John C Klensin wrote: > Ned Freed wrote: [...] >> To the extent RFC 1345 is problematic, it is because its domain >> of applicability is quite limited. But within that narrow >> domain it actually can perform a useful function. > Agreed. And perhaps that suggests a way forward if people are > willing to do the work (from my point of view, your efforts and > Ben's would probably be most of what is needed; I'd be happy to > review and maybe make whatever small contributions I could). +1 A version of Lynx uses these mnemonics for Unicode points when it's forced to limit its output to "codepage 850" characters. The effect was acceptable for en and de documents, or at least better than displaying question marks. I'd probably prefer to see a clearly delimited hex. representation in most cases, but that's a matter of taste and besides not the application Ben has in mind. So if somebody has a real application for these mnemonics they'd clearly wish that it's up to date and correct as far as possible. An I-D also discussing the alternatives and limitations would be fine. > a set of code-mnemonics for a specific, identified, subset of > characters is reasonable and a core problem with 1345 is that > it attempts to be much too broad. I'm not sure about "reasonable", but it's certainly "possible". > If an offspring of 1345 were defined as having, e.g., only > European languages using Roman-derived scripts in scope, then > I think there is more than adequate expertise around to review > a proposal and sufficient stability to not be affected by Unicode > changes, and that there should be no significant issue with IETF > sign-off or even standardization. Drawing the line will be difficult, RFC 1345 tried to cover more than MES-1. Frank _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf