Good monday morning everyone.
I'm a huge fan of MuseScore and would like to have this as part of the primary software for Fedora Jam. What are peoples thoughts on MuseScore? Is it only me that would prefer to have it to having lilypond/frescobaldi. Let the discussion begin :)
-- Jørn Lomax GSoC atudent, Fedora audio Spin CS Student University of Tromsø
On 25 June 2012 07:19:19 Jørn Lomax wrote:
Good monday morning everyone.
I'm a huge fan of MuseScore and would like to have this as part of the primary software for Fedora Jam. What are peoples thoughts on MuseScore? Is it only me that would prefer to have it to having lilypond/frescobaldi. Let the discussion begin :)
-- Jørn Lomax GSoC atudent, Fedora audio Spin CS Student University of Tromsø
Hi:
I'm sure we'll talk about this at today's IRC meeting, but there are a couple of things I'd like to point out on the list.
1.) MuseScore is not a substitute for LilyPond. Both applications have the same end goal, but they represent very different ways of getting there. Why not include both?
2.) This becomes a question of live media size. I didn't do that much checking, but MuseScore appears to be about 8.9 MB, while all the LilyPond requirements and Frescobaldi are about 5.5 MB. This difference is probably meaningless for us--both packages are small.
3.) The problem is the Frescobaldi package uses rumour, which uses fluidsynth, which has a dependency on a SoundFont package, which is very large (120 MB or 137 MB depending which one). If we're going to ship FluidSynth anyway, then LilyPond+Frescobaldi isn't that much extra burden.
4.) We could always ask the frescobaldi package maintainer (Brendan Jones) to remove the dependency on FluidSynth.
5.) When Frescobaldi moved to version 2.0 and stopped using KPart components for the interface, the application became cross-platform (although KDE is cross-platform, the Python bindings still weren't). This is good overall, because now people using Windows or Mac OS X can use Frescobaldi, but it's bad for those of us who are accustomed to the Kate KPart, and rely on its idiosyncrasies (sometimes called "features").
Originally for my own personal use, I branched the KDE-dependent Frescobaldi, renamed it Krescobaldi (obviously), and uploaded it to GitHub.[0] Because I've never used the MIDI capabilities, I don't know anything about it, and I'm the only maintainer, I'm going to remove that functionality. Unconcerned about Windows compatibility, we at Fedora could easily package and ship LilyPond/Krescobaldi on the live media, without a dependency on FluidSynth.
6.) Krescobaldi is a bad idea unless we already plan to ship KDE.
7.) In summary, let's have both.
Christopher.
On Mon, 25 Jun 2012, Christopher Antila wrote:
I'm a huge fan of MuseScore and would like to have this as part of the primary software for Fedora Jam. What are peoples thoughts on MuseScore? Is it only me that would prefer to have it to having lilypond/frescobaldi. Let the discussion begin :)
1.) MuseScore is not a substitute for LilyPond. Both applications have the same end goal, but they represent very different ways of getting there. Why not include both?
+1 And MuseScore can even export to LilyPond. Rosegarden can also export to LilyPond.
3.) The problem is the Frescobaldi package uses rumour, which uses fluidsynth, which has a dependency on a SoundFont package, which is very large (120 MB or 137 MB depending which one). If we're going to ship FluidSynth anyway, then LilyPond+Frescobaldi isn't that much extra burden.
AFAIK Frescobaldi 2.x does not use rumour anymore. It uses portmidi for MIDI playback. It can use Fluidsynth for MIDI playback, but it can just as well use timidity, or an external hardware synth
4.) We could always ask the frescobaldi package maintainer (Brendan Jones) to remove the dependency on FluidSynth.
See 3)
6.) Krescobaldi is a bad idea unless we already plan to ship KDE.
IMO Krescobaldi is a bad idea, period. Wilbert Berendsen has put a lot of work in Frescobaldi's transition from KDE to Qt, and is still actively working on the project. Krescobaldi would be stepping backwards.
7.) In summary, let's have both.
+1
-
Christopher Antila -
Jørn Loamx -
Martin Tarenskeen