Re: i18n name badges
2003-11-23 05:11:35
John,
Ah I see now. Yes, there is leakage in Punycode now and will be for a
while. It is not "nice" but I couldnt think of any encoding which wont
leak. (UTF-8 will give you gibberish if client are not UTF-8 aware or
with the right fonts). Same arguments we have in IDN WG many years ago.
We want to put punycode on the badge so that "eat our own dogfood", then
yes, maybe we should. But given the space constraint on the badges, I
rather put something more meaningful, like an .PNG of the person name.
-James Seng
John C Klensin wrote:
James,
My apologies for being a bit cryptic -- I hoped people would understand
the issues well enough by now to get the point. The difference between
the design of --and hopes for-- IDNA and what is happening is something
the community needs to understand and be better informed about. The
design of Punycode was for machine-to-machine communication, as you
point out. It was never really intended or expected to "leak" into use
environments. Indeed, it was chosen, in preference to some other
options, on the grounds that
* it could be deployed quickly,
* applications could (and would) be speedily updated to
support it and provide proper "native" character set
presentation to users,
* when it did leak, users would find that acceptable
because they could always convert to and from
native/local characters with cut and paste and
readily-available tools.
The reality so far is different. The IDN browser world has turned into
one of conflicting plug-ins, apparently none of which are fully
satisfactory. For other applications, the conversion/ upgrade rate has
been low and most estimates seem to point to its being at least 18 more
months before we see significant progress in applications that have
really wide deployment (and estimates that optimistic often depend on
other conditions or contingencies that might or might not happen). So
IDNA/punycode is leaking a lot, and can be expected to leak a lot for
the next several years, at least. And that implies that the IETF
community should probably understand what it looks like and what the
issues are in converting to and from it, including the fact that, for
expression of "names", nameprep is a lossy conversion. And that is
precisely the "eat our own dogfood" point. More important, we need to
think about --and probably experience-- that point as we consider
approaches for email local parts, which more often contain names about
which people are _really_ sensitive.
john
--On Thursday, 20 November, 2003 11:05 +0800 James Seng
<jseng(_at_)pobox(_dot_)org(_dot_)sg> wrote:
I think having the punycode form have no "value" on a name
badge. Punycode, as it is designed, is meant for
machine-to-machine communication.
But I like the idea of allowing participation to put their own
native names together with their ASCII version on the name
badge especially for the next IETF.
But...
1. What if the person don't have ASCII only name?
(e.g. Patrik Fältström)
2. What if the person have a name which the computer cant be
printed?
(Maybe it is not in ISO 10646, maybe it is but there is no
fonts,
maybe the fonts is there but the rendering is wrong?)
Since we are on the topic of the name badge, it is possible to
somehow tag the family name of the person? (e.g. underline?
bold? captialized?) Not everyone follows the <Last Name>
<First Name> convention. In fact, the concept of "First and
Last" name is quite alien to me.
-James Seng
Dave Crocker wrote:
Fred,
FB> What I would suggest, if we do this, is writing the
person's name *twice*: FB> once in their native character
set, and once in a form that an FB> english-reader can read.
The latter is an established interchange architecture
I believe that was the intention in the proposal. List names
in the same way we always have, AND list them in their
"native" form.
Whether it would helpful to provide a third form -- the ascii
encoding of the native form, as it would be seen in an email
address header -- is a separate question.
/d
--
Dave Crocker <dcrocker-at-brandenburg-dot-com>
Brandenburg InternetWorking <www.brandenburg.com>
Sunnyvale, CA USA <tel:+1.408.246.8253>
|
|