W3C home > Mailing lists > Public > uri@w3.org > October 2004

Re: TAG scheme - some comments

From: Sandro Hawke <sandro@w3.org>
Date: Fri, 22 Oct 2004 15:05:45 -0400
Message-Id: <200410221905.i9MJ5jdf030143@roke.hawke.org>
To: "Hammond, Tony" <T.Hammond@nature.com>
cc: uri@w3.org


> > What 
> > application for tags do you have in mind where parsing the 
> > authorityName is necessary?
> 
> Simply that a TAG processor should be able to recognize a given URI to be a
> TAG URI against the published syntax.

Ah, I see what you mean -- one normally expects the BNF to be fairly
self-sufficient, and we've put a weird exception in the text nearby.

We should change

06:  taggingEntity = authorityName "," date

to 

07:  taggingEntity = authorityName "," date / futureExpansion

and define futureExpansion fairly broadly (but subtract the existing
stuff -- you can do that in the XML spec's BNF -- can you do it in
ABNF?).  Then in the text we can say you MUST NOT mint futureExpansion
tags but MUST NOT reject them.   The first MUST NOT will be relaxed by
a future spec, in chunks, as needed.

> I'm also not a little sceptical about the following injunction:
> 
> > for generation only -- software SHOULD NOT parse tags.
> 
> The TAG has been minted to a specification which defines the two component
> parts of a taggingEntity quite clearly: authorityName and date. I don't
> believe there should be any problem in an application retrieving these
> pieces of information.

No, you're right.  If it matches that branch of the grammar, agents
are welcome to parse it, although I have no idea what good it would do
them.  Forgive me for overstating the case in my earlier message.

   -- sandro
Received on Friday, 22 October 2004 19:03:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 January 2011 12:15:34 GMT