W3C home > Mailing lists > Public > www-tag@w3.org > December 2003

RE: Suggested text for 3.1 (small changes)

From: Williams, Stuart <skw@hp.com>
Date: Mon, 1 Dec 2003 17:33:32 -0000
Message-ID: <E864E95CB35C1C46B72FEA0626A2E80869A3A3@0-mail-br1.hpl.hp.com>
To: "'Chris Lilley'" <chris.lilley@wanadoo.fr>
Cc: www-tag@w3.org


It's common in a number of specs to establish in a set of common prefix
mappings used throughout the remainder of the spec (unless stated otherwise)
eg [1]. I'm not sure how many example prefixes we actually use, maybe not
enough to warrant such a subsection.

> -----Original Message-----
> From: Chris Lilley [mailto:chris.lilley@wanadoo.fr] 
> Sent: 1 December 2003 16:17
> To: www-tag@w3.org
> Subject: Suggested text for 3.1 (small changes)
> Hello www-tag,
> Sending via an alternate email account, chris@w3.org is 
> temporarily hosed.
> Tim Bray wrote:
> TB> 3.1 first numbered step, shouldn't that be xlink:href rather than 
> TB> "XLink href"?  In particular since you use that in step 2 :)
> No. I chose the phrase "Xlink href" to mean "the href 
> attribute in the XLink namespace" specifically because some 
> people seem to think that the entire string "xlink:href" is 
> magical. It isn't, it can be wibble:href and work just fine 
> if correctly declared. Conversely, xlink:href can not work if 
> you assign that prefix to some other URI.
> The name of the attribute is, after all, href, right?

No... it's a qualified name {http://www.w3.org/1999/xlink , href}.
Unqualified 'href' is something else entirely :-)

> So, its  making a point. if that point is not clear then it 
> should be reworded, but not omitted. I suggest clarifying as follows:
> In 3.1 first bullet, replace "identified by the XLink href 
> attribute" with "identified by the href attribute in the 
> XLink namespace"


> In 3.1 second bullet, replace "defines the attribute 
> xlink:href" with "defines the attribute href".

Hmmmm.... not sure I'd go for that. href is the local name of the attribute,
but it is not its fullname. Is there a common notation for writing down
qualified names (not qnames, but qualified names).

> That addresses Tim Brays consistency issue, further improves 
> consistency by calling the attribute href as is done in the 
> quoted section of the XLink spec, and addresses my 'no magic 
> prefix' issue and is generally a better example of the spec 
> spelunking needed to demonstrate in full detail how a link is 
> traversed.

BTW: I'm ok with the "Xlink href" attribute as you originally cast it, but
consistency in the 2nd bullet would then suggest "...defines the Xlink href

> -- 
> Best regards,
>  Chris                          mailto:chris.lilley@wanadoo.fr


[1] http://www.w3.org/TR/soap12-part1/#notation
Received on Monday, 1 December 2003 12:41:10 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:40 UTC