This is indeed one area where software developers have to struggle - Historically, it seems to me that the value of the beat got shorter and shorter, and there are some reasons why grand staff transcriptions may chose to halve the original values: it is easier for instance to show the different voices in 8th notes than quarter notes, because 8th notes can be tied. Since both systems of representation have pros and cons, it makes sense for the software to try and support both the original representation and the tweaked modern approach. this is not easy because two quarter notes and two tied 8th notes are very different things. Also, printers of the period faced some technical and graphical problems just the same as we do today - or worse. So manuscripts may be different from printed works. There are also sometimes geographically based preferences for some system over another. Finally, software users who do send feedback tend to have a significant impact in terms of preferences, particularly if their request makes sense and can be converted into code in a fairly straightforward way. An example of a problem I faced is this: if as Stewart says a minim is represented by a stem with a single tail, then a stem with 2 tails is a quarter note. These in modern notation would not be tie-able, but I am pretty sure that in some manuscripts they are... Then, if you decide they are tied, how many horizontal dashes should they have: 0 because they are quarter notes, 1 because they are the first level of tied notes, or two to match the tails on the stand-alone flag? We could have historical models - like the Lachrimae - to help resolve some of those issues, but if I recall, the viols parts are unbarred. And even if one publication of the time gave some firm evidence, that evidence could be disproved in some other publication of the time... All the while I am writing this I also think about ornament systems that are even worse from those points of view... Django does offer the option to equate one flag tail with a minim (half-note), but it is not the "default" mode - the main reason being I think that our modern mind has got used to the shorter beat value and a maximum of tied notes. There is no manual of "software development for historical music notation" - and this is the exciting part for all of us: users can still directly communicate and influence the direction of the development of those new tools and take a very active part indeed. From the point of view of software, things have evolved at a mind boggling pace in the past ten years: we are living a truly unique period in time, with extraordinary potential. A lot of flexibility has been added to my own software in the past few years, and the main challenge that I see now is that of articulating the complexity that underlies that new flexibility with a fairly simple interface for the user, so they can chose easily what fits their needs best. I'll be happy to make the one tail equal minim the default value if there is a consensus on that point - in the meantime it is at least available as an option. Sorry if I made it sound unnecessarily complicated - Alain
At 01:50 AM 6/9/2004, Stephan Olbertz wrote: >Dear Stewart and all, > >in Django, Stringwalker's successor, you have an option >for the right relation in each file. > >Regards, > >Stephan > >Am 8 Jun 2004 um 13:03 hat Stewart McCoy geschrieben: > > > Dear All, > > > > While examining the Scolar Press facsimile of Campion's "My sweetest > > Lesbia" to be able to reply to Peter Nightingales' query, my mind > > turned to note values. It is perfectly clear from this song, or indeed > > any other song from this period, that the tablature rhythm sign > > > > |\ > > | > > | > > > > means one minim (or half-note). I find it immensely frustrating that > > so many people either misunderstand or choose to ignore that > > relationship. > > > > It was fashionable some years ago to halve note values when > > transcribing lute tablature into staff notation. Diana Poulton and > > Basil Lam do so in _The Collected Lute Music of John Dowland_ (London: > > Faber Music Limited, 1974). In their edition that minim sign is > > transcribed as a crotchet. I have the second edition (1978), which > > includes a few extra pieces which came to light after the first > > edition had been published. In this newer edition the editors no > > longer adhere to their policy of halving note values. Nos 93-6 have > > the correct transcription, nos 97-8 have halved note values, nos > > 99-100 are correct, and nos 101-3 have halved note values. It is a > > vertitable dog's dinner, at least as far as the rhythmic values are > > concerned. > > > > Pascale Boquet, in her _Approche du Luth Renaissance_ (n.p.: n.p., > > 1987) goes one stage worse. She confusingly regards that same minim > > sign as a quaver (quarter note). > > > > Alain Veylit with Stringwalker and Francesco Tribioli with Fronimo > > both get the relationship wrong in their computer software. I think > > both their programmes are excellent in their different ways, and have > > proved immensely useful, yet both make the mistake of automatically > > halving note values. Stringwalker can create instant transcriptions of > > tablature, but with the option of halved or quartered note values, not > > the correct value. Fronimo can reproduce lute songs, but the singer's > > notes have half the value of the notes for the lute. One is left with > > the dilemma: do I give the wrong note values to the singer or to the > > lute? It is confusing performing lute songs prepared with Fronimo, > > since the lutenist reads his tablature with one set of note values, > > while glancing up to the singer's part, which has a totally different > > set of note values. > > > > -o-O-o- > > > > To show how fashions have changed over the years, here are a few > > books where the value of tablature rhythm signs has been halved. > > Note the date of publication: > > > > Thomas Morley, _The First Book of Consort Lessons_, ed. Sydney Beck > > (New York: C. F. Peters Corporation for The New York Public Library, > > 1959) > > > > _Jacobean Consort Music_, ed. Thurston Dart and William Coates, > > Musica Britannica 9 (London: Stainer and Bell Ltd for the Royal > > Musical Association, 2nd edn 1962) > > > > Anthony Holborne, _The Complete Works of Anthony Holborne_, ed. > > Masakata Kanazawa, Harvard Publications in Music 1 (Cambridge, > > Mass.: Harvard University Press, 1967) > > > > Contrast this with more recent editions, where the tablature rhythm > > signs have been transcribed with their correct value. Again note the > > date of publication: > > > > Alfonso Ferrabosco the Elder, _Opera Omnia_, Corpus Mensurabilis > > Musicae 96, vol. 9 (Neuhausen-Stuttgart: Hänssler-Verlag for the > > American Institute of Musicology, 1988) > > > > _Collected English Lutenist Partsongs: 1_, ed. David Greer, Musica > > Britannica 53 (London: Stainer and Bell for the Musica Britannica > > Trust established by the Royal Musical Association, 1987) > > > > Francis Cutting, _Collected Lute Music_, ed. Jan W. J. Burgers > > (Lübeck: Tree Edition, 2002) > > > > -o-O-o- > > > > Older editions tend to have triple time treated differently from > > duple time. This results in a somewhat anomalous transcription in > > _Chansons au Luth_ ed. Lionel de la Laurencie (Paris: Heugel, 1976), > > p. 164. At the top of the page the music is in triple time, and the > > tablature rhythm signs are transcribed with halved note values. Half > > way down the page there is a change of meter to C with a slash, and > > thereafter the rhythm signs are given their correct value. It's all > > rather confusing really. > > > > I leave the final word to Thomas Robinson. On page B2v of _The > > Schoole of Musicke (London, 1603) he writes: > > > > |\ > > | > > | A Minim. > > > > > > Best wishes, > > > > Stewart McCoy. > > > > > > > >
