Re: [Ltru] Issue 63: Replace %xgh with U+00gh in text

[Alexey Melnikov <alexey.melnikov at isode.com>]

Peter Constable wrote:

> Addendum: It seems to me that our reference to Unicode should be normative rather than informative.
>  
This is sensible.

> Peter
>
> -----Original Message-----
> From: ltru-bounces at ietf.org [mailto:ltru-bounces at ietf.org] On Behalf Of Peter Constable
> Sent: Wednesday, June 10, 2009 8:17 AM
> To: "Martin J. Dürst"; Kent Karlsson
> Cc: Alexey Melnikov; LTRU Working Group
> Subject: Re: [Ltru] Issue 63: Replace %xgh with U+00gh in text
>
> From: ltru-bounces at ietf.org [mailto:ltru-bounces at ietf.org] On Behalf Of "Martin J. Dürst"
>  
>
> > > Nit-pick: Unicode (and ISO/IEC 10646) uses tuple syntax when denoting
> > > character sequences. So:
> > >
> > >     lines containing only the sequence "%%" (<U+0025, U+0025>).
> > > ...
> > >     The sequence '..' (<U+002E, U+002E>) in a field-body denotes a range of
> > >      
> > >
> > > >    other subtags in a tag by a hyphen ("-", [Unicode, p. 1077] U+002D).
> > > >        
> > I'm fine with adding the comma. I don't think the IETF uses '<' and '>', 
> > so I'd prefer to leave them out.
> >    
> I'm OK with the comma; I agree that this is IETF, not Unicode, so we should adopt IETF conventions.
>  
>
> > > Still nit-picking: I'm not too keen on including a page number. This
> > > is just as valid for other editions of TUS.
> > >      
> > Fine with me to leave that out.
> >    
> I also would leave out the page number.
>  
>
> > > And does the reference have to
> > > be placed so awkwardly?
> > >      
>
> I also wonder why have the reference to Unicode here--we wouldn't include it with every reference to an individual character, but at the same time there's nothing about a reference here that indicates the same reference applies in every other mention of specific characters. Rather, I think it would make more sense to give a reference to Unicode at the end of the introduction:
>
> <suggested text>
> Specific characters that are mentioned in this document are to be interpreted as characters in the Unicode coded character set ([Unicode]). Character references within ABNF will follow conventions for ABNF given in [RFC 5234]. Character references outside ABNF will use a literal quotation and include the Unicode code position, as recommended in RFC 5137. For example, "-" U+002D.
> </suggested text>
>  
Sure.

> In other respects, I'm OK with the changes Martin suggested.
>