--On Monday, 21 November, 2005 11:16 -0800 "Hallam-Baker,
Phillip" <pbaker(_at_)verisign(_dot_)com> wrote:
I think that a better case to make wrt internationalization is
that it is hard to see how a pure ASCII document is ever
going to provide a satisfactory description of a protocol that
is based on unicode.
This is an entirely different issue from the one that I think
Jefsey was raising and Tom was responding to. Conflating them
gets us nowhere at all.
With regard to the "protocol based on Unicode" issue, first, the
IETF has very little experience so far with such protocols.
"Based on Unicode", to me, is a different situation from "being
able to transmit Unicode". The latter can be easily described
in a pure ASCII document, even though illustrative examples "in
Unicode" might make things more clear. The comments below
could, in principle, change as IETF experience accumulates
although, personally, I'd be a little surprised.
Suppose one had a protocol that was actually "based on Unicode".
Issues with look-alike characters, font variations, and
variations in coding and compression (normalization issues,
UTF-8 versus UTC-2 versus UTC-4 and potentially other
variations), etc., would nearly force characters to be
identified in code point terms, e.g., as U+NNNN, rather than, or
in addition to, just showing the characters in some preferred
font. It is worth noting that the textual parts of the Unicode
standard itself are written in just that way. So, again, the
non-ASCII text becomes an extremely useful illustration of what
is going on or part of illustrative examples or tables. The
text may be easier to follow if those illustrations are present
than if it is not, but everything normative can still be
expressed in ASCII.
And that takes us back to Paul's original comment, at least as I
understood it -- the important thing is the quality of the
writing. Access to fancy formatting and presentation tools may
make good writing easier to follow and, in particular, to let
the reader get the gist of what is going on before trying to
deeply study the text. But adding fancy formatting and
presentation to poor writing will, at best, only hide, for a
while, the fact that the explanations are inadequate. And, by
not forcing the extra discipline that writing without a
dependence on clever illustrations requires, it may make it more
likely that we will get documents whose inadequacies are harder
to detect.
As I said, that conclusion could change as experience with
"protocols based on Unicode" expands. But, so far, we have
almost no experience along that dimension (and, incidentally,
neither do ISO or ITU or IEEE, at least as far as I know).
john
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf