- From: Dan Brickley <danbri@danbri.org>
- Date: Thu, 07 Sep 2006 11:01:10 +0100
- To: Mark Baker <distobj@acm.org>
- Cc: "Ian B. Jacobs" <ij@w3.org>, Chairs@w3.org, www-tag@w3.org
Mark Baker wrote: > > Ian - a late comment, I know. Just tossing it out there ... > > On 9/6/06, Ian B. Jacobs <ij@w3.org> wrote: >> As before, a trailing "/" MAY be used and does not require Director >> approval. > > It seems to me that the lack of a standard approach there may lead to > confusion, as for each short name an author would also have to > remember whether it also needed the trailing slash or not. > > I think it would be better for the spec to pick an option, and say it > SHOULD be used, e.g. "A trailing slash SHOULD be used, but choosing > not to use one does not require Director approval". > > I don't have any strong opinions on which option should be used, > though I suppose having no slash is cleaner, saves a character, and is > consistent with existing W3C namespace naming practice (last time I > looked). It's important to at least allow the use of namespace URIs that end without the / (or that end /# I suppose). Sometimes there are deployment reasons for using a / over #, eg. very large RDF namespaces where you want to do an HTTP GET on term URIs, and not have the bit after the # ommitted by the HTTP library. For example, see the Wordnet 2.0 representation hosted at http://www.w3.org/2006/03/wn/wn20/instances/ see working draft at http://www.w3.org/TR/wordnet-rdf/ cheers, Dan -- http://danbri.org/
Received on Thursday, 7 September 2006 09:58:59 UTC