Am 1999-06-22 um 0:41 h hat Marc Wilhelm Küster geschrieben:
> E. g. in names, a with umlaut is mirrored to ae on level 1 whereas a with
> trema is mirrored to a

Just to avoid a possible misunderstanding: The standard does not say
"in names" but rather "in Namensverzeichnissen" (in lists of proper
names, aka. rolls). This means that, for any given application, a
version of the ordering rules has to be chosen that will hold for
the entire list to be sorted. Particularly, in an encyclopædia, you
would expect the Umlaut "ä" to be treated as "a", in level 1, for all
catchwords, i. e. likewise for proper names and other entries.

> (that is, a German and a French author are treated distinctly even
> if seemingly spelled alike).  One awkward consequence of this is
> that one of the best accepted national ordering standards cannot
> really become a profile of ISO/IEC FCD 14651.

You are right; I have overlooked this consequence until now.

One more reason to drop the distinction between lists of proper names
and other lists and to treat all Umlauts like their respective base
characters, at level 1, as DIN 5007 proposes for its next edition
(last paragraph in the "Erläuterungen" (explanatory comments) section,
p. 7).

Still, DIN 5007 treats diæresis and Umlaut differently, viz. in level 2.
In level 2, an Umlaut goes before any other diacritic (section 6.2.2),
while a diæresis goes almost last (section 6.2.3.1; only U+030B, U+030E,
U+U30F, U+U310, U+0324, U+0333, U+033F and U+0344 would go after it).

But this difference is so subtle that it will hardly ever result in
different orderings of actual phrases. To take effekt on the ordering,
there  would have to be a pair of phrases,
1. identical in all base characters (level 1),
2. identical in all diacritics, and letter-form variants (level 2),
   up to a character, that has either an Umlaut or a diæresis,
3. having another level-2 difference right of item 2, demanding an
   ordering different from item 2.
I could not come up with any such example in natural languages.

> On the other hand, the knowledge of these facts is vanishing in Germany
> itself [...]

Indeed.

Best wishes,
   Otto Stolz