Re: FW: notes from IRI meeting

This represents consensus among representatives from the major browser
vendors and search engine vendors. We also have consensus from the DE and AT
NIC's that mapping eszett to "ss" is the correct approach.

While clearly not a perfect solution, it appears to be optimal considering
all the different factors involved.

Mark


On Thu, Nov 5, 2009 at 09:41, Larry Masinter <masinter@adobe.com> wrote:

> I was going to try to summarize this issue and its resolution
> or hoping the participants would, but I'm not sure what that is.
>
> Larry
> --
> http://larry.masinter.net
>
>
> -----Original Message-----
> From: Ian Hickson [mailto:ian@hixie.ch]
> Sent: Tuesday, October 20, 2009 3:49 PM
> To: Mark Davis ☕
> Cc: Erik van der Poel; Larry Masinter; duerst@it.aoyama.ac.jp
> Subject: Re: notes from IRI meeting
>
> On Tue, 20 Oct 2009, Mark Davis �~X~U wrote:
> >
> > A post-processed URL would not have ß in it. So if you are using only
> > that, or using reverse-DNS lookup, then the display form is lost and you
> > would see "ss", just like you would see case or variant information
> > lost.
> >
> > If we didn't have the IDNA2003 compatibility issue, then retaining ß
> > wouldn't be a problem. We know that the display approach is not perfect;
> > the question is whether it is sufficient to avoid the "same URL going
> > two places" problem, while usually providing the ß/ss distinction to
> > people.
>
> We should make sure we have the buy-in from browser vendors then. If I was
> a browser vendor I'd be very skeptical about a situation in which the
> displayed URL can change without the underlying URL changing, just because
> a script happened to manipulate the URL before the user clicked it.
>
> (Feel free to forward anything I wrote on this thread to a public list.)
>
> --
> Ian Hickson               U+1047E                )\._.,--....,'``.    fL
> http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
> Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
>

Received on Thursday, 5 November 2009 19:32:33 UTC