Re: Blank Nodes and SPARQL

Just a small point:

Jos indicated that he is using blank node labels such as _:e123_14_ 
because NCNAME is currently being used
in the grammar while Dan is using  blank nodes of the form  _:345c23 
(currently not allowed)

I  suggest staying with the NCNAME definition so that you have only one 
kind of name and an optional
prefix.    It means Dan would need to write something like   _:L345C23, 
but I'd prefer not to have names with prefixes being different from names 
without prefixes unless there is a compelling reason to make them 
different.
The fewer special cases the better.

Stan Devitt





Jos De_Roo/AMDUS/MOR/Agfa-NV/BE/BAYER@AGFA
Sent by: public-rdf-dawg-comments-request@w3.org
06/30/2005 07:31 PM

 
        To:     connolly@w3.org
        cc:     Andy <andy.seaborne@hp.com>, Arjohn Kampman <arjohn.kampman@aduna.biz>, 
public-rdf-dawg-comments@w3.org, public-rdf-dawg-comments-request@w3.org, 
Ron Alford <ronwalf@umd.edu>, Seaborne@w3.org, (bcc: Stan 
Devitt/AWKCT/CAN/AGFA/CA/BAYER)
        Subject:        Re: Blank Nodes and SPARQL



Dan Connolly wrote:
[...]
> cwm has a mode where it makes up blank nodes based on line/column
> numbers so you can see where they came from and so they're repeatable.
> So we might have data a la...
>
>  _:l23c14 foaf:mbox <mailto:connolly@w3.org>.

I am using _:e123_14_ where 123 is tripleCount and 14 is documentCount
the 'e' is a recent change after I realized that NCNAME can't start
with [0-9]
true?