PDF file conundrum

Just to reiterate what I was saying on yesterday's call:

David has said that an IR (what I call a "Boothian IR") is a function.
Tim has said that an IR (what I call a "Timothian IR") is an abstract  
document.
Both have said that PDF files are IRs.

I think there is agreement that we say that IRs of any kind (in a web  
architecture context - the term has a long pre-web history of course)  
have awww:representations, even if we don't really understand what  
awww:representations are or how they're related to the IR.

So here is the conundrum:

   Every PDF file is a Boothian IR.
   Every Boothian IR is a function.
   Therefore, every PDF file is a function,
   and every PDF file has awww:representations.

   Every PDF file is a Timothian IR.
   Every Timothian IR is an abstract document.
   Therefore, every PDF file is an abstract document,
   and every PDF file has awww:representations.

Re-run this program filling in a variety of different things for "PDF  
file" and classify the outcomes as true, false, or nonsensical. (If  
you're feeling ambitious, subclassify "true" outcomes as falsifiable  
vs. not falsifiable in the Popper sense.) If you don't understand  
what I mean by a term such as "PDF file", propose a definition that  
is a compromise between what you think I might mean and what you  
would like it to mean (so as to advance your own view).

E.g. try it out with:
   journal article
   DNA sequence
   home page
   blog
   gzip file
   number
   form  (e.g. http://random.org/integers/ )
   pubmed record
   awww:Representation
   web site

I just pick Tim's and David's definitions as examples; obviously  
others, such as the one in AWWW, can be plugged in as well.

Maybe we could make a matrix, with reasons for each determination,  
with "disputed" as a possible entry.

I'm just trying to suggest a method here. We have multiple IR-like  
classes, and multiple other classes that might or might not be  
subclasses of each IR-like class, and before we can talk about  
httpRange-14 or 200s we need to just look at the subsumption matrix  
dispassionately. Of course instead of English we should be using URIs  
that have documented meanings, to make it easier to say definite  
things and more difficult for meanings to squirm about. I'm thinking  
about the best way for us as a group to set up a shared URI  
documentation space.

Jonathan

Received on Wednesday, 28 May 2008 14:27:30 UTC