Re: accent and marcato shouldn't be quantized

[Mark Polesky]

Neil Puttock wrote:
> LGTM.
>
> > do I need to do anything described in CG 8.7 "Adding or
> > modifying features"?  Should I add a @item to
> > changes.tely?
>
> No and no. :)


Well, I did a little more research into this and
unfortunately the situation is a little more complicated
than I had initially hoped.  Ironically, the only clear
thing concerns neither the accent nor the marcato, but the
staccatissimo script.  I'll get to that...

First I need to mention that Kurt Stone and Ted Ross
contradict each other on some points.  By way of visual
example, Kurt Stone implies that the `marcato' should always
appear outside the staff, however his example is misleading
because it shows only the most obvious configuration.

Ted Ross (whose book precedes Stone's by a decade) is
careful to show more exceptional circumstances and states
that the marcato "can be treated similarly to [the accent],
although it can be used within the staff more freely.
Indeed, scanning through my score collection (mostly
Breitkopf & Härtel) revealed far more marcato's inside the
staff (where applicable) than outside.

On the topic of the `accent' script, Kurt Stone writes:
  Although accents are usually placed outside the staff,
  there are situations when it is necessary or desirable to
  place an accent inside the staff, in which case it should
  appear in a space rather than on a line and should never
  be closer to the note-head than a fourth.

The Ross book confirms this but further implies that the
distance between the note-head and the accent should not
exceed a seventh.

To this I'll add my own observations:

In the simplest situations, the accent script appears
outside the staff, even when there is space for it in the
staff.  However, there are a number of situations which may
influence (but not require) a decision to set the accent
inside the staff:
  1) the note-to-script distance exceeds a seventh.
  2) the accented note is part of a polyphonic passage.
  3) the accent is combined with another articulation that
     is normally inside the staff (like a staccato or
     tenuto).
  4) a better slur or tie would result (in which case all
     accents under the slur or tie should be set inside the
     staff).
  5) an ambiguity would be prevented with a nearby note from
     another staff.

On the subject of the staccatissimo script, Stone claims
that it "almost always appears outside the staff" whereas
Ross asserts the opposite: "it is placed next to the note
head wherever possible".  Again, Ross's statements are
better supported in the literature: of all the samples I
found that could fit in the staff (among Schubert, Chopin,
Liszt, and Prokofiev), only one (!) was outside the staff
(near the end of Prokofiev's 8th sonata), which I assume was
an engraver's oversight.


************************

So, I'm changing my initial proposal.  Now I think:
1) marcato should remain quantized.
2) staccatissimo should become quantized.
3) accent should either be
   a) de-quantized, or even better,
   b) programmed "smart", ie. conditional quantizing
      according to the points above (how hard would
      this be?  Are any of the points manageable?)

Comments?

Thanks
- Mark