>» I disagree to the parenthesis. The distinction between upper and lower case
>» letter is usually prescribed by spelling rules etc., whereas the further
>» distinction of small caps is not.  The distinction of upper and lower case
>» is thus more fundamental.
>
>I see what you mean, but you're unfortunately wrong. unicode had to
>choose between a charachter + markup model, or charachter only. In
>fact, `uppercase' is not a charachter concept, but a glyphic one, what
>is at the charachter level is what we call majuscule in french

Personally, I have always found whatever English terminology I have came
across in this matter to be inferior---either there are lots of words for a
concept but none of them fit in any scheme covering all the cases, or there
are no words at all---to what I have encountered in other languages. What I
meant was the character concept, so we may talk about majuscles and
minuscles if you like. I do rather think that upper and lower case are
character concepts though; the glyphic term corresponding to upper case
ought to be (full) capital (but as for what the lower case term would be, I
have no idea).

>(maybe it's what you call uppercase in swedish, after all...),

Directly translated to English, the Swedish terms would be big letter and
small letter. Majuscle and minuscle exist ("majuskel" and "minuskel"
repectively), but hardly anyone uses them. The most common terms in the
typography-aware community is "versal" (upper case), "gemen" (lower case),
and "kapitäl" (small caps), but these are, in your words, glyphical.

>but its use
>may be completely determined by proper markup (end of sentence, proper
>noun, even in german, you'd identify substantives, etc.)

I see what you mean, but you're unfortunately wrong. It is not in general
possible to determine whether a word is a proper noun or not if it appears
in a monocase environment, since quite a lots of proper nouns have once
been nonproper nouns. The reason people can still read it is that they
usually know what to expect, but that's no argument for removing
information. It is usually possible to read a text even if quite a
substantial share of the words, or of the letters, are missing.

My point is that the authorities who define what is proper written language
do this (in the natural languages I know) in a case sensitive way
(distinguish between majuscles and minuscles); they set up rules for which
letters should be majuscles and which should be minuscles. They do not
however set up any rules for which letters should be small caps. I
furthermore conjecture that this is the case for all European languages. It
is because of this that I hold it to be a reasonable approach of unicode to
distinguish between majuscles and minuscles while not having a set of small
caps.

Not that I think the subject is of much relevance, though.

>» Shouldn't c&sc have the same digits as c&lc? (I know foundries like to have
>» c&lc with upper case digits and c&sc with lower case, but there's no
>» typographical reason for that, it's just how they package their product.)
>
>you noticed i was asking for sc digits with the c&sc font, my idea
>being that digits should always be compatible with the surrounding
>average material, hence what glyphs are used to render "small"
>(minuscule) letters.
>
Compability I agree with, but I don't think these sc digits would be any
good. In many fonts these digits would be so small that they would be
almost unreadable! (Not to mension that they, according to you, do not even
exist in most fonts.) There are full capitals in a c&sc font, so why
shouldn't digits be allowed to extend above the x-hieght?

>» One thing which I miss in your list above is all caps for use in lines
>» mainly c&lc, for acronyms and the like. I know some people prefer to set
>» these in sc, but they are intrinsically upper case, so I (any many others)
>» would rather have them in this special all-caps. In practice this special
>» all caps would probably be realised either using medium glyphs or as the
>» all caps at a slightly smaller point size (DEK does this in the WEB user
>» manual).
>
>for that, there should be an all-sc. But that is bad anglo-saxon
>practice, and latex is a good french software, isn't it? ;-)

(Judging from the names of the members of the LaTeX3 project team, I would
rather say it is German software.) I realise a lot of you southern European
( ;-) people would use all-sc for these, but I want to have the medium
all-caps case instance recognised as well.

Lars Hellström