Re: xml:lang [was Re: Outstanding Issues ]

>At 10:15 27/02/2002 -0600, Pat Hayes wrote:
>>>>  >>>Patrick Stickler said:
>
>[...]
>
>>If that is what a  literal is, why have we been using examples like 
>>"35" during the entire datatyping discussion? (Should these have 
>>been ("35", "Mathematics") ?) How does one define the application 
>>of a datatype mapping to a (string, lang) pair ???
>
>When it was first proposed to extend n-triples to include the 
>language tag, the WG decided to hold off until we knew how 
>datatyping worked before making the decision.  As I recall this was 
>discussed at the last face to face.
>
>>>My question was: does anyone have a compelling reason to change 
>>>this.  Do you have one Patrick?
>>>
>>>>  And especially since literals are
>>>>>  now tidy,
>>>
>>>The pair above is just as tidy as "string".
>>
>>I think the point was that literals *without lang* are tidy, which 
>>is my understanding of the current situation.
>
>Just so.  Is there a reason why literals with lang can't also be 
>tidy.  They won't be tidy just on the string part.  Is that the 
>problem?

My problem is that the ground seems to have shifted, and I no longer 
know where it is. My understanding of the entire datatyping 
discussion, all three months or so of it, was predicated on the 
assumption that literals themselves *were* strings. If they are not 
strings, then many of the decisions we took no longer make sense to 
me. So are they strings or not? Dan C says they are but his examples 
seem to say the opposite. You say they are not. I feel like I should 
find somewhere safe to hide until the ground stops moving, as this is 
a very bad time to be trying to build something.

I have an action item to redraft the MT by April, but I cant start 
until I know how this item is going to be decided. So let me modify 
that action item. I will get the revised MT written by one month 
after I know for sure what a literal actually is, OK? And if a 
literal is not a character string, then I want to know what 
implications, if any, the non-string part has for determining 
entailments. For example, if that part is changed, does that change 
the meanings of any triples that contain the literal? Does it change 
them in any particular way?

Pat

PS. Heres an example. Duriing the datatyping discussion, we had a 
suggested distinction between German decimals (using commas) and 
English decimals (using dots), and US dates (MM/DD/YY) versus UK 
dates (DD/MM/YY). Are these examples of the use of language tags, or 
will they be considered to be relevant to language tagging?



-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes

Received on Friday, 1 March 2002 13:05:34 UTC