Re: [EAI] Proposed resolution of List-* and downgrade- Issues

["Frank Ellermann" <hmdmhdfmhdjmzdtjmzdtzktdkztdjz at gmail.com>]

Charles Lindsey wrote:

> What we would want to write is
>    Archived-At: <http://archive.mañana.com>
> which is an IRI (using UTF-8, as always with IRIs)

JFTR, s/as always with IRIs/as always with EAI/ 

EAI always uses UTF-8, while IRIs can be in any charset
determined by the context.  And for EAI that context is
of course UTF-8.  But in a KOI8-R document it's KOI8-R:

http://hmdmhdfmhdjmzdtjmzdtzktdkztdjz.googlepages.com/IDN-IRI-koi8-r.html

> which, upon conversion to a URI, would simply %-encode
> that 'ñ'.

For an <ipath>, <iquery>, or an <ifragment>.  But your
example has the ñ in the <ihost>, and there you would
use IDNA(bis) rules to convert this into an URL:

Archived-At: <http://archive.xn--maana-pta.com/>

Extended by some Message-ID, as Archived-At gives you
the location per message, not only some archive URL.

Nobody (maybe except some XML and LEIRI geeks) has the
faintest idea what a <http://archive.ma%C3%b1ana.com/>
might be.  It is a "syntactically valid" URL in some
non-DNS parallel universe I don't care about, but it
is not produced by following RFC 3987 rules.

Excluding XML LEIRI abominations unrelated to a "real"
world as defined by DNS, Googlebot, and W3C validator.

> IDNAbis/whoever clarifies the matter. They might well
> decree that valid URIs for that purpose have to be in
> punycode

IDNAbis won't touch URIs with a stick.  The IRI to URI
rules are in RFC 3987, and that uses IDNA for <ihost>:
No percent characters in labels of host names, percent
is not LDH.

> mailto-bis is more likely to be whatever Martin 
> proposes, and we are in a better position to lean on  
> Martin :-) 

Mailto-bis IRIs on a KOI8-R page would be interesting:

If they are transformed to UTF-8 (the first step in a
IRI to URI conversion) they might work with EAI UTF-8.

But if they are transformed to a mailto-bis URL they
might not more work, neither EAI nor 2822upd.

Mailto-bis is very near to hopeless, I'd be delighted
if it at least manages all STD 66 and 2822upd hurdles,
ignoring additional mailto-i18n obstacles.

 Frank
-- 
<http://omniplex.blogspot.com/2008/07/html5-asinine-selfish.html>

_______________________________________________
IMA mailing list
IMA at ietf.org
https://www.ietf.org/mailman/listinfo/ima