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

Re: Friendly amendment to Motion 25



> From: "Nate Hayes" <nh@xxxxxxxxxxxxxxxxx>
> To: "P-1788" <stds-1788@xxxxxxxxxxxxxxxxx>
> Subject: Friendly amendment to Motion 25
> Date: Sat, 28 May 2011 12:25:11 -0500
> 
> 
> P1788,
> 
> In light of recent discussions, I submit the attached PDF as a friendly 
> amendment to Motion 25.
> 
> Since there appears to be strong agreement between myself, John and Arnold 
> on the treatment of the min and max operations, I have moved this out of the 
> rationale and into the motion text.
> 
> To capture other ground which appears to have been gained, I also added a 
> new section in the rationale entitled "Towards a Foundation of Decorations" 
> that I hope people will consider.
> 
> Sincerely,
> 
> Nate Hayes
> 

	Nate,

	In future it might be nice if you 'decorated' such friendly
	amendments with the changes from the original motion.  Or,
	at least, tell us what the differences are.  As best I can
	tell the only differences are in the new section 2.4 & a
	new section 3.3 discussing FTDIA.

	No matter.

	In section 2.4 you state that the domain for min & max shall
	be the same as for add.  Indeed, I am fairly sure that the
	domain for min & max must be strictly LARGER than that for
	add.  Min & max must be defined for empty operands & non-
	empty results whereas add is not.  That is, min & max of a
	non-empty operand together with an empty operand must be the
	non-empty operand.  So min & max return results in a domain
	where add would return empty.  Min & max of both empty
	operands is empty just as it is in add.  I'll leave it to
	your sense of parsimony of definition to decide whether that
	case is outside the domain of min & max or inside it &
	defined to be empty.

	You make no mention of it but it must be true that union &
	intersection also differ from add in their domains of
	validity.

	The continuous you describe in definition 1 is C0 continuous.
	This must be so if, as you state later:

		S(sqrt,[0,4]) = D3.

	Definition 3 only applies to functions for which having one
	of their operands empty is OUTSIDE their domain.  Min & max
	are counter examples as may be other functions.

	The entries in the table in section 3 contain all possible
	examples of a decoration crossed with an empty or non-empty
	result.  And yet for many of the entries I cannot think of
	an example that applies.  Can you provide representative
	examples for each entry please?

	Many of the entries seem contradictory or, at least,
	confusing without examples to guide me.  In particular,
	decoration = D0, result = non-empty seems to describe a
	function that manages to return a non-empty result even
	though the cross product of its operands is entirely
	OUTSIDE its domain of definition.  Other entries have
	similar conceptual problems for me.

	Whether your philosophy is tracking or static the one
	fundamental property we need to prove an FTDIA is the subset
	property.  That is, if xx \subset yy then we must have
	S(f,xx) >= S(f,yy) FOR ALL xx, yy, & f.

	And yet you permit (in so many words on page 2):

		  S(sqrt,[0,4]) = D3
		 S(sqrt,[-1,2]) = D1
		S(sqrt,[-3,-1]) = D0

	Thus, were I to have xx = [-3,-1] & yy = [-3,2] we would
	have xx \subset yy but S(sqrt,xx) < S(sqrt,yy).

	You use the word Tracking when you describe your approach
	but this is neither tracking nor static behavior.  And not
	consistent with FTDIA as I have come to understand it.

	And we still have no decoration for 'ill'.

	Yours,

				Dan