Re: [idn] new I-D: Safely Encoding of likeness information into ACE label version 0.2

["Soobok Lee" <lsb@postel.co.kr>]

In summary, I suggest two principles:

  -  disambiguate as many look-alike letters as possible
       only by assistance of  reference fonts and GUIs

  -  normalize as few look-alike letters as possible :
      - only these cases for _identical_ glyph:
        - borrowed glyph
        - sibling glyphs from the same origin

Any constructuve modfication wanted...


Soobok Lee

----- Original Message -----
From: "Mark Davis" <mark@macchiato.com>
To: "Soobok Lee" <lsb@postel.co.kr>
Cc: <idn@ops.ietf.org>
Sent: Wednesday, August 01, 2001 12:31 AM
Subject: Re: [idn] new I-D: Safely Encoding of likeness information into ACE label version 0.2


> Such a disambiguating font is certainly possible, and one of the possible
> approaches to GUI-assists. What we could do is add text to IDNA to recommend
> that domain names be displayed in such a way as to disambiguate confusibles.
>
> Mark
> —————
>
> πάντων μέτρον ἄνθρωπος — Πρωταγόρας
> [http://www.macchiato.com]
>
> ----- Original Message -----
> From: "Soobok Lee" <lsb@postel.co.kr>
> To: "Mark Davis" <mark@macchiato.com>
> Cc: <idn@ops.ietf.org>
> Sent: Tuesday, July 31, 2001 07:31
> Subject: Re: [idn] new I-D: Safely Encoding of likeness information into ACE
> label version 0.2
>
>
> > Mark,
> >
> > How about recommending a disambiguaing  full-unicode-wide reference font
> set
> > for IDN
> > with the best font size in IETF implementation guidelines for IDN-aware
> > OS/application vendors ?
> > The font set will help to reduce the list of pairs of ambiguous characters
> to be
> > compared,
> > and encourge all application vendors to adopt  it as the default IDN font.
> > For example, Greek 'rho' and latin 'p' may  look very different in that
> font,
> > while
> > Cyrillic 'a' (borrowed from latin 'a') and Latin 'a' may still look
> identical.
> > Then
> > we need normalize only 'a' s.
> >
> > Do think this is feasible ?
> >
> >  I found  http://www.unicode.org/charts already uses such good font. Is
> the font
> > for public domain?
> >
> >
> > Soobok
> >
> >
> >
> > ----- Original Message -----
> > From: "Mark Davis" <mark@macchiato.com>
> > To: "Soobok Lee" <lsb@postel.co.kr>; "Eric Brunner-Williams in Portland
> Maine"
> > <brunner@nic-naa.net>
> > Cc: <idn@ops.ietf.org>
> > Sent: Tuesday, July 31, 2001 10:56 PM
> > Subject: Re: [idn] new I-D: Safely Encoding of likeness information into
> ACE
> > label version 0.2
> >
> >
> > > This is a pipe-dream. See my messages of
> > > Wednesday, July 18, 2001 09:45
> > > Thursday, July 19, 2001 08:20
> > >
> > > One might be fooled by how easy it is to come with with a simple list
> > > between two scripts, such as Greek and Latin. But then the complications
> > > arise: are rho and p to be considered confusable? Then there are
> extended
> > > characters, like Latin small letter open e (ɛ) and epsilon (ε).  You
> rapidly
> > > (I almost wrote "rabidly") slide off into gray areas even with just two
> > > scripts. When it comes to comparing all the characters of the world, in
> many
> > > combinations of fonts, the problem quickly becomes intractible.
> > >
> > > We have had a lot of experience dealing with huge numbers of characters.
> It
> > > is harder than it looks.
> > >
> > > Mark
> > > —————
> > >
> > > πάντων μέτρον ἄνθρωπος — Πρωταγόρας
> > > [http://www.macchiato.com]
> > >
> >
> >
>
>