[sc34wg3] Re: Reducing redundant constructs

Patrick Durusau sc34wg3@isotopicmaps.org
Fri, 15 Jul 2005 10:20:00 -0400


Lars Marius Garshol wrote:

>* Patrick Durusau
>| Yes, set theory does say exactly what you have recited below.
>So we agree on that. :)
>| The question is: What is the "set" in question.
>Clearly, yes. TMRM says it's only the set of properties.
>| If it is only the set of values, then yes, your reading is correct.
>The definition of a proxy seems pretty clear to me.
>| If the set in question is an identifier, plus a set of values, then
>| the fact that sets of values are "identical" does not lead to the
>| same result.  At least if the identifiers are different. If they are
>| the same, then again, you are correct.
>| That may not get us any closer to agreement but it is now clear what
>| I am contending?
>Yes, it is, but you seem to be contradicting your own draft. Not sure
>where the resolution to that lies.
Yes, I see what you mean.

Hmmm, let me see if I can simply say the problem without offering a 
solutiion first:

proxy = set of properties

so, as representatives of subjects, a set of proxies is a set of a set 
of properties. (This I understand to be your position.)

However, we know that proxies have identity as constructs, separate and 
apart from their representing subjects.

In order to reach my position, there would have to be a set of **** that 
is composed of subject proxies and their identifiers.

I put **** as I don't know what to call it.

The problem is that as a practical matter, proxies have identifiers in 
order for them to be addressed.

So, if we continue with the draft as written (I think you are correct in 
your reading), then there is no rule to deal with those identifiers for 
"identical" proxies.

That is a problem as other subject proxies may have references to the 
identifiers for subject proxies that may be "identical," the fate of 
which under the current draft is unspecified. (A bad thing.)

The more I think about it, although I would prefer to avoid 
contradicting my own draft, the more I think it is good that this issue 
came up. I think resolution will require discussion face to face in 
Montreal (and perhaps beyond).

Hope you are having a great day!


Patrick Durusau
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model
Member, Text Encoding Initiative Board of Directors, 2003-2005

Topic Maps: Human, not artificial, intelligence at work!