Guido commented: | On Wed, 28 May 2003, John Chambers wrote: | > | Do you see anything else? | > My jcabc2ps clone has a flock of "little" extensions. The biggie is | | John, what you wrote is really mouth-watering... I hope Jean-Fran�ois will | manage to include most (if not all) of your extensions.
Yeah, to some people. Of course, to others, the only thing interesting is Western "classical" notation. That's easy to understand, and for a lot of people that's all that's ever needed. I'm generally mixed up with a crowd that plays a lot of Eastern European, Balkan and Middle-Eastern music. The needs of that music are somewhat different. This is the source of much of the abc2ps cloning. Different musical crowds have very different needs, and usually see little reason to bother with the "weird" stuff that their music doesn't need. One of the funnier is the question of musical fragments. Most abc software only accepts complete "tunes". It's not obvious to many musicians why you'd want anything else. But if you're trying to write music documents, you quickly become frustrated by this. Yeah, you can embed what you want in a complete tune, and then use image-munging software to cut out all the extraneous stuff. But on a piece of paper, I can write just a couple of notes on a bare staff, with no need for completeness. If you're used to using abc, it's obvious that abc could easily handle fragments. Then you wouldn't have to futz with all that insane image stuff. So why not do it? | I've been waiting for years to see the merge of existing abc2ps clones. It | looks like the time has come at last. It won't necessarily be easy. And whether the clones that do tablatures will be included is another question. One of my motivations, of course, is the needs of my Tune Finder. It sees all the abc that people have put on the web, and you might be very impressed by the variety. Some of my extensions have been to handle some of this. There is still abc Out There that can't be rendered correctly by my stuff. It'd be very useful to have a tool that can handle it all. One of the constraints on the code running behind a web site is that it isn't being run by a human sitting at a computer. It's programs running on machine X trying to handle a request from machine Y, but it can't directly access Y or interact with the user on Y. It has to be programs that can be run from a "script". This is why I rewrote the command-line parsing so that jcabc2ps can be run as a normal unix "filter" program under the control of another program. This isn't a totally trivial change, especially since it meant reworking all my Makefiles. (But with vi this is usually just one or two global 's' commands, and the result is a smaller Makefile.) The question of other character sets is interesting. The usual first guess is some header code giving the character set. But this is not sufficient. I have some songs in my collection with lyrics in Russian and Finnish, for example. Or if you want a really fun case, try Russian and Yiddish. Currently jcabc2ps handles 8-bit chars just fine, so files in the 8859-1 char set work without any problem. This handles Finnish, but I can't get Russian and/or Yiddish w: lines on the same page. I've read a lot of stuff about Unicode, but so far my main reacion is "That sounds nice; how the hell do you actually use it?" What do I put in my plain-text file so that when I dump it to my printer, it comes out correct on the paper? And it'd be nice if I could see it on the screen first. So far, the info I read is mostly just confusing. To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
