At 16:07 06/12/12, Doug Ewell wrote: >Martin Duerst <duerst at it dot aoyama dot ac dot jp> wrote: > >> What about the following compromize: >> >> 1) For names that consist of two pieces, list the uninverted form, >> assuming that it can always be inverted. > >Fine. > >> 2) For names that consist of three or more pieces, list the uninverted >> form as well as all the possible inverted forms. Sorry, by "possible", I meant "sensible", which I think we can take as "listed in the source standard". Sorry for the confusion. Regards, Martin. >Not fine. Apparently my earlier point about "and Barbuda Creole English, Antigua" was a bit too subtle. Let's try another example with ALL POSSIBLE inverted forms: > >Description: Santa Ana de Tusi Pasco Quechua >Description: Ana de Tusi Pasco Quechua, Santa >Description: de Tusi Pasco Quechua, Santa Ana >Description: Tusi Pasco Quechua, Santa Ana de >Description: Pasco Quechua, Santa Ana de Tusi >Description: Quechua, Santa Ana de Tusi Pasco > >Silly, isn't it? And not all of them even make sense. > >And as I said before, if you really want ALL POSSIBLE inverted forms -- >for whatever reason -- there is no need to list ANY of them, for the same reason that there is no need to list all possible inversions of a two-word name: because it is obvious. Listing all of these redundant forms would do nothing but chew up space in the Registry, which will be big enough already without them. > >There are 2 six-word names, 12 five-word names, 98 four-word names, and 504 three-word names. That's a lot of pointless repetition. And that doesn't even count hyphens. > >I understand adding only the uninverted form, or only the inverted form(s) given in the source standard. I could even *sort of* go along with adding all the versions from the source standard, inverted and uninverted, although that seems quite unnecessary. But creating our own inverted forms, just to make sure every mathematical indexing possibility is covered, makes no sense at all. Strongly opposed. > >> There are variants on this, e.g. in 2), don't list the uninverted form to reduce duplication. The proposal is based on the assumption that two-piece names can always be inverted, and that names with three pieces or more are relatively rare. If that's not true, please correct me. > >See statistics above. > >> This would allow to hopefully satisfy both those who want to see the uninverted names, because they are more straightforward, and those who want to be able to invert them without creating wrong inversions. > >My preferences, in order: > > 1. uninverted forms only > 2. inverted forms from 639-3 only > 3. both 1. and 2. above > 999. anything else > >-- >Doug Ewell * Fullerton, California, USA * RFC 4645 * UTN #14 >http://users.adelphia.net/~dewell/ >http://www1.ietf.org/html.charters/ltru-charter.html >http://www.alvestrand.no/mailman/listinfo/ietf-languages > #-#-# Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University #-#-# http://www.sw.it.aoyama.ac.jp mailto:duerst at it.aoyama.ac.jp _______________________________________________ Ltru mailing list Ltru at ietf.org https://www1.ietf.org/mailman/listinfo/ltru
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.