Jan Nieuwenhuizen wrote: > Well i'm surely not going to argue that Lily has no bugs, or even that > Lily is lacking trivial features that most other notation programs have > (could you put an url to your 'examples', that talks a bit easier). Well, I looked at this a while back, printed out some examples, and scurried over to the music dept. Since then, I've not been able to ghostview any of the output lily generates (ghostview says "Error: /undefinedfilename in (lily.ps)" because I've not installed it in elec-eng) but I did send some questions off to the group about the points I mentioned. BTW, I did mean "Star-Spangled Banner" in the original post --- I must be starting to suffer short-term memory loss due to lack of coffee 8-) The original quesions were all answered and we made a policy decision to go with lilypond. I've attached them at the bottom to show you what I mean, but I've already had infomative answers back from Han-Wen Nienhuys and others. Let me say again: we're comitted to using Lilypond, because we think it will eventually be the best package, because it is open. I want our code to be open too. We can't use MUP for that reason, although I do agree with Bob that it /seems/ more mature, I'd have to agree with jeff that prop. formats are dangerous. But then, you see, I'm not wanting to get nice printed output yesterday, so I'm not really a "real user". > But hearing that someone finds another programs output //that much// better > than Lily's made me curious! > So today i took a look again at Mup http://www.Arkkra.com/, and converted > one of MUPs samples (sic.) to mudela. > > > Looking forward to being proved wrong 8-) > > Judgement of what is 'fine output' will always be rather subjective; > there's no right or wrong here, so you may best see for yourself: > > PostScript output of MUP and Lily: > http://www.Arkkra.com/doc/star.ps > > http://www.xs4all.nl/~jantien/lilypond/input/test/out-www/stars-and-stripes.ps.gz > I'd really like to have a look at this, but I can't at the moment. BTW, it's just plain PS, not gzipped, yes? Any chance of a printable version? Otherwise I can run across to music and get somebody to run lilypond against it. I hope things have improved in newer versions we've not put up here yet... also, the .png files have mysteriously disappeared... At least most of the points I raised in my attached email wouldn't be an issue with a WYSIWYG front end: you would see the problems on screen and avoid them. Our project is also tied in with some musicological work, probably using humdrum, so as you can imagine it's a big piece of code. We're still arguing about architecture so far, so please don't expect working stuff soon 8-) All the best, Nick/ -- Dr. N.J.Bailey----------------------------------------------- Lecturer in Electronic and Electrical Engineering University of Leeds, Woodhouse Lane, Leeds, LS2 9JT. UK.------------------------------------------------- http://www.elec-eng.leeds.ac.uk/staff/njb/
[EMAIL PROTECTED] writes: > Hello, Han-Wen! > > Just in the process of refining our grant application for this here > musicological project which should include a graphics front-end which can > generate Lilypond source. We've been looking around and would like to pester > you with some questions about some buglets that might be demonstrated by the > examples. These are sub-optimalities in the printed output, and we are > wondering if they can be overcome by simply changing the Lilypond input. > We're not asking you to do that! Just to tell us if the things we've got > here are a result of the source you typed, or are in some way "hard-coded" > into the project. If it is the former, our score editor will of course > produce totally perfect output for lilypond to work with and have no problems > (yes, really 8-), but if it is the latter, we will possibly need to > contribute some fixes (if you will let us break your code 8-). I'm trying to > get a handle on this to estimate total project man-hours. > > > http://www.cs.uu.nl/people/hanwen/lilypond/mutopia/out-www/standchen.ps.gz > > Bar 1: the pianissimo marking collides with the quaver noteheads > (?bad kerning) Lilypond `bug'. Solving it is difficult without major overhaul, because the pp and quaver aren't logically connected. (I put `bug' in quotes, because to me a bug is something that does not conform to specifications; solving the pp problem is not in the specs..) > Bar 71 (last page): Bad collision between phrase mark, forte mark and > triplet "3" idem. In general, avoiding graphical collisions is difficult to do automatically. > Throughout: ties and slurs uniform over whole length (not thicker in > the middle: ?font) This is tunable, I think. > Throughout: Words aren't centre-justified under noteheads; "Lei- > der", not "Lei - > der", and no use of "_" ("mich!" on last page, not "mich_______!") Lilypond. Needs fixing, but shouldn't be very difficult. > Page 2: penultimate system, middle bar: tie cuts thorough > semiquavers. > Last page. Ties unusual (in LH both go down. Shouldn't they go in > opposite directions?) > Lilypond `bug', but might be tunable somehow. Fixing this shouldn't be too difficult. -- Han-Wen Nienhuys, [EMAIL PROTECTED] ** GNU LilyPond - The Music Typesetter http://www.cs.uu.nl/people/hanwen/lilypond/index.html
