Re: Auto-transposition

[David Kastrup]

Thomas Morley  writes:

> 2017-12-18 16:08 GMT+01:00 Kieren MacMillan :
>
>> I suppose I could use single variables more if Lilypond had better
>> "at runtime" methods of reusing material (e.g., inline variable
>> definition)
>
> Have a look at
> https://lilypondforum.de/index.php/topic,195.msg1231.html#msg1231
> Though, I have no clue about the consequences...

A number of things wrong with it.  For one, it's not well-defined just
when the variable is available.  For another, there is no point in using
a string? argument when you actually need a symbol? argument.  And worst
of all, the command does not bother creating a copy, so any destructive
manipulations on one variable (like \relative, \transpose, and others)
will magically appear in the other variable.

Not particularly enthused about the side effect in the middle of music
either: people might expect \tag to have an effect on whether or not
some sequence is defined in some branch, but it won't.  But that's not
all that dissimilar to surprises about \relative and \tag .

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi Harm,

> Have a look at
> https://lilypondforum.de/index.php/topic,195.msg1231.html#msg1231

Ooh! It even works outside of the parent variable (as would be absolutely 
necessary in my usage):

%%%  SNIPPET BEGINS
\version "2.19.80"

tee = #(define-music-function (name mus) (string? ly:music?)
  (ly:parser-define! (string->symbol name) mus)
  mus)

testing_this = {
  g'1
  \repeat volta 2 \tee MusicRi {
c''1
e''1_\markup \halign #-1.5 "Fine"
  }
  \repeat volta 2 {
g''1
d''1_\markup \halign #-0.2 \column {
 \right-align "D.S. al Fine" \right-align "e poi"
}
  }
  \MusicRi
}

another_test = {
  R1
  \MusicRi
  R1*2
  \MusicRi
}

\score {
  <<
\new Staff \testing_this
\new Staff \another_test
  >>
}
%%%  SNIPPET ENDS

> Though, I have no clue about the consequences...

Hmmm… It would be nice to hear from others about whether this construction 
makes sense, is "safe", etc.
Certainly the ability to define variables "in-line" would be a bit of a 
game-changer for my Lily-coding.

Thanks!
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Thomas Morley]

2017-12-18 16:08 GMT+01:00 Kieren MacMillan :

> I suppose I could use single variables more if Lilypond had better "at 
> runtime" methods of reusing material (e.g., inline variable definition)

Have a look at
https://lilypondforum.de/index.php/topic,195.msg1231.html#msg1231
Though, I have no clue about the consequences...

Cheers,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Wol's lists]

On 18/12/17 15:08, Kieren MacMillan wrote:

Hi Saul,


\resetRelativeOctave belongs IMO always as part of the definition of a music 
expression, ideally on the line directly preceding note entry.

Agreed.

Until it's a repeated phrase, but not necessarily the same octave. iirc 
(as I said it's a long time ago) this phrase kept cropping up in the music.


Horses for courses, and I think I didn't put \resetRelativeOctave in the 
variable itself because it didn't work for some reason.


Cheers,
Wol

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi David,

> This is, in part, a result of our documentation being so granular.

True.

> Putting it another way: I doubt that many of the issues with \relative rear 
> their heads in the examples in the docs. 

So true. And thus newbies aren't really [explicitly] taught how to use relative 
mode correctly (e.g., “as early as possible”), so when they start to engrave 
anything more complex than the simple doc examples, they run into problems 
(hence all the questions posted to the list).

> A discussion of the pros and cons of each method with examples could be 
> appropriate in perhaps the LM, or in 
> http://lilypond.org/doc/v2.19/Documentation/notation/writing-pitches

Agreed.

Best,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi Saul,

> \resetRelativeOctave belongs IMO always as part of the definition of a music 
> expression, ideally on the line directly preceding note entry.

Agreed.

> Personally, I think it is preferable to define all of the music for each 
> context by explicitly typing it in a single variable. Using music variables 
> for fragments or phrases which are then transformed or combined is a recipe 
> for code that is difficult to read and debug, and IMO virtually never helps 
> the compositional process in the long run. There may be use cases where that 
> sort of thing is appropriate, but I imagine they must be very far from the 
> kind of work I do.

Hmm… Essentially every form of music that I compose and engrave benefits from 
using and combining music variables:
  1. pop and musical theatre forms (repeats, transposed last verses, etc.);
  2. choral music (choral unisons, strict counterpoint, etc.);
  3. new concert music (row manipulations, etc.);
  4. orchestral works (repeats, counterpoint, doublings, etc.).

