Re: Is this possibly a bug?

[Mark Polesky]

David Kastrup wrote:

> > You are correct. How about "this rule never prints accidentals that
> > [appear/are (already)] in the key signature"?
> 
> Sounds much better than my proposal, but does not cover naturals.
> Whichwere the original point of the original incorrect statement.
> 
> If somebody can propose something which sounds as nice as your variant
> and still conveys the part of the original statement that had been
> correct, that would be perfect.

How about "this rule only prints accidentals for pitches that are not
in the [diatonic scale of the] current key."

> I would probably add to "dodecaphonic" the sentence "Since the key
> signature is essentially ignored, specifying it does not make much
> sense." [Somebody please check whether this is right: if the key
> signature is, say, f major, and we specify dodecaphonic and have a note
> B sharp, do we get # before it (in which case the key signature is
> indeed ignored) or do we get h# before it (in which case it isn't)?
> Also I would consider it likely that the score for transposing
> instruments should _not_ transpose the key, but that is probably better
> done manually).]

With regard to the printing of accidentals, the key signature is not
"essentially" ignored, it is "completely" ignored.

However, a composer using a key signature may legitimately decide to use
"dodecaphonic". Some of the most atonal works come with a key signature.
For example, the first movement of Messiaen's "Quartet for the End of
Time" is (perplexingly) in B-flat major. There are many reasons for
doing this - psychological, referential, humor... So it's a bad idea to
use the phrase "does not make much sense" in the docs (even if we do
feel that way). Better to keep the tone as factual an non-judgemental as
possible. You could instead say "The key signature has no effect; it is
completely ignored."

> But now to 'forget:
> 
> "Like dodecaphonic, this rule does not remember (and consequently
> revert) any accidentals, but prints them only when differing from the
> key signature."

Personally I think what's there now is clearer.

Thanks for your input!
- Mark