp. The eye is an extension of the brain; the retina feeds directly to the brain. The structures in your eye that interpret light immediately begin processing, even before the brain proper becomes aware of the input. The visual cortex in the brain can also be said to be an extension of the eyes. The eye/brain dichotomy breaks down, and we are left with something better described as an "eye-brain continuum". You can't find pieces. Instead, you can find aspects of the structure within the whole that are specialized for different tasks: some cells process color, others process detail, others process dark and light, etc. 

In nature, the boundaries between one thing and another are often semantic constructs that we have imposed on things. A tree, for example, exists by virtue of its climate, rain, sun, and soil. If you change any of these contextual factors, the One Thing we call a tree will cease to exist. So, the tree might better be called a "tree-ecosystem continuum".

We generally give new names to things that have different forms, even if the things are part of a larger whole. The fact that we have words for something makes us think that the thing is a stand-alone item that can be separated from its context. A tree has named bits like: bark, roots, leaves, branches, twigs, etc. While you can remove some of these things without destroying the whole tree, you cannot easily replace them or exchange them with other bits. Even if you could remove the top of a tree and graft it on to a different tree, you could not claim that the grafted tree is "the same" as the original.

When we design artifacts that are "modular" or "subcomponents", we intentionally create smaller units that are assembled into a more complex system. A screw may be a small component of something (and, parenthetically, might exist in its current form because of standard screw heads, threading sizes, etc). Several screws combine with other stuff to create an assembled object. The final product might work as a whole, but the subcomponents, including the screws, are not an integral part of the whole. They are replaceable parts. You can replace all the screws with different ones having the same thread and size, and the whole object will still be intact. You can't do that with leaves of a tree.

This notion relates with my Labels and Things entry, but I wanted to expand it a bit.

The way we think about the world-- in terms of parts working together as whole systems-- may be misleading compared with how real things actually form and interoperate. A tree does not enlist a bunch of leaves to help it achieve its goal of transforming sunlight into energy; the tree IS a bunch of leaves. A group of cells in a growing tree morph from branch-growing cells into leaf-growing cells, based on a context that includes the tree's own genetic pattern.

I personally have hands, arms, shoulders, and other innumerable named bits. I don't think of these as helpful subcomponents: these are part of me.

The way industry produces manufactured goods, and the way software is manufactured, presupposes a concept of subcomponents and assembly that does not reflect how nature really works to create systems.


Click here to visit the Radio UserLand website.
Click to see the XML version of this web page.
Click here to send an email to the editor of this weblog.
© Copyright 2005 Steve Land.
Last update: 4/21/2005; 8:22:47 AM.