OK, but one question is:  Suppose a program sees:

M:2/4 Q:1/4=120
   ...
M:6/8 Q:3/8

With just this information, should  the  second  be  displayed  as  a
dotted  eighths note, or as <quarter> = <dotted eighths>?  This might
not have been at all what was intended. The intent could have been to
switch  to  6/8 but not specify a tempo.  I can see this leading to a
lot of confusion and wildly different implementations.

This is personally mostly an "intellectual interest" point;  I  don't
really  see  myself  using  this  much,  since  I  rarely feed ABC to
players. But this could easily lead to bad ABC on my part.  If I were
transcribing  some  printed  music and saw a tempo indication such as
1/4=3/8 (as notes of course), I'd probably do my  best  to  translate
this to a Q line.  But I'd probably get it wrong, since I wouldn't be
a regular user of this feature. This is similar to all the extant ABC
that has wildly incorrect tempos in the current notation.

In any case, I'd prefer that the software just do what  I  type,  and
not  try  to outsmart me.  If I write Q:1/4=3/8, I'd hope the obvious
two notes and '=' would be displayed. If I write only Q:3/8, I'd hope
that  only  a  dotted  eighths  note  would be displayed, despite its
obvious lack of meaning to me and most other musicians.

It does occur to me that if the software will just  blindly  put  out
the  contents  of  the  Q  line with fractions converted to notes, it
would work to notate the conventional Balkan rhythm notations:
  Q: 4/8 3/8 4/8=70

Let's see if it works with abc2ps ...

...  Nope.  But I spent about 20 minutes looking at the write_tempo()
routine, and I figured out why it failed.  It was just a missing test
of the return value of sscanf().  It's  fixed  now,  and  looks  real
pretty.  Since my clone (jcabc2ps) is what my Tune Finder uses, I can
now use this notation on the Web and Balkan musicians  will  get  the
notation  that  they expect.  I think I'll now go off and add this to
all the irregular tunes in my .../Intl/ directory.

Any other Balkan musicians out there who think this would be a useful
thing  to somehow get into the ABC standard?  Or maybe we should just
say "Hey, it's not a violation of the current standard; it's just  an
undocumented  case which obviously oughta work.  It is standard music
notation, after all."

(I'm glad you encouraged me to look into this. ;-)


Laurie writes:
| This is back to the question of "does the notation tell the program what to
| print or does it describe the music".  The answer is that it describes the
| music and software can figure out how to express that in any other medium
| (for instance sound waves, MIDI codes or tadpoles on 5 bar gates).  I
| *suggest* that a good thing to print for this case would be
| <note shape denoting old beat> = <note shape denoting new beat>
|
...
I wrote:
| To get it treated as anything other than a mistake, you'd have to say
| Q:1/2=3/2 and the displayed/printed output would have to show the two
| notes with the '=' in between. (And I suppose we should also state in
| the  spec that the first length is the old beat and the second is the
| new, or half the people writing abc players would do it the other way
| 'round. ;-)
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html

Reply via email to