Explicitly typing everything into a single variable for each context is 
time-consuming and error-prone (though admittedly Frescobaldi has *some* tools 
that alleviate *some* of those issues). I suppose I could use single variables 
more if Lilypond had better "at runtime" methods of reusing material (e.g., 
inline variable definition) — but given the current toolbox, I shudder to think 
how much less efficient my work would be if I was limited to single variables 
per context.

Best,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Saul Tobin]

I would highly recommend against this type of coding style.
\resetRelativeOctave belongs IMO always as part of the definition of a
music expression, ideally on the line directly preceding note entry.

Personally, I think it is preferable to define all of the music for each
context by explicitly typing it in a single variable. Using music variables
for fragments or phrases which are then transformed or combined is a recipe
for code that is difficult to read and debug, and IMO virtually never helps
the compositional process in the long run. There may be use cases where
that sort of thing is appropriate, but I imagine they must be very far from
the kind of work I do.

On Dec 15, 2017 7:54 AM, "Wols Lists"  wrote:

On 15/12/17 13:45, David Wright wrote:
> On Fri 15 Dec 2017 at 10:02:19 (+), Wols Lists wrote:
>> On 15/12/17 06:20, Saul Tobin wrote:
>>> Relative mode makes perfect sense if you're entering music that cares
>>> mainly about the relationship between notes within a phrase (i.e. most
>>> music). IMO absolute mode might be easier from the perspective of the
>>> software, but it's not how most musicians think, and that's
>>> important. Maybe the documentation could do a better job explaining the
>>> semantics of relative mode and when to use \resetRelativeOctave?
>>>
>>> I take exception to the idea that relative mode ought to be deprecated.
>>> I've been using exclusively relative mode to compose for almost ten
>>> years, and I think it's great.
>>
>> I think Han-Wen actually wrote \resetRelativeOctave for me :-)
>>
>> But if you don't understand relative then it will mess you up.
>>
>> Does anybody (not me :-) want to write a little update for the docu that
>> will make both relative mode and \resetRelativeOctave (hopefully) clear?
>>
>> It originated when I was (iirc) transcribing Chattanooga Choo-Choo, and
>> there's a repeated phrase, so I thought I'd define it as a variable.
>> OOOPPSS! The starting and ending notes are a fifth or more apart, and
>> the phrase repeats with nothing else in-between. The resulting staircase
>> was spectacular!
>>
>> If somebody would care to take that as hint for putting an example in
>> the docu, that's fine by me! :-)
>
> Just use \relative early.
>
I was thinking more along the lines of (note this is NOT TESTED)

phrase = { c f d g }

\relative { \phrase \phrase \phrase }

\relative { \resetRelativeOctave \phrase \resetRelativeOctave \phrase
\resetRelativeOctave \phrase }

Cheers,
Wol


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Flaming Hakama by Elaine]

> Subject: Re: Re: Auto-transposition
>


> By the way, whoever is working on the Changes document for upcoming 2.20,
> don't forget to mention about \fixed!
>
>/Mats
>

I volunteered to do this, so I will make sure to work on describing this
feature.


>  > I don't think there is a clear advantage to use relative vs absolute.
>
> I think there are clear advantages and disadvantages to both. The main
> problem I see is the insistence (implicit in the documentation) that
> relative mode is the best for newbies to start with, and the volume of list
> posts from newbies having problems with relative mode makes it clear that
> that is not true. Perhaps if the documentation were make crystal clear,
> with \resetRelativeOctave used in every example, etc., then I could feel
> comfortable backing down from my crusade to try to save newbies from the
> pitfalls we have [unintentionally] laid for them.
>
>
>
> Best,
>
> Kieren.
>

This is, in part, a result of our documentation being so granular.

Putting it another way: I doubt that many of the issues with \relative rear
their heads in the examples in the docs.

A discussion of the pros and cons of each method with examples could be
appropriate in perhaps the LM, or in
http://lilypond.org/doc/v2.19/Documentation/notation/writing-pitches

Adding \resetRelativeOctave *everywhere* is not consistent with the MWE
nature of the examples, and is probably not needed in most cases, since it
(seems to me) makes sense to use it at the start of a new phrase, or
anywhere where the content is not dependent upon what came before.
Whereas, most doc examples consists of a single, or even a partial, phrase.

I certainly don't think it is necessary that all examples should use the
same mode of note entry.  It might be good to start revising examples where
using \fixed organically makes more sense.

A most useful approach (but probably prohibitively laborious way?) might be
to add the ability to toggle the docs dynamically to show the example in
either \relative or \fixed.


