Re: bang ! in turtle

Sorry Nathan...

I must admit I am really afraid of overcomplicating things. The beauty if Turtle is its simplicity. Simplicity meaning that Turtle, as it is today, can be made understandable with a minimal knowledge of RDF, if any. Adding additional syntaxes may jeopardize this feature and make Turtle a much more geekier language. That may backfire on the community big time; we may end up in the RDF/XML situation making the language too complicated!

On Mar 1, 2011, at 18:30 , Nathan wrote:

> When I was playing about with JSN3 [1] a few months ago, I also made up  a few more shorthands which were quite interesting / useful:
> 
> Shorthand:
>  a	rdf:type

We have that already

>  ^a	[ owl:inverseOf rdf:type ]

As Sandro said, this is just an explanation, right? Formally, what you describe is a bnode that can be inferred to be an inverse, etc, etc

>  >	rdfs:seeAlso

What justifies picking this specific predicate over anything else? Surely dc:title or dc:comment is more frequent than seeAlso, do we want to add a sign for those? This would have no end...

>  =	owl:sameAs

I am wary of bringing in a vocabulary item from a large animal like OWL. Also: why not then owl:equivalentProperty, owl:equivalentClass, ... again, this has no end.

>  =>	log:implies
>  <=	log:implies, but in the other direction
> 

These are of course useful _if_ we defined a syntax for a rule language, or a serialization of RIF. Alas!, this is _not_ something for this WG.

> predicate prefixes:
> 
>  ^x	[ owl:inverseOf x ]
>  >x	[ link:listDocumentProperty x ]

I am not even sure I understand this! I do not know what link:listDocumentProperty is...

> 
> example of the last one, since not many people use/know "link:listDocumentProperty"
> 
>  :me >foaf:knows <http://example.org/my-friends> .
> 
> meaning: look here for a list of relations where the subject is :me and the predicate is foaf:knows - in this case, here's a doc to look in to find people I know, can also be seen as type hinted seeAlso, comes in most useful. more info on this predicate is in the Linked Data design Issue.


What is this in terms of triples? I am serious, I simply do not understand what this is! And that might be significant: we'd create an overcomplicated structure. And we should be very very careful in my view.

Ivan


> 
> [1] http://webr3.org/apps/specs/jsn3/
> 
> Cheers,
> 
> Nathan
> 
> Antoine Zimmermann wrote:
>> That would be a very nice feature.
>> I often see myself writing stuff like:
>> s1  rdf:type  C .
>> s2  rdf:type  C .
>> ...
>> sN  rdf:type  C .
>> when I would like to write:
>> C is rdf:type of s1, s2, ..., sN .
>> I'd be happy if the syntax was
>> :s  is :p of  :o .
>> but caret is ok too (please avoid '!').
>> AZ.
>> Le 01/03/2011 17:41, Nathan a écrit :
>>> actually ^ may be better.. such that
>>> 
>>> :a :b :c .
>>> 
>>> could be written as:
>>> 
>>> :c ^:b :a .
>>> 
>>> meaning
>>> 
>>> :c [ owl:inverseOf :b ] :a .
>>> 
>>> meaning:
>>> 
>>> :a :b :c .
>>> 
>>> cheers
>>> 
>>> Nathan wrote:
>>>> Hi All,
>>>> 
>>>> Is there any chance we can include ! in turtle to prefix predicates
>>>> and inverse the directionality, such that
>>>> 
>>>> :a :b :c .
>>>> 
>>>> could be written as
>>>> 
>>>> :c !:b :a .
>>>> 
>>>> (as opposed to :c is :b of :a)
>>>> 
>>>> Cheers,
>>>> 
>>>> Nathan
>>>> 
>>> 
>>> 
> 
> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf