W3C home > Mailing lists > Public > semantic-web@w3.org > July 2010

Re: predicates as verbs

From: Bernard Vatant <bernard.vatant@mondeca.com>
Date: Thu, 1 Jul 2010 22:41:04 +0200
Message-ID: <AANLkTimPfMCEdDDnx2LvIDUNh5n7GlV-lR7T5gZNb7Q1@mail.gmail.com>
To: Pat Hayes <phayes@ihmc.us>
Cc: semantic-web at W3C <semantic-web@w3c.org>
Hi Pat

This entire discussion is absurd.
>

My point, exactly. But if I did not want to sound rude, I know you care less
about this kind of diplomacy. Actually you have somehow an image to maintain
in this domain :)
And you could not resist jumping in this absurd discussion, either ...



> "Verb" is a grammatical category. "Relation" is a semantic category.
>

Indeed ...


> In English, virtually any relation can be expressed using a verb phrase,
> and almost any verb-described action can also be nominalized.
>

The latter point is less natural in French, but in theory is also true.



> So by all means, if it makes you feel better, express your RDF or logical
> content using verbs. Or, if you prefer, use nominalized expressions. It
> makes no - literally, NO - difference to the meaning of the RDF.
>

Of course. But it can hinder comprehension by humans, in particular when the
human is not a native english speaker. I have often questions about
predicates such as skos:broader.
Does  "A   skos:broader  X"   means  "A is broader than B"  or " A has
broader concept B". It happens that it is the latter, but for some people
it's counter-intuitive, and you can't always say RTFD, can you?



> BTW, the cupola "is", while technically (grammatically) a verb, can be used
> to predicate almost any property of any thing. The sentence "My cat is
> black" has a verb in it, as required by the English grammatical rules (btw,
> the ones you and I learned in village school are almost all wrong)
>

Be careful on what you say here. MY school teacher was my father, he could
not have been wrong, particularly on things as important as grammar !
But my children have learnt very different ones indeed. Not sure they are
less wrong. Waiting for the next generation (grand-daughter just learning to
read ...)


> , but its logical form is a simple predication: Black(myCat), which has no
> verb-like entity in it anywhere. Most simple facts are like this.
>

I must agree with you on logic ground (no choice). But the point if that
this simple (binary?) predication has to be "triplified" in some way to
enter RDF land. You will need some predicate to link the concept "Black" and
the entity "myCat". Adding verbosity, so to speak. Same for Cat(Lune) (Lune
is the name of my cat) which expands to

:Lune   rdf:type   :Cat

etc

But in the 6000+ languages of the world, maybe there are some where this
"state of affairs" Black(myCat) will be represented by a unary predication,
like "raining" or "sunset".

Bernard



> On Jul 1, 2010, at 12:03 PM, Bernard Vatant wrote:
>
> Hi Henry
>
> I guess we should not confuse the grammatical kind of the natural language
> term used to label the predicate, which is not necessarily a verb
>
> :myCat   :color   :black               (noun)
> :myCat   :hasColor   :black         (verb)
> :myCat   :colored  :black            (adjective)
>
> ... with the underlying semantics of the triple : a sentence in which the
> predicate is in the verb position and carries more or less explicitly the
> function of a verb
> And I think this is the point of Paola question.
>
> The label used, if it's not a verb, is most of the time (always?) a
> shortcut (ellipse) for a more verbose (verbal) form, even in your example
>
>
>  (2 2) plus 4.
>>
>
> is a shortcut for
>
>  (2 2)  'has result by addition'  4
>
> or any other similar expression using "yields" "gives", whatever ...
>
> I don't see any predicate making sense which cannot be expanded in natural
> language into a verbal expression, at least in languages I know of, close to
> english in their use of verbal forms. French does not differ on this.
> A triple is an assertion. Verbs are the natural way to make assertions. At
> least that's what I've be taught in my village school in France back in the
> 50's. "A well-formed sentence contains a verb."
> Granted that was the time of ink and quills, but ...
>
> Cheers
>
> Bernard
>
>
> --
> Bernard Vatant
> Senior Consultant
> Vocabulary & Data Engineering
> Tel:       +33 (0) 971 488 459
> Mail:     bernard.vatant@mondeca.com
> ----------------------------------------------------
> Mondeca
> 3, cité Nollez 75018 Paris France
> Web:    http://www.mondeca.com
> Blog:    http://mondeca.wordpress.com
> ----------------------------------------------------
>
>
> ------------------------------------------------------------
> 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
>
>
>
>
>


-- 
Bernard Vatant
Senior Consultant
Vocabulary & Data Engineering
Tel:       +33 (0) 971 488 459
Mail:     bernard.vatant@mondeca.com
----------------------------------------------------
Mondeca
3, cité Nollez 75018 Paris France
Web:    http://www.mondeca.com
Blog:    http://mondeca.wordpress.com
----------------------------------------------------
Received on Thursday, 1 July 2010 20:42:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:20 UTC