More misc CG stuff.

[lilypond.git] / Documentation / devel / doc-work.itexi

@c -*- coding: us-ascii; mode: texinfo; -*-
@node Documentation work
@chapter Documentation work

@menu
* Introduction to documentation work::  
* Texinfo crash course::        
* Documentation policy::        
* Tips for writing docs::       
* Updating docs with convert-ly::  
* Translating the documentation::  
@end menu


@node Introduction to documentation work
@section Introduction to documentation work

Our documentation tries to adhere to our @ref{Documentation
policy}.  This policy contains a few items which may seem odd.
One policy in particular is often questioned by potential
contributors: we do not repeat material in the Notation Reference,
and instead provide links to the @qq{definitive} presentation of
that information.  Some people point out, with good reason, that
this makes the documentation harder to read.  If we repeated
certain information in relevant places, readers would be less
likely to miss that information.

That reasoning is sound, but we have two counter-arguments.
First, the Notation Reference -- one of @emph{five} manuals for
users to read -- is already over 500 pages long.  If we repeated
material, we could easily exceed 1000 pages!  Second, and much
more importantly, LilyPond is an evolving project.  New features
are added, bugs are fixed, and bugs are discovered and documented.
If features are discussed in multiple places, the documentation
team must find every instance.  Since the manual is so large, it
is impossible for one person to have the location of every piece
of information memorized, so any attempt to update the
documentation will invariably omit a few places.  This second
concern is not at all theoretical; the documentation used to be
plagued with inconsistent information.

If the documentation were targeted for a specific version -- say,
LilyPond 2.10.5 -- and we had unlimited resources to spend on
documentation, then we could avoid this second problem.  But since
LilyPond evolves (and that is a very good thing!), and since we
have quite limited resources, this policy remains in place.

A few other policies (such as not permitting the use of tweaks in
the main portion of NR 1+2) may also seem counter-intuitive, but
they also stem from attempting to find the most effective use of
limited documentation help.



@node Texinfo crash course
@section Texinfo crash course

@menu
* Texinfo introduction::        
* Sectioning commands::         
* LilyPond formatting::         
* Text formatting::             
* Syntax survey::               
* Other text concerns::         
@end menu


@node Texinfo introduction
@subsection Texinfo introduction

The language is called texinfo; you can see its manual here:
@uref{http://www.gnu.org/software/texinfo/manual/texinfo/}

However, you don't need to read those docs.  The most important
thing to notice is that text is text.  If you see a mistake in the
text, you can fix it.  If you want to change the order of
something, you can cut-and-paste that stuff into a new location.

@warning{Rule of thumb: follow the examples in the existing docs.
You can learn most of what you need to know from this; if you want
to do anything fancy, discuss it on @code{lilypond-devel} first.}


@node Sectioning commands
@subsection Sectioning commands

Most of the manual operates at the

@example
@@node Foo
@@subsubsection Foo
@end example

@noindent
level.  Sections are created with

@example
@@node Foo
@@subsection Foo
@end example

@itemize
@item
Please leave two blank lines above a @@node; this makes it
easier to find sections in texinfo.

@item
Sectioning commands (@@node and @@section) must not appear
inside an @@ignore.  Separate those commands with a space, ie @@n
ode.

@end itemize


@node LilyPond formatting
@subsection LilyPond formatting

@itemize

@item
Use two spaces for indentation in lilypond examples.  (no
tabs)

@item
All text strings should be prefaced with #.  LilyPond does
not strictly require this, but it is helpful to get users
accustomed to this scheme construct.  ie @code{\set
Staff.instrumentName = #"cello"}

@item
All engravers should have double-quotes around them:

@example
\consists "Spans_arpeggio_engraver"
@end example

Again, LilyPond does not strictly require this, but it is a useful
standard to follow.

@item
Examples should end with a complete bar if possible.

@item
If possible, only write one bar per line.  The notes on each
line should be an independent line -- tweaks should occur on their
own line if possible.  Bad:

@example
\override textscript #'padding = #3 c1^"hi"
@end example

Good:

@example
\override textscript #'padding = #3
c1^"hi"
@end example

@item
Most LilyPond input should be produced with:

@example
@@lilypond[verbatim,quote,relative=2]
@end example

@noindent
or

@example
@@lilypond[verbatim,quote,relative=1]
@end example

If you want to use \layout@{@} or define variables, use

@example
@@lilypond[verbatim,quote]
@end example

In rare cases, other options may be used (or omitted), but ask first.

