Re: A different symbol for {|

If <<>> here is suggested as different {|..|} and the inside is still a 
  PredicateObjectList

i.e.

    :a :knows :b << :since 1900 >>

that is the two triples

    :a :knows :b .
    <<:a :knows :b>> :since 1900 .

it seems to work technically i.e. the grammar is not ambiguous.

    :a :knows << :x :p :o >>

is still one triple, with object triple term <<>>

     Andy

On 07/01/2021 09:11, Miel Vander Sande wrote:
> Technically,  you'd be making a fourth term, which would be outside of 
> the triple structure. It would also conflict with the named graph 
> element in for instance nquads.
> 
> I also see some issues with having multiple annotations. Would the <<>> 
> then support linebreaks?

yes if a PredicateObjectList
unless we change the white space rules which would be big.

> Your proposal also looks a bit like using the blank node [ ] notation in 
> the graph position,  so I see some possible confusion there

Yes. It would not extend to NQ* and allowing for (future) triple terms 
in the graph position for "generalized RDF".  It may be possible but as 
a change to the grammar tool assumptions.

There is a lot to be said for easy of implementation for better uptake.

     Andy

> 
> Best,
> 
> Miel
> 
> 
> On Thu, 7 Jan 2021, 09:46 Laura Morales, <lauretas@mail.com 
> <mailto:lauretas@mail.com>> wrote:
> 
>     OK fair enough. Is there a reason why this cannot be used btw?
> 
>          :a :knows :b << :since 1900 >> .
> 
> 
> 
> 
>     Sent: Thursday, January 07, 2021 at 9:37 AM
>     From: "Miel Vander Sande" <miel.vandersande@meemoo.be
>     <mailto:miel.vandersande@meemoo.be>>
>     To: "Laura Morales" <lauretas@mail.com <mailto:lauretas@mail.com>>
>     Cc: public-rdf-star@w3.org <mailto:public-rdf-star@w3.org>
>     Subject: Re: A different symbol for {|
> 
>     Knowing that :knows is universal, I find this is more confusing in
>     the same way Andy pointed out that  :s :p "abc"@{ :a:b } . reads
>     like modifying "abc". Plus, we are digressing from the triple model
>     again and I don't think that will make things easier in the long
>     run. RDF simply isn't PG and so is the syntax.
> 
>     With this thread in mind, the only minor change I see fit is going
>     from {| |} to <| |> (or whatever other character to replace |)
>

Received on Thursday, 7 January 2021 16:58:06 UTC