Re: Fwd: change "URL" to "web address" throughout the HTML 5 spec (Issue-56 urls-webarch)

As discussed in the last telephone conference, I have updated the
requirements document and the specification with a short paragraph in the
Terminology section that explains that we use "URI" where "URI reference"
may need to be used.

Cheers,
Silvia.

On Sun, Aug 23, 2009 at 12:30 PM, Silvia Pfeiffer <silviapfeiffer1@gmail.com
> wrote:

> On Sun, Aug 23, 2009 at 7:26 AM, Raphaël Troncy<Raphael.Troncy@cwi.nl>
> wrote:
> >> The term URI doesn't seem to include relative references according to
> >> what I forwarded. So, the creation of web addresses such as
> >> "../test/video.ogv#t=12.50" is not covered when using the term URI.
> >> This was what triggered my email.
> >
> > I'm not sure I understand the issue :-(
> > Do you claim that: ./resource.txt#frag01 is *not* a valid URI?
>
> Yes, it's a valid URI reference, but not a valid URI.
>
> > It is according to Wikipedia,
> >
> http://en.wikipedia.org/wiki/Uniform_Resource_Identifier#Examples_of_URI_references
>
> Not quite.
>
> According to the standard, URIs and URI references are not the same,
> see http://en.wikipedia.org/wiki/Uniform_Resource_Identifier#URI_reference
> (also states "protocol documents should not allow for ambiguity").
>
> When we talk about fragments, we actually always talk about URI
> references. "In order to derive a URI from a URI reference, software
> converts the URI reference to "absolute" form by merging it with an
> absolute "base" URI according to a fixed algorithm." Take a look at
> the standard to see the difference:
> http://labs.apache.org/webarch/uri/rfc/rfc3986.html#uri-reference .
>
> So, if we want to be correct, we should use "URI reference" everywhere.
>
> Silvia.
>