Re: [N-Quads] N-Quads spec should say it is a superset of N-Triples

My recollection is that there was not consensus about what language to use
to say that the same document with text/turtle vs application/trig produced
a graph and the other a data set. Nor about the direction of the
relationship (subset, superset) between them. There's some language in the
primer but that's about it.

--gavin

On Sat, Feb 28, 2015 at 5:23 PM, David Booth <david@dbooth.org> wrote:

> On 02/28/2015 12:40 PM, Andy Seaborne wrote:
>
>> On 27/02/15 23:22, David Booth wrote:
>>
>>> For whenever the next update is made . . .
>>>
>>> The N-Quads spec
>>> http://www.w3.org/TR/2014/REC-n-quads-20140225/
>>> should say up front that every N-Triples document is also an N-Quads
>>> document, rather than forcing the reader to guess or try to figure it
>>> out from comparing the two specifications.
>>>
>>> Similarly, the N-Triples spec should say that N-Triples is a subset of
>>> N-Quads.   It already says that it is a subset of Turtle:
>>> "N-Triples is an easy to parse line-based subset of Turtle [TURTLE]."
>>>
>>
>> Slightly "devil's advocate" but while it is a subset by syntax, it is
>> not exactly a subset in all ways.
>>
>> When bytes are treated as N-Quads it means the default graph of a
>> dataset; as N-triples, it does not say that.
>>
>
> That might be worth saying too.
>
> Thanks,
> David Booth
>
>
>