Richard Robinson writes:
| On Fri, 29 Dec 2000, Gianni Cunich wrote:
| > Be serious, John, the REAL abc users aren't likely to learn perl, python or
| > whatever you might use, or to hack the last abc2ps clone
First, let me make one thing clear - I have no intention of being
serious at all times. After all, we're talking about music, and
common speech says that you "play" music. There is overly much
seriousness here. I think we should fight against this at every
opportunity. It's PLAY, dammit. *PLAY*; y'hear me?
| But, the people that do things like this are as real as the people that
| don't. Everybody exists. The people that hack things mostly do so because
| they want that behaviour to use in their abc, no ?
Yup. Of course, as a long-time techie nerd type, I've gotten quite
used to the idea that a large percent of the population considers me
less real than most. I grew up in a family for which seeing something
published in a book was itself grounds for ignoring. There's a lot of
this sort of thing going around. There are plenty of musicians around
who are openly contemptuous of people who read music. Computerized
music notation is utterly beneath their contempt, as are those of us
who take such stuff seriously. This ain't gonna change.
But we should consider that abc is not just a music notation, it is a
*computer* music notation. There are now several million programmers
in the world, and a lot of them are musicians. To the programmers
amongst us, ABC is "just data", as amenable to programming as any
other sort of data. I fully understand the attitude of musicians who
aren't about to learn the first thing about programming, but I'm not
one of them.
Some here might have noticed a pattern to my suggestions in the past:
ABC should be kept simple to type, read and parse. These are all
facets of the same sort of simplicity, of course. But in addition to
usability by humans, one of the nice things about ABC is the ease of
writing code to do things with it. This is because it's easy to
parse. In fact, to your typical perl hacker, the suggestions of a
single "standard" ABC parser is a bit of a puzzle. Why bother? All it
takes is a simple pattern. At most a few lines of code. Yeah, it's
hard in those low-level languages like C, C++ or java. But if you use
a language with even the most primitive string parsing operations, an
ABC parser is merely a bit of throwaway code, easier to type than to
find in your archives. Unless we materially increase the complexity
of ABC's syntax, this is going to be true for some time.
This simplicity is, I'd guess, much of the reason behind the recent
suggestion that a lot of ABC users quickly unsubscribe from this
list. They subscribe to ask a few questions. When those questions are
answered, they know everything they'll ever need to know to use ABC
for their purposes. The developers' discussions are not all that
interesting, so they unsubscribe. Not out of animosity, but rather
because all they need is to type in some tunes, email them to friends
or other lists, and so on. This is something the typical user can
learn in a very short time. Unless they find that they need something
a bit more sophisticated than notes, bar lines and chords, we never
hear from them again.
This is one of ABC's real strengths, of course. This simplicity is
most of the reason why the latest run of my Tune Finder's search bot
found somewhat over 81,000 ABC titles on the Web. No other music
notation is even within two orders of magnitude of this.
(Which doesn't stop me from bitching about things ABC lacks ... ;-)
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
