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

SUO: RE: Re: Exposition




Dear Jon,

> ¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
> 
> Strand 3.  Embedded Tables & Extended Relations
> 
> ¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
> 
> JA:  There were a few issues about "embedded", "indexed", "keyed",
>      or "ordered" relational tables that were initially raised in
>      the present frame, and so I will try to address them now, at
>      least until the running accumulation of bits grows too large.
> 
> JA:  This filiation is offspring to your comment that,
>      with regard to your putative method of reduction:
> 
>      | MW:  Where I come from this method would
>      |      be informally known as reification.
> 
> JA:  Having seen more of your elaboration of what you are 
> actually doing,
>      I begin to see this more as a matter of two kinds of 
> embedding that
>      you are carrying out on the given relation, and one of 
> these brands
>      of embedding, extension, or extenuation does bear a 
> relation to the
>      operation that is classically known as "hypostatic 
> abstraction", or
>      else as "reification".  But all of these operations act 
> to increase
>      the arity of the relation, so I really do not see how 
> they help the
>      case that you want to establish.  I have sorted out 
> these questions
>      for separate discussion in this note.
> 
> JA:  Concreteness in the beginning is hardly misplaced,
>      so let us consider a discreetly simple example of
>      a relational database, just to pick one at random
>      from off the grainy strand beneath our feet, here:
> 
> JA:  Relation L(&) = {<x, y, z> in B^3 : x & y = z}.
> 
>      o-----o-----o---------o
>      |  x  |  y  | z = x&y |
>      o-----o-----o---------o
>      |  0  |  0  |    0    |
>      |  0  |  1  |    0    |
>      |  1  |  0  |    0    |
>      |  1  |  1  |    1    |
>      o-----o-----o---------o
> 
> JA:  There are many different ways to embed a k-adic relational table
>      in a more spacious and sumptious table, that is, one that enjoys
>      either more rows or more columns or both, and with whatever form
>      of extenutaion is chosen to extend the relation thus represented
>      either to larger samples of its initial cartesian product domain,
>      or else to relations of augmented audacity, arete, or valoration.
> 
> JA:  But you have already instanced for me the two most common types
>      of embedding or extension, and so it is meet that we take these
>      up first.  Let us just contemplate their basic forms all in one
>      place, and leave off discussing the details to later in the day.
> 
> JA:  Extension 1.  Abstract Hypostasy, Reflective Entitlement.
> 
>      o---------o-----o-----o---------o
>      |    t    |  x  |  y  | z = x&y |
>      o---------o-----o-----o---------o
>      | "L(&)"  |  0  |  0  |    0    |
>      | "L(&)"  |  0  |  1  |    0    |
>      | "L(&)"  |  1  |  0  |    0    |
>      | "L(&)"  |  1  |  1  |    1    |
>      o---------o-----o-----o---------o

MW: I don't follow this.
> 
> JA:  Extension 2.  Indexing, Keying, Ordering, Sequencing.
> 
>      o---------o-----o-----o---------o
>      |    n    |  x  |  y  | z = x&y |
>      o---------o-----o-----o---------o
>      |    1    |  0  |  0  |    0    |
>      |    2    |  0  |  1  |    0    |
>      |    3    |  1  |  0  |    0    |
>      |    4    |  1  |  1  |    1    |
>      o---------o-----o-----o---------o
> 
> JA:  Some of the names, of course, are not yet standard.
> 
> MW:  In the interests of clarity, this can be seen as an
>      intermediate step in the transformation I am making.
> 
> MW:  Having introduced the index (the reification step) ...
> 
> Gasp!
> 
> MW:  then each of the components of each row is related to that index.
>      This gives you a basis for creating a relation for each 
> column, so:
> 
> MW:  n  = {1,2,3,4}
> 
>      x = {<1,0>, <2,0>, <3,1>, <4,1>}
> 
>      y = {<1,0>, <2,1>, <3,0>, <4,1>}
> 
>      z = {<1,0>, <2,0>, <3,0>, <4,1>}
> 
> MW:  I fail to see why you think x, y, z are not relations.
>      I have swopped 1 tradic relation for 3 dyadic relations
>      plus a set.
> 
> Unhuh, and are these three 2-adic relations
> just hanging out there, blowing in the wind,
> as if they were to wash up on the shores of
> your certainty, each one a detached message,
> all dispersed in welters of many-splintered
> brands and colors of bottles, that the wine
> darks seas of the world give up to you, and
> you, on receipt of this gift, how would you
> even know that each message had any bearing
> on the meaning of the others, if not by way
> of some clue that you can persieve or purse
> seine or just plain abdeuce to be contained
> therein, read between the letters and lines?

MW: Not at all, they are linked by the objects
they contain (though it would be sensible not
to use 1,2,3,4 as the index as you have done
since at least 1 is also used to represent
a different object in the matrix).

> And are these characters, in their pedigree,
> any kind'o'kin to the gang'o'four you hinge
> them on, or do they hang but separately, as
> nothing but instances of one dummy variable
> after'n'other?

MW: Very colourful, but you lost me.
> 
> MW:  It is relatively straight forward to do the
>      transformation in each direction, so they
>      are equivalent.
> 
> Not so.
> 
> You must remember this:
> A set is still a set,
> as the index goes by.
> 
> Consider an "indexed 3-adic relation",
> which is really this 4-adic relation:
> 
> o---------o-----o-----o---------o
> |    n    |  x  |  y  | z = x&y |
> o---------o-----o-----o---------o
> |    1    |  1  |  1  |    1    |
> |    2    |  1  |  0  |    0    |
> |    3    |  0  |  1  |    0    |
> |    4    |  0  |  0  |    0    |
> o---------o-----o-----o---------o
> 
> This is a distinct 4-adic relation that might
> well arise in any number of applications from
> what is fundamentally the same 3-adic relation.
> Indeed, there are 4! = 24 distinct such 4-adics
> that may easily arise in one setting or another
> out of the same 3-adic relation that we had in
> the beginning.  

MW: I don't follow this move. To me 1,2,3,4 are (or 
should be) unique object identifiers in the Universe
of Discourse (1 isn't here so there is a problem).

> So the mapping is not 1 to 1,
> but 1 to 24, and this does not make for any
> kind of invertible transformation, nor any
> kind of equivalence between formal objects.
> 
> You have come to the very threshold of teridentity, my friend.

MW: Save me, save me!!!!
> 
> | You tell him to come in sit down
> | But something makes you turn around
> | The door is open you can't close your shelter
> | You try the handle of the road
> | It opens do not be afraid
> | It's you my love, you who are the stranger
> | It is you my love, you who are the stranger.
> |
> | Leonard Cohen, 'The Stranger Song'
> 
> Jon Awbrey
> 
> ¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
> 


Regards  
      Matthew
===============================================================
Matthew West                    http://www.matthew-west.org.uk/

Principal Consultant                   Shell Visiting Professor
Operations & Asset Management            The Keyworth Institute
Shell Services International            The University of Leeds
http://www.shellservices.com/  http://www.keyworth.leeds.ac.uk/

H3229, Shell Centre, London, SE1 7NA, UK.
Tel: +44 207 934 4490 Fax: 7929 Mobile: +44 7796 336538
===============================================================