I was just reading the Wikipedia entry for "deconstruction", the
post-modern theory of Derrida's, and I started thinking about my view
that all representations are 'text' on a computer because they are all
represented as data. I had mentioned that at an art demonstration and
I started thinking that images expressed as printed data are not
usually readable by human beings. But then I thought that if a color
image is expressed as a black and white image (and presented in an
appropriate field pattern for images) it would be. This is a form of
compression so it is a kind of abstraction that needs to be produced
by an algorithm. I then thought that all sorts of characteristics of
an image, like texture, could also be produced by other similar
algorithms to make them understandable to human beings. (Texture
defined as variations occurring within certain sized areas is an
example. My understanding is that a general algorithm to detect any
kind of texture would be much more complicated.)  (By the way, It is
interesting, in the context of this message, that the word 'texture'
starts with the word 'text'.)
None of this may be radically new or even interesting but I think the
recognition of a few of these concepts might be useful. This does not
involve the idea of presenting data to be human-readable, but thinking
of how abstraction might be used to produce recognition. First of all,
a useful abstraction might rely on an algorithm not only to get it out
of a data (or a 'text') but the data or some characteristic of the
data might need to be put through a transformation by the algorithm.
Although this is not radically different it is a new way of looking at
'abstraction.' Secondly, different kinds of transformative
abstractions of data-text might be needed to build a configuration of
abstractions that could then be used in recognition and subsequent
analyses. I think that is a new and interesting way of looking at the
concept of abstraction.
Jim Bromer

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T586df509299da774-Mc99e2ce822f4ac6665821c81
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to