Re: For RDF Concepts: term-equality and value-equality.

On 5/9/13 2:31 AM, Pat Hayes wrote:
> Well, strictly speaking, TWO literals are never term-equal, because if they are, then there is only one of them. Two expressions might be tokens of the same literal, under the conditions you describe.
>
> But maybe this is too pedantic, sigh.
>
> However, suggest:
>
> Two literals are term-equals // Two literals are term-equal
>
> I think this is just grammar: A equals B = A and B are equal.
>
> Two literals can share equal values // Two literals can have the same value
>
> or even better, Two literals can have the same referent...  except then you would need to say earlier that literals have referents (just like IRIs do). Which is correct but I guess might be confusing. Sigh.

It isn't confusing. In my eyes, that's another point of critical clarity 
that will aid RDF comprehension. Literals and IRIs are identifiers that 
have referents. Then we can toss the "closed world" vs "open world" 
analogies in the bin when talking about effects of RDF semantics on data 
management etc..

Kingsley
>
> Pat
>
>
>
> On May 8, 2013, at 1:06 PM, David Wood wrote:
>
>> I've added some changes to this section that I think addresses everyone's concerns.  Please let me know.
>>    https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-concepts/index.html
>>
>> Regards,
>> Dave
>> --
>> http://about.me/david_wood
>>
>>
>>
>> On May 8, 2013, at 11:23, Andy Seaborne <andy.seaborne@epimorphics.com> wrote:
>>
>>>
>>> On 08/05/13 15:21, Sandro Hawke wrote:
>>>> On 05/08/2013 10:12 AM, Pat Hayes wrote:
>>>>> On May 7, 2013, at 4:13 AM, Andy Seaborne wrote:
>>>>>
>>>>>> RDF Concepts says:
>>>>>>
>>>>>> [[
>>>>>> Literal equality: Two literals are equal if and only if the two
>>>>>> lexical forms, the two datatype IRIs, and the two language tags (if
>>>>>> any) compare equal, character by character.
>>>>>> ]]
>>>>>>
>>>>>> I think it would be useful to spell out "term equality" and "value
>>>>>> equality" as important concepts.
>>>>>>
>>>>> Blech. I strongly dislike having "kinds" of equality. Equality has one
>>>>> meaning, and it does not admit of degrees or kinds. This is a
>>>>> difference between literals and literal values, not two kinds of
>>>>> equality. We already draw out the distinction between literals and
>>>>> literal values.
>>> I don't see anything about testing values in concepts - I think it is useful in "concepts" to put literal equality and value testing close together.
>>>
>>> The important point, which continues to confuse people, is that
>>>
>>> "1"^^xs:integer
>>> "+1"^^xs:integer
>>>
>>> are different terms.
>>>
>>> How we express that, I don't mind.
>>>
>>> Text way down in a modified 5.5 isn't helpful where as something at the point of talking about literal equality is more reader-focused.
>>>
>>> 	Andy
>>>
> ------------------------------------------------------------
> IHMC                                     (850)434 8903 or (650)494 3973
> 40 South Alcaniz St.           (850)202 4416   office
> Pensacola                            (850)202 4440   fax
> FL 32502                              (850)291 0667   mobile
> phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
>
>
>
>
>
>
>
>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen