[smufl-discuss] Re: Clef change glyphs, and SMuFL 0.9

[notenlektorat]

[Ok, I realize in hindsight that I shouldn't have put off sending this for a
week, but anyway, here are (regardless of their current value) some
additional thoughts from me concerning the Great Clef Change Crisis of
2014:]

After reading up a bit on what a stylistic set actually is I have to admit
that I have come around to Daniel's point of view [that is, by now, his
original view of the matter]. I am not sure, though, if I have understood it
completely. Would it be correct to say that the three stylistic alternates
given for accidentals (uniE240.ss0, uniE241.ss01, uniE242.ss01) are not
necessarily an exhaustive set and could therefore be augmented by a font
designer as preferred (for example: the double sharp could be included as
uniE243.ss01, even though it is not explicitly presented in the
documentation)? If so, I would be happy to scale down my request in that
matter to simply including a similar set of examples for stylistic
alternatives for clef changes (or rather: clefs of a smaller size, see
below) within the SMuFL documentation, for the sole reason of raising
awareness for the issue with any font designer using those guidelines and to
encourage them to actually include such a set, since the use of smaller
clefs is primarily not a question of typography, but a necessity of
notational grammar.

In favour of including clef changes as glyphs of their own there has been
made the comparison of differently sized clefs with upper and lower case
letters. Upon some reflection I must say that this argument does not hold:
an upper and lower case letter pair will actually consist of two completely
different symbols. But the clef used for clef changes is exactly the same
symbol as the standard size clef, there just is a convention to show it
smaller when there is a clef change. And while I formerly was of the opinion
that inclusion of special clef change glyphs would be justified because of
them having their own unique meaning, that is actually not correct: the
meaning of a smaller clef is not "here be a change of clef", but rather
indication of a reference pitch's position on the staff, which is in fact
the exact same meaning that the regular sized clef has (the clef change is
just the result of this indication).* It's not even another symbol with the
same meaning (like the variety of quarter tone accidentals or standard flags
versus straight flags). It is literally the same symbol for the same
meaning. Since the meaning of the symbol does not change, but only it's
typographical context (and with that a single aspect of its appearance),
this would be more like super- or subscript, which, as I believe, is not
implemented with duplicate code points for every possible symbol.**

In this respect, smaller clefs are quite different from other musical
symbols that just may happen to have identical proportions, but different
sizes (or even the same size). The ornament comma looks just like the breath
mark and the comma for time signatures. A plus symbol denotes an actual plus
(again within the context of time signatures) and it also stands for the
concept "closed" in wind notation as well as the "dead stroke" in percussion
notation. My personal favourite, however, is a symbol that single-handedly
signifies the following within SMuFL so far:

-string harmonic
-wind open
-fingering hole
-percussion open
-guitar open
-fretboard open
-chord symbol diminished
-combining void dot
-function theory ring

(Those are just the ones that are roughly the same size and could be easily
mistaken amongst themselves.)

Semantically, it would be wrong to substitute all those glyphs with a single
comma glyph, a single plus glyph and a single circle glyph, to be rescaled
as context demands, even if separate glyphs for each of these many symbols
might end up looking virtually identical within a given font. But it would
not be wrong to use a scaled-down standard size clef for indicating a clef
change (in fact, it would be exactly correct). To do so could be, however,
potentially problematic for matters of typography and readability.
Therefore, the core question is not whether clef changes should get their
own code points or not (they should not), but how the inevitable scaling of
clefs should be addressed.

With this in mind, I'd like to suggest not to deal with scaled versions of
arbitrarily singled-out symbols, but to instead reserve one stylistic set
(probably the last of the twenty) exclusively for alternate versions
intended for small (effective) printing sizes. Font designers would be free
to provide as many symbols for this set as they see fit (again: scaling down
symbols is not intrinsically wrong, it just comes with certain problems),
and there should be clear recommendations in the guidelines for which
symbols to include. Clefs should be on the top of that list. To give font
designers more control over these issues, intended scaling factors for cue
size, grace note size and size for smaller clefs should be providable as
meta-data.

Let's say that a font designer provides a stylistic set for smaller clefs
(with the issue of clef changes in mind) and draws them on the premise of
scaling the full size clefs down to 63%. Now let's assume a scoring
application has its own default scaling factor for clef changes, which would
be, for the sake of the argument, at 66%. Setting aside any global scaling
of the stave size, wouldn't that mean that due to the override of the scale
factor by the application the clef would not be scaled down enough to
trigger the use of the stylistic set the designer intended for smaller
clefs? If, however, there was meta-data providing an intended scale factor,
the (optional) rule for the application could be something like: scale down
the full size clef to 63% (as stated in the font's meta-data), and if the
resulting size is smaller than the effective size mentioned by Daniel (how
would that size be determined, anyway?), use the stylistic set. The way I
understand it, without providing the actually intended scale factor (or size
limit) in such cases there will always be a small but inacceptable risk of a
stylistic set being used (or not used) erroneously.

[All this having been said, I want to add that, at this point, I would not
oppose the addition of the three most commonly used clefs in smaller size as
code points, as announced by Daniel. However, while it is the most pragmatic
thing to do, this solution does introduce a high-level inconsistency and
does dodge the actual problem, since the next-most commonly clefs are not
exactly obscure (I'm looking at you, Unpitched Percussion one and two) and
would thus get second-class treatment.]

This concludes this much too late sermon.

Alex




* by the way: the inclusion of two differently sized guitar tabulature clefs
strikes me as a similar case; are they really both necessary? They even come
with additional stylistic alternates for, strangely enough, taller sizes,
and I would argue that, just as with clefs, the size difference carries no
information of its own. The information about the number of strings comes
from the stave, not from the clef. The size adjustment seems to be more a
thing of typographic convenience. Because of that I would suspect that they
might be reduced to one code point, but I'm no expert in guitar notation.

** There are superscript digits, but I would suppose that those are actually
symbols for exponentiation.


#############################################################
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]>