Re: IRIs and bidirectional formatting characters

--On Wednesday, August 26, 2015 00:39 -0400 Andrew Sullivan
<ajs@anvilwalrusden.com> wrote:

>...
>> Even ASCII, right? rn vs m et al.
> 
> I think that's oversimplifying.  In the case of rn vs. m and
> so on, with a clear font and sufficient size it is at least
> possible to see the difference.  With formatting characters,
> it is not possible _by definition_ to see them, any more than
> control characters and so on.
> 
> The various different confusion problem cases are already hard
> enough without adding the additional complication of lumping
> them all together as though they are one problem.  Since we
> can make a distinction, between these different classes, it's
> probably wisest to keep the distinction in mind.

Let me say that a little more strongly and differently.   It is
possible to usefully address and make progress on "the various
confusion problems" only by making careful distinctions and then
addressing the categories differently.   To at least some
extent, even the Unicode decision to treat all varieties of
compatibility equivalence as part of a single category is not
supportive of sorting out confusion issues.   Whether intended
or not, the only goal that is served by lumping all types of
possible confusion is to "prove" that the problems, as a class,
are so nearly hopeless that we should just give up on all of
them.   I hope that is a conclusion that no one really wants.

>> No, this is still a matter of research. To be clear, the IETF
>> has published
>> 
>>   https://tools.ietf.org/html/rfc3987
>>   https://tools.ietf.org/html/draft-ietf-iri-bidi-guidelines

And, to be further clear, the second of these is a expired
Internet Draft, not something "published" by the IETF or having
any particular standardization or equivalent status.  That
status of the first is, in practice, unclear.  It is nominally
(still) a Proposed Standard but the working group to consider
needed revisions and that would presumably have replaced it to
resolve several issues collapsed without producing a result,
leaving those issues unaddressed and hence the spec somewhat in
limbo.

> There's no question that IRIs are a mess in this aspect, and
> the topic is not getting improved by the fairly low engagement
> around the IETF of people worried about i18n.  It could use
> some help in this area, actually, so if anyone has any spare
> cycles (ha!) I have some mailing lists to suggest.

:-(  But definitely agreed, especially speaking as one of those
who is around the IETF and worried but getting rather thoroughly
disillusioned by the low engagement Andrew describes.

best,
    john

Received on Wednesday, 26 August 2015 19:02:55 UTC