@item
Inspirational headwords are produced with

@example
@@lilypondfile[quote,ragged-right,line-width=16\cm,staffsize=16]
@{pitches-headword.ly@}
@end example

@item
LSR snippets are linked with

@example
@@lilypondfile[verbatim,lilyquote,ragged-right,texidoc,doctitle]
@{filename.ly@}
@end example

@noindent
excepted in Templates, where `doctitle' may be omitted.

@item
Avoid long stretches of input code.  Noone is going to read
them in print.  Please create a smaller example.  (the smaller
example does not need to be minimal, however)

@item
Specify durations for at least the first note of every bar.

@item
If possible, end with a complete bar.

@item
Comments should go on their own line, and be placed before
the line(s) to which they refer.

@item
Add extra spaces around @{ @} marks; ie

@example
not:          \chordmode @{c e g@}
but instead:  \chordmode @{ c e g @}
@end example

@item
If you only have one bar per line, omit bar checks.  If you
put more than one bar per line (not recommended), then include bar
checks.

@item
If you want to work on an example outside of the manual (for
easier/faster processing), use this header:

@example
\paper @{
  #(define dump-extents #t)
  indent = 0\mm
  line-width = 160\mm - 2.0 * 0.4\in
  ragged-right = ##t
  force-assignment = #""
  line-width = #(- line-width (* mm  3.000000))
@}

\layout @{
@}
@end example

You may not change any of these values.  If you are making an
example demonstrating special \paper@{@} values, contact the
Documentation Editor.

@end itemize


@node Text formatting
@subsection Text formatting

@itemize

@item
Lines should be less than 72 characters long.  (I personally
recommend writing with 66-char lines, but don't bother modifying
existing material.)

@item
Do not use tabs.

@item
Do not use spaces at the beginning of a line (except in
@@example or @@verbatim environments), and do not use more than a
single space between words.  `makeinfo' copies the input lines
verbatim without removing those spaces.

@item
Use two spaces after a period.

@item
In examples of syntax, use @@var@{musicexpr@} for a music
expression.

@item
Don't use @@rinternals@{@} in the main text.  If you're
tempted to do so, you're probably getting too close to "talking
through the code".  If you really want to refer to a context, use
@@code@{@} in the main text and @@rinternals@{@} in the @@seealso.

@item
Variables or numbers which consist of a single character
(probably followed by a punctuation mark) should be tied properly,
either to the previous or the next word.  Example:

@example
The variable@@tie@{@}@@var@{a@} ...
@end example

@item
To get consistent indentation in the DVI output it is better
to avoid the @@verbatim environment.  Use the @@example
environment instead if possible, but without extraneous
indentation.  For example, this

@example
@@example
  foo @{
    bar
  @}
@@end example
@end example

@noindent
should be replaced with

@example
@@example
foo @{
  bar
@}
@@end example
@end example

@noindent
where `@@example' starts the line (without leading spaces).

@item
Do not compress the input vertically; this is, do not use

@example
    Beginning of logical unit
    @@example
    ...
    @@end example
    continuation of logical unit
@end example

@noindent
but instead do

@example
Beginning of logical unit

@@example
...
@@end example

@@noindent
continuation of logical unit
@end example

This makes it easier to avoid forgetting the `@@noindent'.  Only
use @@noindent if the material is discussing the same material;
new material should simply begin without anything special on the
line above it.

@item
in @@itemize use @@item
on a separate line like this:

@example
@@itemize
@@item
Foo

@@item
Bar
@end example

Do not use @@itemize @@bullet.

@item
To get LilyPond version, use @@version@{@} (this does not work
inside LilyPond snippets).  If you write "@@version@{@}" (enclosed
with quotes), or generally if @@version@{@} is not followed by a
space, tere will be an ugly line break in PDF output unless you
enclose it with

@example
@@w@{ ... @}

  e.g.

