RE: Extension Relation Type Comparison - LC Comment on draft-nottingham-http-plink-header-07

> -----Original Message-----
> From: Mark Nottingham [mailto:mnot <at> mnot.net]
> Sent: Wednesday, January 27, 2010 8:28 PM
> To: Eran Hammer-Lahav
> Cc: apps-discuss <at> ietf.org
> Subject: Re: Extension Relation Type Comparison - LC Comment on draft-
> nottingham-http-plink-header-07
> 
> 
> On 26/01/2010, at 4:03 AM, Eran Hammer-Lahav wrote:
> 
> > The current text about comparing extension relation types is unclear:
> >
> >   When extension relation types are compared, they MUST be compared as
> >   URIs in a case-insensitive fashion, character-by-character.  Because
> >   of this, all-lowercase URIs SHOULD be used for extension relations.
> >
> > What does it mean "compared as URIs"?
> >
> > It is clear that these two URIs would be deemed equivalent:
> >
> > http://example.com/rel/type
> > HTTP://example.COM/rel/TYPE
> >
> > But are they also equivalent to:
> >
> > http://example.com:80/rel/type
> 
> None of those are equivalent; it specifies case-insensitive, character-by-
> character. "As URIs" alludes to the fact that an extension type might be
> serialised in a non-URI form; e.g., as a CURIE, if that's your cup of tea.

I have talked to a few folks about this and they all assumed 'as URI' means that outside the lower-case
exception, normal URI comparison rules apply. I agree with your intention though that these should be
compared as case-insensitive strings.

EHL