Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

SUO: RE: Discussion Period on Motion by Matthew West




Dear Jon,

See comments below.


Matthew West
Principal Consultant
Shell Information Technology International Limited
Shell Centre, London SE1 7NA, United Kingdom

Tel: +44 20 7934 4490 Other Tel: +44 7796 336538
Email: matthew.west@shell.com
Internet: http://www.shell.com


> -----Original Message-----
> From: Jon Awbrey [mailto:jawbrey@att.net]
> Sent: 02 September 2003 16:22
> To: West, Matthew R SITI-ITPSIE
> Cc: SUO
> Subject: Re: Discussion Period on Motion by Matthew West
> 
> 
> o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
> 
> JA = Jon Awbrey
> MW = Matthew West
> 
> JA: When I think about these things in systems theory terms,
>     it always starts with some state space X that contains
>     all of the logically possible states of the system you
>     want to think about.  Your first stab at dealing with
>     process will then be in terms of a "path" p : R -> X,
>     where R is a real line that you think of as the time
>     domain.  Physical laws and practical constraints are
>     then expressed as subsets of X and subsets of the
>     set {p : R -> X} of all possible paths through X.
> 
> MW: The states you [describe?] are essentially complex
>     properties of a system.
> 
> JA: This is where I begin to have trouble making translations
>     into a language that I understand.  Let me think about a
>     state x in a state space X -- some people will be fussy
>     here and want to say "configuration space", but because
>     I start with non-deterministic systems from the outset
>     I don't make that distinction.
> 
> MW: So far so good.
> 
> JA: One thing that we are talking about when we say "the system"
>     is a thing that can be in various states, and a thing that
>     can pass through a line or a sequence of states.
> 
> MW: Agreed.
> 
> JA: In that case, we can say that the state is a complex 
> property of this
>     system-thing.  We say that the system is in the state x 
> in the set X,
>     and there are projection maps p_j : X -> R, just taking R 
> as a typical
>     example, and that we think of as "measuring" x in X with 
> regard to the
>     property corresponding to p_j, and giving us a real number.
> 
> MW: I think I'm still with you.
> 
> JA: So I think I have a way of understanding the statement that
>     a state is just a complex property of a system.  Some people
>     will call that system-thing the "reified system".
>  
> MW: It is that which takes on the state.
> 
> Yes, the picky point that I think some people are trying to 
> make is that
> the data of experience or the results of measurement are what 
> we really
> have on hand, whereas the system-thing is in the bush, as it were, at
> some remove from immediate impressions, the object of possibly many
> and possibly competing hypotheses that we form to explain why the
> data are as they are.

MW: I'm not one of them, as you have probably guessed. As noted below
you can have any individual object you want as long as you can 
demonstrate it has a spatio-temporal extent.
> 
> JA: They will say that it is nothing but a "hypostatic 
> abstraction" that
>     personifies the measurements we make and that enter our experience
>     and that are the only real things that we ever know of 
> this system,
>     but I don't see the harm of positing such objects, so long as we
>     stay conscious of what we are doing.
> 
> MW: One of the things I like about 4-dimensionalism is that 
> it is very unfussy
>     about what objects there are.  If it is some extent in 
> space-time, you can
>     have it.  Why you are interested is up to you.  So we are 
> allowed systems.
> 
> Seems eminently reasonable to me.
> 
> MW: The time chunk of the system that possesses the state is 
> the relevant
>     possible individual, that is itself a temporal part of 
> the whole life
>     of the system.
> 
> JA: Almost.  Let me try.  Instead of a reified-system-at-an-instant,
>     the thing that is in the state x in X, you want to talk about a
>     "time chunk" as the thing that has the properties p_j that afford
>     a basis for defining the state x in X, along with all of the other
>     properties that the system has at that moment?
> 
> MW: Sounds good.  In the limit it can be reified-system-at-an-instant.
> 
> Check.
> 
> JA: In practice, both empirical and theoretical practice,
>     we commonly approximate continuous trajectories with
>     discrete sequences of states, and computational work
>     pretty much forces us to start with discrete and even
>     finite cases, so it's best to have concepts that cover
>     both indifferently, at least, until push comes to shove.
> 
> MW: Indeed.  I think we have this covered.
> 
> Whew!
> 
> JA: Given the variety of different ideas about data models
>     that we've seen, it might also help if you could say
>     a few words about how you see them.
> 
> MW: The structure and meaning of data, is the few words version.
>     A longer version can be found in:
> 
>     http://www.matthew-west.org.uk/Documents/princ03.pdf
> 
> JA: I could not download or read this successfully,
>     and I think my Adobe reader is pretty current.
>     Do you have another format?
>  
> MW: The site seems to have been having some delays which have made
>     connecting and downloading difficult.  It seems ok now though.
> 
> MW: I have just generated a PDF V1.3 version of the document,
>     complete with bookmarks for the TOC.  Try that.  If it
>     still doesn't work I can e-mail it or a Word document.
> 
> Will try again.
> 
> Jon Awbrey
> 
> o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
> 
>