@@w@{"@@version@{@}"@}
@end example

@end itemize


@node Syntax survey
@subsection Syntax survey

@itemize
@item
@@c - single line comments
  "@@c NOTE:" is a comment which should remain in the final
  version.  (gp only command ;)
@item
@@ignore ... @@end ignore - multi-line comment

@item
@@cindex - General index. Please add as many as you can.  Don't
  capitalize the first word.
@item
@@funindex - is for a \lilycommand.

@item
@@example ... @@end ignore - example text that should be set as a
  blockquote.  Any @{@} must be escaped with @@@{ @}@@
@item
@@itemize @@item
A @@item
B ... @@end itemize - for bulleted lists.
  Do not compress vertically like this.

@item
@@code@{@} - typeset in a tt-font. Use for actual lilypond code or
  property/context names.  If the name contains a space, wrap
  the entire thing inside @@w@{@@code@{  @}@}.
@item
@@notation@{@} - refers to pieces of notation, e.g.
  "@@notation@{cres.@}".  Also use to specific lyrics ("the
  @@notation@{A - men@} is centered").  Only use once per subsection
  per term.
@item
@@q@{@} - Single quotes. Used for `vague' terms.
@item
@@qq@{@} - Double quotes.  Used for actual quotes ("he said") or for
  introducing special input modes.

@item
@@tie@{@} - Variables or numbers which consist of a single character
  (probably followed by a punctuation mark) should be tied
  properly, either to the previous or the next word.  Example:
    "The letter@@tie@{@}@@q@{I@} is skipped"

@item
@@var - Use for variables.
@item
@@warning@{@} - produces a "Note: " box. Use for important messages.

@item
@@bs - Generates a backslash inside @@warning.
    Any `\' used inside @@warning (and @@q or @@qq) must be written as `@@bs@{@}' 
    (texinfo would also allow \\, but this breaks with PDF output).

@end itemize



@node Other text concerns
@subsection Other text concerns

@itemize

@item
References must occur at the end of a sentence, for more
information see @@ref@{the texinfo manual@}.  Ideally this should
also be the final sentence of a paragraph, but this is not
required.  Any link in a doc section must be duplicated in the
@@seealso section at the bottom.

@item
Introducing examples must be done with

@example
     . (ie finish the previous sentence/paragaph)
     : (ie `in this example:')
     , (ie `may add foo with the blah construct,')
@end example

The old "sentence runs directly into the example" method is not
allowed any more.

@item
Abbrevs in caps, e.g., HTML, DVI, MIDI, etc.

@item
Colon usage

@enumerate

@item
To introduce lists

@item
When beginning a quote: "So, he said,...".

This usage is rarer.  Americans often just use a comma.

@item
When adding a defining example at the end of a sentence.

@end enumerate

@item
Non-ASCII characters which are in utf-8 should be directly used;
this is, don't say `Ba@@ss@{@}tuba' but `Baßtuba'.  This ensures
that all such characters appear in all output formats.

@end itemize




@node Documentation policy
@section Documentation policy

@menu
* Books::                       
* Section organization::        
* Checking cross-references::   
* General writing::             
* Technical writing style::     
@end menu

@node Books
@subsection Books

There are four parts to the documentation: the Learning Manual,
the Notation Reference, the Program Reference, and the Music
Glossary.

@itemize

@item
Learning Manual:

The LM is written in a tutorial style which introduces the most
important concepts, structure and syntax of the elements of a
LilyPond score in a carefully graded sequence of steps.
Explanations of all musical concepts used in the Manual can be
found in the Music Glossary, and readers are assumed to have no
prior knowledge of LilyPond.  The objective is to take readers to
a level where the Notation Reference can be understood and
employed to both adapt the templates in the Appendix to their
needs and to begin to construct their own scores.  Commonly used
tweaks are introduced and explained.  Examples are provided
throughout which, while being focussed on the topic being
introduced, are long enough to seem real in order to retain the
readers' interest.  Each example builds on the previous material,
and comments are used liberally.  Every new aspect is thoroughly
explained before it is used.

Users are encouraged to read the complete Learning Manual from
start-to-finish.


