OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: [xml-dev] XML-with-datatypes (was....)

[ Lists Home | Date Index | Thread Index ]

From: Peter Hunsberger [mailto:peter.hunsberger@gmail.com]

>Yes, I suppose so, but I don't seem to have much time to do any
>exercise these days (sigh)...

Me too.  If my blood sugar doesn't kill me, the lawn will.

>Given other posts of yours, I suppose you're trying to figure out what the
>to category theory means?  

That's a good place to start.  I will reread the wikipedia entry when I can 
right after I finish and condense the book on liability law for 9-1-1. It's 
strange what one has to know to wrangle coders and salesmen these days, but
that issue on the TAG list makes me realize it's not just my problem.

>You already know my leanings towards graph theory; in as much as XML
>get's more people working on good ways to do generalized graph
>traversal I'm all for it.  

Yep. You fit well on the CG list. I had forgotten how much Sowa had 
influenced topic maps.  Many concepts are converging and maybe that 
means AI's time is upon us.  Pioneers, arrows in the back, all of that. 
AI could drive strong-typing front and center.  We get to keep XML loose 
because we keep humans in the loop.  OTOH, the semantic web is predicated 
on machine-to-machine communications and automated discovery where typing 
hints have higher value.

>However, every time I see someone wanting
>to glue an artificial relationship into XML some part of me screams
>for the generalization of graphs.  OTOH, precomposed containers of
>relationships (elements with attributes) do remove lots of the tedium
>of explicitly building out every relationship...

Indeed.  It's in the way(s) that we use them.

The strong typing thread (last episode) led some to read the articles 
from Java's creators, etc.  Good stuff.  Over time and many screw ups, 
I finally got it through my dull brain why XML succeeded, and most of 
it has to do with semantic free languages FOR expression.   Until the 
bottom layers work and call out their needs/deficiencies, theory and 
top-down imposition of managed code limits reach.  Or simply, the 
deeper the semantics, the smaller the reach.  Semantics have to added 
'very delicately'.  It is somewhat like recording: build a solid 
rhythm section then be very choosy about what you put over that. 
The hallmark of mediocrity is a busy arrangement.

Shukrivar!  Thank god!  I want to work on a new song called, "They 
Pay Me to Tell You This!"  I started it last week, then today via 
Elliotte saw Paul Graham's article and thought "independent invention 
strikes again; it must be time for this song."



News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS