Changing DISALLOWED (was Re: Reserved general punctuation)

Debbie Garside debbie at
Sat May 3 00:24:21 CEST 2008

Hi Andrew

Originally I came in on this in support of not having DISALLOWED completely
set in stone - for ever.  I think everyone seems pretty much on the same
wave length wrt this now.  I have no problem with changes taking 6 months
but I think 2 years is pretty steep if you look back at Mark's original
argument.  Balance is definitely the key.  If there is scope for putting in
an update mechanism within this RFC then it should be done.  However, if it
proves too complex and will delay the end result beyond acceptable
timescales then the minimum that should be added is the scope to update the
RFC should it prove necessary in the future.  I do think that going through
the complete RFC process for one character would be pretty steep too!  To go
back to Mark's original argument, if it were your language you would want it
but would IETF setup another WG for one character - I don't think so!  This
is where a mechanism would help.

That said, I have no idea whether this scenario will come up a week after
the RFC is published or 10 years after or never.  But the point is, neither
does anyone else.



> -----Original Message-----
> From: idna-update-bounces at
> [mailto:idna-update-bounces at] On Behalf Of
> Andrew Sullivan
> Sent: 02 May 2008 21:20
> To: idna-update at
> Subject: Re: Changing DISALLOWED (was Re: Reserved general
> punctuation)
> On Fri, May 02, 2008 at 05:58:53PM +0100, Debbie Garside wrote:
> > leisure.  I just think that some sort of mechanism for updating
> > according to "Mark's 3" should be incorporated unless we can
> > realistically say that an RFC can be delivered within the
> timescales
> > you specify - in which case I would be happy.
> I'd like to hear the reason why anything under 6 months is
> really important for the case I thought we were talking
> about, which is the one where things move from DISALLOWED to
> PVALID (maybe with some context rules).
> Remember, the reason the code point is DISALLOWED in the
> first place is that it's supposed to be somehow harmful to or
> incompatible with the DNS.  This doesn't mean it's not
> normally in a word, or anything else -- it means "harmful to
> the DNS".  As John said several times in Philadelphia, you
> can't write literature in the DNS.
> Therefore, if we assume that "internationalized LDH" is what
> we're aiming for, and something somehow makes it incorrectly
> onto the list of "not-iLDH, therefore not allowed", I think
> taking a year or two to change that is perfectly acceptable.
> (If you don't actually like the LDH-to-iLDH approach, that's
> a different matter.  That's an argument that this WG needs to
> be rechartered instead.)
> If taking a couple years to get such a change right is
> acceptable, what is the purpose of designing option 3 now?  I
> grant John's observation that it might turn out that we need
> such a mechanism, because these cases are really common after
> all.  But surely it's prudent to get some experience before
> inventing that mechanism?
> A
> --
> Andrew Sullivan
> ajs at
> +1 503 667 4564 x104
> _______________________________________________
> Idna-update mailing list
> Idna-update at

More information about the Idna-update mailing list