On 30/01/2004 15:29, Michael Everson wrote:I see from a brief comparison of the IPA chart (http://www.arts.gla.ac.uk/IPA/fullchart.html) with the Latin alphabet collation chart that there seems to be a general principle that separate IPA characters are collated separately at the first level. Apart from diacritics, I can see only one exception to that principle, c cedilla. I assume that this was a deliberate decision, and it seems to be a sensible one for IPA usage. This principle applies including the 8 or so small caps in IPA. Other UPA variants e.g. rotated letters are sorted separately at the top level, and so I suppose that the same principle applies to UPA.
At 15:00 -0800 2004-01-30, Peter Kirk wrote:
Nor is that how mathematical alphanumeric symbols are used. But they are still given compatibility and collation data as if they were. I am simply looking for some consistency, and less confusion for the ordinary user of the collation charts who shouldn't see these letters highlighted at the top level, but rather hidden among a whole lot of other font variants used only for special purposes.
Wrong. The math symbols are not used in words in plain text which are conventionally sorted. The Uralic Phonetic Alphabet letters certainly are.
Is the issue that fixing these weights is more bother than it's worth, as Ken suggested? Or is it that these weights are actually correct, because it is what is wanted by the users, the UPA community (and for 11 of the which are in the IPA block, the IPA community)? If the latter, I withdraw my objection. But I would want to see some evidence that the phoneticians actually want these to be sorted as separate letters at the top level. Have they been asked?
The change I was suggesting, to treat small caps as a font variant, would compromise this principle for both IPA and UPA. For this reason I want to withdraw the suggestion.
-- Peter Kirk [EMAIL PROTECTED] (personal) [EMAIL PROTECTED] (work) http://www.qaya.org/
