[wbs] Sean Palmer response to 'Syntax of QNames in N3'

Here are the answers submitted to 'Syntax of QNames in N3' (the public) for
Sean Palmer.


---------------------------------
What do you think of each option?
----
On a scale of  ... 1=totally unacceptable, 2=don't like but can live with,
3=don't{care| know}, 4=happy with this; 5=Absolutely wonderful
 * Do not allow "-" in qnames in N3.   To serialize in N3 something which
has a "-" in it, use the full (or relative) URI within angle brackets. 
The conservative choice, in that it restricts the language.

rated 2

 * Do not allow "-" in qnames in N3.  To serialize in N3 something which
has a "-" in it, use the munging described in the Notation3 article to
encode sequences of "-" and "_" into sequnces of "_".

rated 1

 * Allow "-" in qnames in N3.  If you need to make operators like "-"
later, force it to be surrounded by spaces, and disallow the qname which
consists of "-" alone with no prefix.  This constrains the language least.

rated 5


Comments (or a URI pointing to your comments): 
People have been using hyphens in Notation3 QNames for some time now, and
since CWM supports them it'll be a step towards conservatism which isn't
even necessary: since the tokenization rules are clear, it's still
possible to use space-padded hyphen as a mathematical operator. The fact
that people use hyphens in QNames anyway shows that there's an expectance
which is unlikely to just disappear.

The binary munging idea is nice, but is far too much burden and not easy
to explain to someone new to Notation3 in under a minute.


These answers were last modified on 21 February 2005 at 17:19:54 E.S.T.
by Sean Palmer

Answers to this questionnaire can be set and changed at
http://www.w3.org/2002/09/wbs/1/RDF-N3-Syntax/ until 2005-03-31.

 Regards,

 The Automatic WBS Mailer

Received on Monday, 21 February 2005 17:25:02 UTC