Fundamental Compressionist Philosophy.

[Andrew Stanworth]

Subject: Fundamental Compressionist Philosophy.

In reply to Gerry,

| OK, I'm guilty :-) on two counts: 1) being an academic, 2) coining the
term
| "Computing as Compression".
|
| In defence of the latter, I chose 'as' rather than 'is' because so many
| people, including the highly respected Alan Turing, have not seen
computing
| in these terms.

That is a very academic way of seeing things ; )  I admit to being a rat
that's abandoned the academic ship.  A very fat, floating, rat that intends
to sit back and say 'this is the way the Titanic ought to have been built'
in a document
on my site, hopefully in the immediate future.  My personal views relating
to academia aside, my own feeling is that 'new' ideas need some
evangelising - and casc/cisc falls into that category - if nobody shouts
loud enough, who is going to hear about it above all of the noise?

I am grateful to Alan Turing - the idea of a universal basis to computation
is one of the two sexiest philosophical ideas I have ever come across -
Parmenides furnishing the other.  I liked Turing's idea so much that I
wanted to know the identity of the component(s) that actually made universal
computers, universal.  Although in my original posting on this topic I
stated that all universal computers have two elementary components, an
output device and a switching device, in truth there is only one device,
'the two input-state and two output-state switch', since all fixed output
can be reduced to a hardwired switch where both output states are set to be
identical.

A further point of note, of which I am fully aware, is that a two-input and
two-output state switch is the ultimate in reductionism (i.e. the
fundamental 'atom') in physics as well - or, more correctly, is a different
way of describing the same experience.  To understand this, consider that
'change' is the central construct around which all conceptions of our
understanding of reality are based.  An understanding of the nature of
change itself, directly gives rise to the concepts of matter, space, time,
energy and consciousness.  The elucidation of this is a follows - change
must have 'something' to change, which gives rise to the concept of
'matter',  matter must exist 'somewhere', which gives rise to the concept of
'space', the difference between one change and another gives rise to the
concept of 'time', the impetus which actually effects the change gives rise
to the concept of 'energy', the awareness of 'what' exists 'where' and 'how'
it should be changed gives rise to  'consciousness'.

Now, try asking a physicist what 'time' actually is and how it works, or
what 'space' actually is and how that works, or what 'matter' actually is
and how that works, or what 'energy' actually is and how that works, or what
'consciousness' actually is and how that works.  I guarantee that no
physicist will be able to deliver a satisfactory answer that defines any of
these concepts in and of itself, without recourse to circular reasoning -
though they use them every day and would have no subject without them.  The
truth is that nobody knows where these things come from, or how they work,
or why they work.  They are just present in our experience as facets of the
changing environment we inhabit - our thoughts have no existence at the
deeper level necessary to penetrate these mysteries.

If we should now consider the humble concept of the 'switch', I can
demonstrate (albeit awkwardly in such a small space) that it is also a valid
description of the aspects of 'change' - change must have 'something' to
change, which gives rise to the concept of 'output', the difference between
one change and another still gives rise to the concept of 'time', the
impetus which actually effects the change still gives rise to the concept of
'energy', the awareness of 'what' exists 'where' and how it should be
changed still gives rise to a rudimentary level of 'consciousness' (and
sounds a hell of a lot like a switch to me!!) but now also allows room to
include the concept of 'input'.  'Space', you may have noticed has been
missed out, but this is still consequence of 'something' which is now 'input
& output', existing 'somewhere' which is 'space'.

| There is usually more than way of looking at any topic so I
| feel it is safer to use a title that acknowledges that there may be other
| ways of viewing the same area of interest. This is "future proof" because
it
| means that other theories can be put forward as alternative views without
| the need to 'destroy' older views. Quantum theory and relativity are
| alternative views of the physical world that rub along quite happily, each
| with their own strengths and weaknesses. Why not the same for alternative
| theories of computing?

My view, not incompatible with this, is that the only valid 'truth' is
utility, or fitness to satisfy an intended purpose.  In other words, if it
works use it, if doesn't work don't use it, if it only works sometimes then
only use it sometimes.  Thus, to my mind, in your analogy Newtonian physics
shouldn't be left out of the picture.  More interestingly - I contend that
just as the physicists may describe their experiences of reality in terms of
fundamental 'particles' a computer scientist may do the same but use
fundamental 'switches' - the big realisation is that both descriptions are
functionally identical - you could say that they are both 'universal
computers' for mapping the universe.  The one using mass and energy, the
other using input/output and energy.

| Rather than using a 'switch', I have borrowed an idea that seems to lie at
| the heart of most of the standard techniques for information compression:
| any relatively long repeating pattern can be replaced by a relatively
short
| 'identifier', 'code',  'name' or 'reference'.

You are not the only respondee that seems not to want to talk about my
switches ; )

Undaunted. I would like to propose three equations:-

    a) 'a functioning identifier' = 'data' + 'a system of non-identical
output-state switches'

    b) 'pure data element' = 'all outputs of a switch existing in the same
state'

    c) 'a functioning identifier' = 'a system of same-output-state switches'
+ 'a system of non-identical output-state switches'

| The neat thing about using this traditional idea is that decompression can
| be done in exactly the same way as the original process of compression:
find
| patterns that match each other and then merge or 'unify' the matching
| patterns. In the original compression, it is the relatively long patterns
| that are matched and unified. In the process of decompression, it is the
| code patterns that are matched and unified.

It seems that you are describing how it (i.e., casc) works, and I am
describing what 'it' (i.e., cisc) actually is.  My main reason for joining
this list is that it seems clear that all of the people here intuitively
grasp the 'bigness' of the casc/cisc idea (apart from the fact that I have
somewhat of an intellectual fetish for compaction - there, I've admitted
it!!) and my theorising would seem to provide a concrete set of foundations
to build from (which is surely beneficial to validate the convictions of all
here)  - i.e., my point is that, in operation, the fundamental unit which
makes information processing systems 'universal' can be described in terms
of making effective use of available resources (i.e., through the twin
processes of the compaction & the concommitant extraction of compacted
information - or to give it another description 'evolution and natural
selection').

| This is most fully explained in my articles and reports about natural
| language processing
(http://www.sees.bangor.ac.uk/~gerry/NL_processing.htm)

I shall nosey around them : )

Thanks for your thoughts (switches!?!),

Andrew.
e-mail: andrew at seaca.org
site: www.seaca.org