If we do this, it should be WE (the IETF engineers) that do it and NOT another thing we request the secretariat to do. We should eat our own dogfood by writing, testing and then GIVING an implementation that is compatible with the current label making system to the secretariat. It's probably not too hard to do, given how the pre-reg badges are produced (I am guessing), but may be much harder for the on-site badges that are made using those little thermal printers. >From experience I can tell you that even adding the 3 extra characters we have in Scandinavian can cause all sorts of interesting problems when mixed with different output equipment. Add multi-byte characters and you may see line-printers shutting down and opening their lids indicating "paper out" or generating spurious form-feeds. OK, so we don't use line printers any more, but you get the idea :-) [If anyone still remembers how to make a line printer attached to an IBM 370 do this by sending just the right sort of code, you get extra points]. The good news is that modern operating systems (ahem) tend to have Unicode and other nice support built-in, but this doesn't necessarily mean that output is a breeze. Ole Ole J. Jacobsen Editor and Publisher, The Internet Protocol Journal Tel: +1 408-527-8972 GSM: +1 415-370-4628 E-mail: ole@xxxxxxxxx URL: http://www.cisco.com/ipj <control>-L here: On Wed, 19 Nov 2003, John C Klensin wrote: > --On Wednesday, November 19, 2003 11:15 -0800 Fred Baker > <fred@xxxxxxxxx> wrote: > > > At 08:23 AM 11/19/2003, Peter Saint-Andre wrote: > >> Proposals for making email addresses fully internationalized > >> were a hot topic in Minneapolis. I'd like to suggest a more > >> modest reform: fully internationalized IETF name badges. > >> IETF 59 might be a fine venue for rolling those out... > > > > No problem, as long as nobody expects anyone in particular to > > actually be able to *read* the name badges. I don't read Han > > (simplified or traditional), Korean, Kanji, Cyrillic, Arab > > (either alphabet) or a variety of other alphabets. I manage > > with umlauts and such, because I can make a noise and the > > other person can say "yes, that's me, the way you pronounce my > > name is...". But I have no clue how to start in a > > non-ascii-like alphabet, and frankly with tonal languages such > > as Chinese my western mouth is likely to injure the person's > > name trying to get it out. > > > > Aside: I had a Taiwanese employee once who would periodically > > give me lessons on how to say her name. It sounded to *me* > > like I was pronouncing it her way. One can only wonder what > > she was hearing... > > > > Just speaking for myself, one of the things I really like > > about name badges is being able to determine, upon inspection, > > what to call the person standing in front of me. > > > > BTW, while I understand that many Asians can read each other's > > writing, I don't think that implies they can read Cyrillic or > > Arab either. They're in a similar boat, if not the same one. > > > > What I would suggest, if we do this, is writing the person's > > name *twice*: once in their native character set, and once in > > a form that an english-reader can read. The latter is an > > established interchange architecture. > > Fred, this is exactly what I was suggesting, only partially in > jest. Native character set, plus punycode, which is much more > precise than a transliteration. If we don't like the punycode > form, we probably need to think about what we are doing to > users in the absence of a serious presentation layer. > > john > >