
               GNU Music project - manifesto

           Han-Wen Nienhuys and Jan Nieuwenhuizen

Contents

     1: Goal
2: Requirements
3: Components
4: Programs

1: Goal

The public deserves free tools for composing and printing.

2: Requirements

Emacs and  serve as useful examples of what programs by the
GMP should be.

high-quality
     The software should be of high-quality from the appli-
     cation view.  For example, the notation should be high-
     quality from an engraving point of view, like

high-quality
     The software should be of high-quality point of view,
     like all GNU software, it should have no limits, be
     fast, etc.

tweakable
     Printed music has a lot of styles, and special symbols.
     It may be unfeasible to provide and maintain  lots of
     code that is hardwired into the system. The tools
     should be extensible/programmable like Emacs and TeX.

easy to use.
     That is, for technical users (that can read a manual).
     The learning curve should be as flat as possible but
     not at the expense of comfort of use and power.

3: Components

A set of music fonts
     In development, the Feta font.

A typesetting engine
     In development, included in LilyPond.  A system with
     rules on how to set properties of items to be printed
     (up/down directions, breaking, dimensions, etc) It
     should be suited to interactive typesetting (so, incre-
     mental algorithms are needed)

A display engine
     which can display clear notewriting in (say) an X-

     window Ideally the system should cooperate with the
     typesetting engine

An ASCII language
     In development, LilyPond has a language.  Having an
     ASCII format which enables urtext, and easy sharing
     (via mail and news forums) encourages cooperation and
     exchange of music.

A printing engine
     In development, included in LilyPond.

An input system
     The natural way to enter composed music is singing or
     playing it. The GMP should have module which can take
     keyboard input or microphone input and convert it to
     computer data. (microphone input would be difficult)

sequencing
     (have no clue about this)

A scanning system
     Having a system which can produce mudela from printed
     scores,  greatly simplifies creating a collection of
     music

A music-understanding system
     (difficult) A system to generate accompaniments, fig-
     ured bass, automatic accompaniment, etc.

A performing system
     A system which can play credible performances of
     abstract music representations.  LilyPond has a bare
     bones system, but it cannot (yet) be described as
     "credible".

4: Programs

o    A noninteractive typesetter, suited for batch jobs, and
     typesetting existing music. This would couple the ASCII
     language, the printing engine and the typesetting
     engine LilyPond is currently representing this section.

o    A GUI for composing. This would combine the display
     engine, the input system and the typesetting engine.

o    Libraries for reading and writing various
     audio/music/notation formats.
