Simon Wascher <[EMAIL PROTECTED]> writes:
> For reasons of scientific quality I need to include as much as possible
> information of the original source within the file, as near to the
> original as possible. So the file should be a representation of the
> playback *and* the the display.
The ABC file should be an approximate representation of the *musical
content* of a piece. Anything else is likely hoping for too much -- as
far as playback is concerned, an ABC file of Bach's Goldberg Variations
would probably be closer to the sort of modern edition of the work that
you can buy in a music shop than to either Bach's original manuscript
(on the `display' side) or Glenn Gould's recorded interpretation of the
piece (on the `playback' side). Both exploit dimensions of
`presentation' that are simply not available in ABC notation, nor likely
to be, and while the ABC edition of the piece would be nice to have
around the house and let us have all kinds of fun with it, it would most
probably never be the one or the other.
As I said earlier on, if you want a near-facsimile replica of an
original source, ABC is probably not the right vehicle for your work --
at least not without major tweaking of the standard and without
restricting yourself to a particular output program (since the standard,
fortunately, doesn't prescribe printed ABC output in enough detail for
the various notation programs not to differ greatly from one another in
just how they display things). Note that ABC in its current form doesn't
even let you specify whether the stems of notes go up or down in a
sequence like `ABc', so if you want to imitate a given score as closely
as possible (we're talking `science', after all) using ABC, you have
still quite a way to go before you're done.
I don't know a lot about Lilypond but it might be worth checking out for
your purposes. In any case if you're after a specific look to your sheet
music you would do well to publish PDF as well as ABC, since even if you
stick all your little presentation things into the ABC standard others
will not be able to reproduce your output with anything approaching
`scientific' precision unless they run exactly the same software that
you do. Similarly, if you want to capture or express all the details of
the performance of a piece of music then something like a MIDI file may
really be the way to go, rather than ABC. All three formats -- ABC, PDF,
MIDI -- fill different `ecological niches' that do overlap to some
extent, but any one of the three could never fully replace either of the
others.
I'm not saying that you shouldn't try to make ABC do everything that you
want. I'm just trying to caution you not to expect too much from ABC as
it is, or ABC as it can be while still resembling what ABC is today
enough to be recognizable. If music notation schemes were vehicles, then
ABC originally started out as something like a bicycle -- very cheap and
easy to use but still able to get us to all sorts of interesting places
as long as they are not too far away from home. Now since its original
invention the ABC bike seems to have acquired all sorts of useful
accessories (like a lamp, 21 gears, panniers, and a radio), but that
doesn't mean that we will ever be able to use it to pull a 25-ton
trailer -- not without losing the ability of carrying it down the
basement stair for the night, anyway, and that may be something that
many of its users aren't ready to give up.
> Do not say this is *not* the target of abc, there is no such thing as a
> restriction against display matters anywhere in the standard. Its *your*
> ideology about abc. For me abc is a way to create compressed, easily
> exchangeable, freeware, playback active representations of a musical
> source.
Guess what -- with me it's just the same. However I've been around the
block often enough to have found out that it is a good idea to keep the
actual information and the details of its presentation quite separate.
It is not as if this was just some stupid `ideology' -- it is sound
engineering practice, founded on a large body of experience in designing
similar systems, which is available to anyone who deigns to enter a good
computer science research library. (In fact, an hour or two on the Web
reading up on the philosophy behind XML should give one a reasonably
good idea of what the data/presentation dichotomy is all about, without
one actually having to bother moving one's lazy self away from the
computer.)
You appear to have this learning experience still ahead of you. I wish
you well; may your epiphany not be too painful when it comes. As it
will, eventually.
Anselm
--
Anselm Lingnau .......................................... [EMAIL PROTECTED]
What we call 'Progress' is the exchange of one nuisance for another nuisance.
-- Henry Havelock Ellis
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
