<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2802" name=GENERATOR></HEAD>
<BODY style="MARGIN: 4px 4px 1px; FONT: 10pt Microsoft Sans Serif">
<DIV>Thanks James,</DIV>
<DIV>Good questions all. As with all standards, the language question is an important one, and not one to be taken lightly. My current opinion (an only my personal one) is that we should decide on the standard elements in one language (and English appears to be the one that continues to surface, whether we all like that or not) and then figure out (and as technology character sets--Unicode not yet being universally available) how to make those work in all languages. I also personally believe that is important to include the source language and characters and to translate (or transliterate) those into other language(s)...whether we choose a single language or go for the UN languages.</DIV>
<DIV> </DIV>
<DIV>However, the latter is a long range goal and we rely on those of you who are fluent in those languages to assist us, now, and in making the standards work in other languages (and probably metadata traditions).</DIV>
<DIV> </DIV>
<DIV>ISO dates are an interesting question. For taXMLit, we decided that we should capture dates as they are in the original publication and that any conversion/interpretation belongs in another (and later, still to be developed--see the explanation document) layer. For other standards, we might consider a different solution, but for a full content standard, I believe it best to record what is there and then to add one or more interpretations to other standard(s) or opinions for someplace else (which is closely linked and equally easily acceptable).</DIV>
<DIV> </DIV>
<DIV>Cheers,</DIV>
<DIV>Anna<BR><BR>>>> Nozomi Ytow <nozomi@biol.tsukuba.ac.jp> 09-Feb-2006 2:11:16 PM >>><BR>Dear all,<BR><BR>thanks for documents, Anna.<BR><BR>The following questions came to my mind. These<BR>are relevant to Level 1, but some also relevant<BR>to contents of higher levels.<BR><BR>Do we assume ASCII only for human-readable portion,<BR>or do we allow non-ASCII including east European,<BR>Cyrillic and Asian characters?<BR>If we allow non-ASCII, do we expect character<BR>normalisation (senu Unicode)? Two or more code points<BR>for a single character is not uncommon in Unicode.<BR>If we allow non-ASCII, do we need to restrict some fields<BR>to ASCII, even if the fields values are non-ASCII in <BR>the original literature?<BR><BR>Do we use standards such as ISO to display date?<BR><BR>Cheers,<BR>James<BR><BR><BR><BR><BR><BR>_______________________________________________<BR>TDWG-Lit mailing list<BR><U><A href="mailto:TDWG-Lit@lists.tdwg.org">TDWG-Lit@lists.tdwg.org</A></U> <BR><U><A href="http://lists.tdwg.org/mailman/listinfo/tdwg-lit_lists.tdwg.org">http://lists.tdwg.org/mailman/listinfo/tdwg-lit_lists.tdwg.org</A></U> <BR></DIV></BODY></HTML>