W3C home > Mailing lists > Public > www-rdf-interest@w3.org > December 2000

Re: Towards a final consensus re statements & statings

From: Pierre-Antoine CHAMPIN <champin@bat710.univ-lyon1.fr>
Date: Fri, 08 Dec 2000 16:18:32 +0100
Message-ID: <3A30FBC8.C6C1BE7C@bat710.univ-lyon1.fr>
To: Brian McBride <brian_mcbride@hp.com>
CC: www-rdf-interest@w3c.org
Brian McBride wrote:
>
> [a nice digest of the spec]
> 
> Can someone point out where m&s suggets or implies
> that a reified statement represents a stating?

I guess M&S *suggests* more that it *implies*,
that is probably why we had those long discussions, and why I'm no more sure which option I prefer...
I will try to point it out, and then do critics

 1.  the notion of stating is a useful notion, and M&S does not mention it at all, like the authors have "forgotten" it. Since the concepts of reified statement/stating are quite near, it is easy to interpret that the authors... would have written "stating" if they had made the distinction in the first place.

 2.  it is very easy to write a piece of RDF with two reified statements being identical, and M&S does *not* make it illegal. Since a statement is unique, it looks like a bug in the model, unless reified statements are in fact statings.

 3.  the example for reification (namely "Ralph says that Ora was the creator of the page") tastes just like a stating. One would like to add the date when Ralph said that, hence the problem raised by Jonas Liljegren : if I also want to state that "Pierre-Antoine said that Ora was the creator of the page, at another date", then which date applies to which stater ?

  st1: [Ora, creator, page]
  st2: [st1, saidBy, Ralph]
  st3: [st1, saidAt, 01/12/99]
  st4: [st1, saidBy, Pierre-Antoine]
  st5: [st1, saidAt, 01/12/00]



Those are the main reasons I had to think that reified statements were in fact statings.
Now, here are critics for each argument, who lead me to think maybe I was wrong :

 about 1.  That argument is not the stronger one... Actually, I would rather stick to the spec as long as I can... So, what about 2. and 3. ? :)

 about 2.  It is actually very easy to write a piece of RDF where any resource is duplicated, not only statements. And it is not considered a bug... although we might have troubles handling it ! Example :
  
  <rdf:Description about="http://www.w3.org/">
    <as:contentQuality> Excellent </as:contentQuality>
  </rdf:Description>
  <rdf:Description about="urn:WWWConsortium">
    <as:homepage>
      <rdf:Description id="localID"/>
    </as:homepage>
  </rdf:Description>
  <rdf:Descritpion about="mypage.html">
    <as:pointsTo rdf:resource="#localID"/>
  </rdf:Description>

where #localID and http://www.w3.org/ are in fact the same thing.

  about 3.  I already submitted the idea below on the list, but I will go further : the problem raised by Jonas is not a problem ! If we do consider that the reified statement is really a statement rather than a stating, then the date should not be a property of st1, but rather of st2 and st4 !

  st1: [Ora, creator, page]
  st2: [st1, saidBy, Ralph]
  st3: [st2, at, 01/12/99]
  st4: [st1, saidBy, Pierre-Antoine]
  st5: [st4, at, 01/12/00]

The statements st2 and st4 are actually statings, not because *every* reified statement be a stating, but because of the particular meaning of their predicate "saidBy".


Here is my belief about the statement/stating polemics just now.
If I forgot some important point in the pro-stating list... let me know !

  Pierre-Antoine

-- 
Sometimes I think the surest sign that intelligent life exists elsewhere in the
universe is that none of it has tried to contact us.
(Bill Watterson -- Calvin & Hobbes)
Received on Friday, 8 December 2000 10:19:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:47 GMT