[smufl-discuss] Re: Private use area?

[David Webber-2]

From: Daniel Spreadbury

> Dave's argument is that he wants to use a single point size, matching the
staff size, to draw all symbols, which is why his font in Mozart contains
smaller clef glyphs for clef changes. However, he readily admits that his
ambition to use a single point size is thwarted by the requirements of
drawing cue notes and grace notes,  ...

Cue notes yes - grace notes no.

My drawing procedure actually creates three fonts for music items:

1.  for standard music:  this includes special symbols for an acciaccatura
(each way up) with a crossed tail, and small black and white note bulbs for
constructing appoggiature and beamed groups of grace notes.  [And
accidentals are present in the fonts also at "grace note size", which are
also used fro drawing in conjunction with trills, turns and mordents when
required.]

2. for cue notes (at 75% the size of the standard items).   As we discussed,
this is really inescapable as not only notes but also other music symbols
may need to be cued.  Indeed there is the concept of a cue stave: a small
stave showing a whole cued in part, which may need all the symbols of normal
music (including clef changes which are smaller than the normal sized clef
changes).

3. For chord symbols, largely because different authors scale chord symbols
above a part at different sizes relative to the music itself.  But SMuFL
doesn't cover this, so it is irrelevant.

My default choice (for consistent appearance) is that all three are the same
font at different sizes but they don't have to be (and with SMuFL fonts no.
3 cannot be).

The font section is thus schematically:

if( cue item ) use cue font
else if( chord symbol ) use chord font
else use standard font

although cue items and chord symbols tend to be drawn together in groups,
and so font switching can be reduced quite a bit.

> By comparison, SMuFL encodes 41 recommended clefs, and additionally a
further 9 optional stylistic alternates, and 62 optional ligated forms
(primarily to accommodate numbers above and below G and F clefs). Which of
these are "important" enough to require dedicated clef change glyphs? Just
the G, C, and F clefs? What about the historical variants? What about the
clefs denoting the octave transposition of an instrument? What about the
clefs denoting transposition of any interval? <

Just plain C,G,F clefs.

To be brutally honest, I think the inclusion of all these variant clefs is
SMuFL is way over the top, especially considering that the simplest
variation (clef changes are smaller than start-of-line clefs) is omitted.
But (see immediately following)....

My philosophy of the so-called 'transposing clefs' is simply that the
attached 8 (or very occasionally 15) rather than the clef itself,  denotes
"the octave transposition of an instrument" as you very accurately put it.
So Mozart draws this as a separate glyph above/below the clef when required.
But I suppose if some popular software needs single symbol "octave clefs", I
suppose that is grounds for including them.

However, the notation with 8's by clefs, with one exception, is itself
historical, and rendered obsolete by the growth in popularity of transposing
instrument notation since the 18th century.   Publications for old consorts
(recorders, crumhorns, etc) sometimes hang on to the 8s for the octave
transposing members, but those instruments don't employ clef changes so they
don't need small clefs with 8s.   Modern 8ve-transposing instrumental parts
(piccolo, guitar, contrabassoon, etc) don't use 8s anyway (and also don't
change clef).    French horns can change clef, but they sound a 5th below
(or sometimes in obsolete parts 4th above) written pitch, so 8's are out of
the question.  Other clef-changing instruments: cello, bassoon, trombone,...
don't use 8's anyway: in fact the change of clef is there specifically to
avoid having to play at the octave!

The exception I referred to is tenor voice, where the 8 is retained to
emphasise that this is the only choral part which is not at written octave.
Tenor voice tends either to be written alone (open score) sounding an octave
lower than written on treble clef, OR (closed score) together with a
baritone/bass part at pitch on bass clef.   As far as I know it is not usual
to switch between the two conventions in a single written part.   So again
we can do without 8's on clef changes.

Stylistic variations of (eg) C and F clefs:   there are different ways of
drawing these, and I'd have to check what SMuFL has done about it.   My
inclination would be that different clef designs actually belong in
different fonts - like different designs of letters.   But if SMuFL includes
different F clefs, say, then it *may* need matching styles for clef changes.

> It is common to require a variety of different symbols at different
effective sizes even when drawing music for a single staff size.

Not really (and I say this from direct experience).    There are a few
instances only:  notes vs grace notes; accidentals for each;  clefs;  tails
for appoggiature and normal notes; numbers for time signatures and tuplets.
Compared even with your 62 optional clef forms, this is nothing!   And small
clefs are about the commonest.

>...There is also some debate over the exact amount by which the
size of a clef should be reduced for a clef change:

Elaine Gould 'Behind Bars' p7 (yes that early in the book!)  states
two-thirds.   You don't really have to look further.   Again different fonts
could make different choices if they wanted to look stylistically different.

> the arithmetic
required to achieve a different scale factor is complicated considerably
if the clef change glyphs have already been scaled by some arbitrary
amount by the font designer.<

???  For a clef change you just select the clef-change symbol (small clef)
and print it.   No arithmetic necessary - that's the point!

> How is a designer to accomplish a consistent and harmonious typographic
> concept while having no fine control over glyph size for a completely
> common-place use case?

???  Now I'm confused: all the font designer has to so is include the symbol
at 2/3 the size of the other one.   But doesn't his concept apply to every
symbol in the font?   Isn't it the font designer's job to do this?

> The approach I'm taking for Bravura, for what it's worth, is to create a
stylistic set containing optical variants designed to be drawn at smaller
sizes, so that the stroke thicknesses etc. match the larger size. I have
based the glyphs I've drawn so far around a nominal reduction in size to
75% normal size, and adjusted the strokes that they will appear as thick
as full-size glyphs when drawn at 75% of the size. This includes the
common clef glyphs. The plan is for our application to work out when a
glyph is going to be drawn smaller than a certain real size (not point
size, but the actual effective size) and automatically select the optical
variant stylistic set for that glyph. (I believe LilyPond does something
similar, except it switches to a different version of Emmentaler
altogether; being based on Metafont, it's easy for the LilyPond developers
to generate subtly different weights of the font to be used at different
staff sizes, analogous to the different punches used at different rastral
sizes in the days of hand engraving.) <

This sounds unnecessarily complicated!

> Anyway, I remain unmoved on this point. Adding dedicated clef change
glyphs for all clefs duplicates too many glyphs; adding dedicated clef
change glyphs only for the three most commonly-used clefs is arbitrary;

Not so: I have explained exactly why it is not arbitrary above.

> it
is not the case that adding any number of dedicated clef change glyphs
would obviate the need for drawing glyphs at different point sizes for the
same effective staff size; and font designers and application vendors have
the freedom to use the "private use area" if they disagree. <

Which I guess is what I'd have to do.    It seems very inelegant though.

Dave

David Webber
Mozart Music Software
http://www.mozart.co.uk/



#############################################################
This message is sent to you because you are subscribed to
  the mailing list <[hidden email]>.
To unsubscribe, E-mail to: <[hidden email]>
To switch to the DIGEST mode, E-mail to <[hidden email]>
To switch to the INDEX mode, E-mail to <[hidden email]>
Send administrative queries to  <[hidden email]>