@item
Notation Reference: a (hopefully complete) description of LilyPond
input notation.  Some material from here may be duplicated in the
Learning Manual (for teaching), but consider the NR to be the
"definitive" description of each notation element, with the LM
being an "extra".  The goal is _not_ to provide a step-by-step
learning environment -- do not avoid using notation that has not
be introduced previously in the NR  (for example, use \break if
appropriate).  This section is written in formal technical writing
style.

Avoid duplication.  Although users are not expected to read this
manual from start to finish, they should be familiar with the
material in the Learning Manual (particularly ``Fundamental
Concepts''), so do not repeat that material in each section of
this book.  Also watch out for common constructs, like ^ - _ for
directions -- those are explained in NR 3.  In NR 1, you can
write: DYNAMICS may be manually placed above or below the staff,
see @@ref@{Controlling direction and placement@}.

Most tweaks should be added to LSR and not placed directly in the
.itely file.  In some cases, tweaks may be placed in the main
text, but ask about this first.

Finally, you should assume that users know what the notation
means; explaining musical concepts happens in the Music Glossary.


@item
Application Usage: information about using the program lilypond
with other programs (lilypond-book, operating systems, GUIs,
convert-ly, etc).  This section is written in formal technical
writing style.

Users are not expected to read this manual from start to finish.


@item
Music Glossary: information about the music notation itself.
Explanations and translations about notation terms go here.

Users are not expected to read this manual from start to finish.

@item
Internals Reference: not really a documentation book, since it is
automagically generated from the source, but this is its name.

@end itemize


@node Section organization
@subsection Section organization

@itemize

@item
The order of headings inside documentation sections should
be:

@example
main docs
@@predefined
@@endpredefined
@@snippets
@@seealso
@@knownissues
@end example

@item
You @emph{must} include a @@seealso.

@itemize
@item
The order of items inside the @@seealso section is

@example
Music Glossary:
@@rglos@{foo@},
@@rglos@{bar@}.

Learning Manual:
@@rlearning@{baz@},
@@rlearning@{foozle@}.

Notation Reference:
@@ruser@{faazle@},
@@ruser@{boo@}.

Application Usage:
@@rprogram@{blah@}.

Installed Files:
@@file@{path/to/dir/blahz@}.

Snippets: @@rlsr@{section@}.

Internals Reference:
@@rinternals@{fazzle@},
@@rinternals@{booar@}.
@end example

@item
If there are multiple entries, separate them by commas but do not
include an `and'.

@item
Always end with a period.

@item
Place each link on a new line as above; this makes it much easier
to add or remove links.  In the output, they appear on a single
line.

("Snippets" is REQUIRED; the others are optional)

@item
Any new concepts or links which require an explanation should go
as a full sentence(s) in the main text.

@item
Don't insert an empty line between @@seealso and the first entry!
Otherwise there is excessive vertical space in the PDF output.

@end itemize

@item
To create links, use @@ref@{@} if the link is within the same
manual.

@item
@@predefined ... @@endpredefined is for commands in ly/*-init.ly
FIXME?

@item
Do not include any real info in second-level sections (ie 1.1
Pitches).  A first-level section may have introductory material,
but other than that all material goes into third-level sections
(ie 1.1.1 Writing Pitches).

@end itemize


@node Checking cross-references
@subsection Checking cross-references

Cross-references between different manuals are heavily used in the
documentation, but they are not checked during compilation.
However, if you compile the documentation, a script called
check_texi_refs can help you with checking and fixing these
cross-references; for information on usage, cd into a source tree
where documentation has been built, cd into Documentation and look
for check-xrefs and fix-xrefs targets in 'make help' output.  Note
that you have to find yourself the source files to fix
cross-references in the generated documentation such as the
Internals Reference; e.g. you can grep scm/ and lily/.


@node General writing
@subsection General writing

@itemize

@item
Do not forget to create @@cindex entries for new sections of text.
Enter commands with @@funindex, i.e.

@example
@@cindex pitches, writing in different octaves
@@funindex \relative
@end example

@noindent
do not bother with the @@code@{@} (they are added automatically).
These items are added to both the command index and the unified
index.

Both index commands should go in front of the actual material.

