Well, it's quite self-evident to me: Lilypond supports guitar, banjo, and maybe mandolin. Period.
If you read the docs, you can see that you can define the number of strings and set their open-string pitches. You have one font, numbers, one type of flag, attached to the numbers like they would be to noteheads, and no direct entry. (By which I mean, you enter pitch/duration values just like for notehead notation, not string/fret.) Back in the day (can it really be that long ago?) I had quite a few discussions with Lilypond folk, both developers and some of the users, about revamping the tab system to make it universally useful. Since I don't program C++ and don't have time to give away in that volume, I was unable to respond to the usual "It's all documented and the source is right there, why don't you fix it?" Further, about the time I was starting to get somewhere with suggested improvements (which eventuated in their Petrucci-like notation font, at least, although I hardly am claiming responsibility for it or for causing someone to actually make it) the developers went from "Labor of love" to "Buy our love": if you wanted an improvement or addition and could pay them to work on it, they'd give it a try. I'm also not rich. So Lilypond remains a lute-insensitive system. I spent some time, back then, trying to develop a coherent regime for describing those features of tablature which would enable a Knuth-like approach. My hope was that, with mf fonts for appropriate tab letters, numbers, and time-indicators (flags), and a well-enough designed set of keywords, one could enter tablature (string, fret, duration) or notation, convert between them (mostly for piano setting of parallel tab lines) and change their format, say from french to italian notation or even from dm tuning to vielleton. If I could make it work for lutes and theorboes and guitars and banjos, then it would work pretty well for anything with multiple strings and frets, I figured. Of course, the number of choices is mind-boggling (and some of them are very specific to a couple of instruments, but not others, like theorbo 6x8, 7x7, 8x6 choices, etc), and the basic idea of Lilypond is anti-tab in the first place. That last statement deserves some elucidation. Lilypond approaches notation from the viewpoint that each note can be quantified by its pitch and duration (not a bad start). When the time comes to apply those notes to a specific instrument, the only things that are of real importance are clef, key signature, key (i.e., it allows for transposition) and arrangement of lines (4-line staves, piano grandstaves, etc.) This is a reasonable position for a couple of modern-orchestral musicians whose goal is to make computers output Barenreiter-quality printed music. It is also, basically, Knuthlike. (Knuth invented TeX and metafont, working from the concept that all letters and all printing share a certain set of reasonable parameters, and by finding those parameters and making them variable, you should be able to change between any printing style and any other printing style, all while preserving the constants, like good white-space use, no orphan or widow lines, etc.) When it came to tab the attitude was, we're composers, not lutenists: we want to write music, and let the machine choose where each note appears on the tab. So there is little option to force a specific note to a specific string, and no real mechanism to prevent impossible finger stretches. The real charm, for me, at least, of multi-stringed instruments is that you have choices: you can play an open string or finger a fret and get the same note (choosing the top e string on guitar, for instance, you can get the same note at the fifth fret of the b string, the 9th fret of the G string, the 14th fret of the d string, etc.) Likewise, you can finger a note high on a low string, which is 'above' an open string, making close-chords or dissonances which would be otherwise impossible. Machines with simple programs don't deal well with this kind of thing. Lilypond doesn't either. You can enter a chord, have it appear in guitar notation and add graphic symbols to indicate fingering or which string to play, but there's no connection between that and how the Tab would appear, and may never be. So their concept was to be able to write music, not "music-for", and then make it music-for by dropping it into an "I make music for-" box (say, "I make music for lute" or "I make music for piano") and have it come out notated correctly. The first part makes a nice concept, the last part doesn't happen without hand-modifying and hand-fudging the code. Every fudge means you can't transpose or otherwise use the machine's ability to change things for you, and puts you back to using a wide-nib pen blind by having a friend hold it, and you describe what to write. Django can set tab very well, but even here, you can end up with impossible tabs, which are very difficult to correct. (Like when a chord ends up notated in stratospheric fret letters on lower strings, but you can't pull the note to the right string because there's another stratospheric letter there already!) Even still, it's worth paying for if you're doing tab, and its worth the learning experience. Finale, by the way, suffers like this for tab, and worse, because of typographic errors that should never happen. Looking through Amherst Early Music's wonderfully-useful Odhecaton is instructive on how many ways Finale can stab you in the back, and it doesn't even have tab in it! Sibelius, I have less familiarity with, but it seems that the more powerful and capable (and expensive) the computer-based editor, the harder it becomes to do what you actually wanted to do. Amherst, for instance, employed a Finale expert, and still ended up with crashes between rests and notes and barlines. One final note, on this entirely-too-long nearly-off-topic rant: If you have something like Django which you use once-a-year, get a notebook. Make (http://blog.makezine.com/) have a lab notebook in their store which is perfect for this purpose, but any good lab-notebook-like-object will do. Then, take notes on what you do and learn. Draw a star next to everything you figure you'll need to know how to do again, and, starting from the back of the notebook, make an index with a descriptive line and the page number of the star. (If there are more than one star on a page, consider pageno.starno notation!) Then, when you pick the notation software up again, start with the notebook, and remind yourself of what you learned before. This is standard lab procedure, I've carried it into hand-manufactury (I make one-off lasers) and I use it for software like Django (and autoCAD) which I use rarely and which have a learning curve. ray On Thu, Dec 4, 2008 at 7:31 AM, Spring, aus dem, Rainer <[EMAIL PROTECTED]> wrote: > "quite self-evident then" > > ??????? > > What do you mean? > > > Best wishes, > > Rainer aus dem Spring To get on or off this list see list information at http://www.cs.dartmouth.edu/~wbc/lute-admin/index.html
