W3C home > Mailing lists > Public > semantic-web@w3.org > August 2007

Re: what is a plain literal?

From: Garret Wilson <garret@globalmentor.com>
Date: Fri, 10 Aug 2007 06:57:59 -0700
Message-ID: <46BC6EE7.8070905@globalmentor.com>
To: Bijan Parsia <bparsia@cs.man.ac.uk>
CC: Semantic Web <semantic-web@w3.org>

Bijan,

Bijan Parsia wrote:
>
> On Aug 10, 2007, at 2:09 AM, Garret Wilson wrote:
>
>>
>> "Plain literals are considered to denote themselves, so have a fixed 
>> meaning."
>>
>> Is this a typo?
>
> No.
>
>> It's not the most natural sentence
>
> I have no problem with the part you have problem with. (I'm not a fan 
> of the "are considered to"  part rather than just "Plain literals 
> denote themselves...".

Actually, the part that didn't seem natural to me is, "so have a fixed 
meaning." It almost seems like a word is missing. I just wanted to make 
sure. (It almost sound like this sentence: "Plain literals are 
considered to denote themselves, so have a nice day.")

>
>> ---does it mean, "and thus always have a fixed meaning", or "and and 
>> so have no fixed meaning?"
>
> I have no idea why you are getting the second reading. Replace "so" 
> with "thus". If you replace it with "thus not" you've clearly changed 
> the meaning.

That's why I'm asking. "So" and "thus" are adverbs; thus, with a 
compound predicate, I'm used to seeing a conjunction.

>
> The meaning is, as they say, fixed.

So I'll consider the sentence to say, "Plain literals are considered to 
denote themselves, and so have a fixed meaning." Thanks.

>
>> Next question: how do plain literals differ semantically from typed 
>> literals with a datatype URI of xsd:string?
>
> They can have a lang. Datatyped literals cannot. It was a big deal. It 
> is clearly a wart.

You've stated a fact about the restrictions RDF places on the xsd:string 
datatyped literal in the RDF abstract syntax. This fact is true. But my 
question was, "How do plain literals differ semantically from typed 
literals with a datatype URI of xsd:string?" That RDF places a 
restriction on one and not the other tells me that RDF treats them 
differently, but I already know that! (One is a typed literal, the other 
is a plain literal---of course!) As I pointed out, that's begging the 
question---how is the *meaning* different?

Put another way---if I never, ever use language tags, why would I ever 
want to use a plain literal over a typed literal with xsd:string? If 
they mean different things, then I have a design decision to make when 
creating an ontology, based upon the difference in meaning between the 
to. What is that difference in meaning, if any, so that I can make an 
informed decision when creating my ontologies?

Thanks,

Garret
Received on Friday, 10 August 2007 13:58:12 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:17 GMT