@@cindex entries should not be capitalized, ie

@example
@@cindex time signature
@end example

@noindent
is preferred instead of @qq{Time signature},   Only use capital
letters for musical terms which demand them, like D.S. al Fine.

For scheme functions, only include the final part, i.e.,

@example
@@funindex modern-voice-cautionary
     and NOT
@@funindex #(set-accidental-style modern-voice-cautionary)
@end example

@item
Preferred terms:

@itemize

@item
In general, use the American spellings.  The internal lilypond
property names use this spelling.

@item
List of specific terms:

@example
canceled
simultaneous    NOT concurrent
measure: the unit of music
bar line: the symbol delimiting a measure   NOT barline
note head   NOT notehead
chord construct   NOT chord (when referring to <>)
@end example

@end itemize

@end itemize


@node Technical writing style
@subsection Technical writing style

These refer to the NR.  The LM uses a more gentle, colloquial
style.

@itemize

@item
Do not refer to LilyPond in the text.  The reader knows what the
manual is about.  If you do, capitalization is LilyPond.

@item
If you explicitly refer to @q{lilypond} the program (or any other
command to be executed), write @code{@@command@{lilypond@}}.

@item
Do not explicitly refer to the reader/user.  There is no one else
besides the reader and the writer.

@item
Avoid contractions (don't, won't, etc.).  Spell the words out completely.

@item
Avoid abbreviations, except for commonly used abbreviations of foreign
language terms such as etc. and i.e.

@item
Avoid fluff (@qq{Notice that,} @qq{as you can see,}
@qq{Currently,}).

@item
The use of the word @q{illegal} is inappropriate in most cases.
Say @q{invalid} instead.

@end itemize


@node Tips for writing docs
@section Tips for writing docs

In the NR, I highly recommend focusing on one subsection at a
time.  For each subsection,

@itemize

@item
check the mundane formatting.  Are the headings (@@predefined,
@@seealso, etc.) in the right order?

@item
add any appropriate index entries.

@item
check the links in the @@seealso section -- links to music
glossary, internal references, and other NR sections are the main
concern.  Check for potential additions.

@item
move LSR-worthy material into LSR.  Add the snippet, delete the
material from the .itely file, and add a @@lilypondfile command.

@item
check the examples and descriptions.  Do they still work?
@strong{Do not} assume that the existing text is
accurate/complete; some of the manual is highly out of date.

@item
is the material in the @@knownissues  still accurate?

@item
can the examples be improved (made more explanatory), or is there
any missing info?  (feel free to ask specific questions on -user;
a couple of people claimed to be interesting in being
@qq{consultants} who would help with such questions)

@end itemize

In general, I favor short text explanations with good examples --
@qq{an example is worth a thousand words}.  When I worked on the
docs, I spent about half my time just working on those tiny
lilypond examples.  Making easily-understandable examples is much
harder than it looks.


@subsubheading TWEAKS

In general, any \set or \override commands should go in the
@qq{select snippets} section, which means that they should go in
LSR and not the .itely file.  For some cases, the command
obviously belongs in the @qq{main text} (i.e. not inside
@@predefined or @@seealso or whatever) -- instrument names are a
good example of this.

@example
\set Staff.instrumentName = #"foo"
@end example

On the other side of this,

@example
\override Score.Hairpin #'after-line-breaking = ##t
@end example

clearly belongs in LSR.

I'm quite willing to discuss specific cases if you think that a
tweaks needs to be in the main text.  But items that can go into
LSR are easier to maintain, so I'd like to move as much as
possible into there.


It would be @qq{nice} if you spent a lot of time crafting nice
tweaks for users... but my recommendation is @strong{not} to do
this.  There's a lot of doc work to do without adding examples of
tweaks.  Tweak examples can easily be added by normal users by adding
them to the LSR.

One place where a documentation writer can profitably spend time writing
or upgrading tweaks is creating tweaks to deal with known issues.  It
would be ideal if every significant known issue had a workaround to avoid
the difficulty.


@node Updating docs with convert-ly
@section Updating doc with convert-ly

cd into Documentation and run

@example
find . -name '*.itely' | xargs convert-ly -e
@end example

@noindent
(This also updates translated docs.)




@node Translating the documentation
@section Translating the documentation