> I was thinking more along the lines of (note this is NOT TESTED)
> >
> > phrase = { c f d g }
> >
> > \relative { \phrase \phrase \phrase }
>
> Yes, that's exactly the problem with putting \relative around
> constructions rather than the variables themselves, illustrated
> by "mover".



Since \relative { } is an absolute construct, this is easily solved by:

phrase = \relative c' { c f d g }
\relative { \phrase \phrase \phrase }




David Elaine Alt
415 . 341 .4954   "*Confusion is
highly underrated*"
ela...@flaminghakama.com
skype: flaming_hakama
Producer ~ Composer ~ Instrumentalist
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

Flaming Hakama by Elaine  writes:

> On Fri, Dec 15, 2017 at 7:57 AM, David Kastrup  wrote:
>
>> Flaming Hakama by Elaine  writes:
>>
>> > Actually, I think it had to do with midi output for bass clarinet.   The
>> > sample I used was not transposed the octave, and I had to compensate by
>> > transposing the music down.
>>
>> According to Wikipedia, bass clarinet is written in treble clef but
>> played one octave lower than written (after subtracting the usual B♭
>> transposition).  So, uh, I have no idea?
>
> Correct, I'd entered the music in the printed octave,
> and wanted to hear it in the sounding octave.

\transposition c

should do the trick for Midi.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Wols Lists]

On 15/12/17 17:48, Flaming Hakama by Elaine wrote:
> > Actually, I think it had to do with midi output for bass
> clarinet.   The
> > sample I used was not transposed the octave, and I had to
> compensate by
> > transposing the music down.
> 
> According to Wikipedia, bass clarinet is written in treble clef but
> played one octave lower than written (after subtracting the usual B♭
> transposition).  So, uh, I have no idea?
> 
> 
> 
> Correct, I'd entered the music in the printed octave, 
> and wanted to hear it in the sounding octave.
> 
Which is actually why I always like \relative when entering music. I
just alter the reference pitch up or down an octave to suit ...

I gather \fixed would have the same effect - I might have to try
changing to that because it would probably help avoid me getting stuff
going all over the place - I do tend to enter all the notes before I
compile and check it so a couple of missed commas or apostrophes results
in ledger lines galore :-)

If you want to print it with the treble rather than the treble_8 clef,
I'd be inclined to enter it in the correct octave with \relative then
\transpose it up the octave just for printing.

Cheers,
Wol

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Flaming Hakama by Elaine]

On Fri, Dec 15, 2017 at 7:57 AM, David Kastrup  wrote:

> Flaming Hakama by Elaine  writes:
>
> > David Kastrup  writes:
> >
> >> Flaming Hakama by Elaine  writes:
> >>
> >>> And the behavior of \relative that is weirdest:
> >>>
> >>> * The octave of \relative is lowered when used with bass clef.
> >>>
> >>> The fix for the clef issue is to use a \transpose when you use it
> >>> with bass clef.
> >>
> >> That one is nonsense.  \relative does not interact in any way with the
> >> bass clef.
> >
> >> Best guess: choral music with tenor (injudiciously) written in \clef
> >> "treble" rather than \clef "treble_8" confused your sense of octave.
> >
> >
> > Actually, I think it had to do with midi output for bass clarinet.   The
> > sample I used was not transposed the octave, and I had to compensate by
> > transposing the music down.
>
> According to Wikipedia, bass clarinet is written in treble clef but
> played one octave lower than written (after subtracting the usual B♭
> transposition).  So, uh, I have no idea?
>
> --
> David Kastrup
>


Correct, I'd entered the music in the printed octave,
and wanted to hear it in the sounding octave.


Elaine
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Fri 15 Dec 2017 at 15:52:28 (+), Wols Lists wrote:
> On 15/12/17 13:45, David Wright wrote:
> > On Fri 15 Dec 2017 at 10:02:19 (+), Wols Lists wrote:
> >> On 15/12/17 06:20, Saul Tobin wrote:
> >>> Relative mode makes perfect sense if you're entering music that cares
> >>> mainly about the relationship between notes within a phrase (i.e. most
> >>> music). IMO absolute mode might be easier from the perspective of the
> >>> software, but it's not how most musicians think, and that's
> >>> important. Maybe the documentation could do a better job explaining the
> >>> semantics of relative mode and when to use \resetRelativeOctave?
> >>>
> >>> I take exception to the idea that relative mode ought to be deprecated.
> >>> I've been using exclusively relative mode to compose for almost ten
> >>> years, and I think it's great.
> >>
> >> I think Han-Wen actually wrote \resetRelativeOctave for me :-)
> >>
> >> But if you don't understand relative then it will mess you up.
> >>
> >> Does anybody (not me :-) want to write a little update for the docu that
> >> will make both relative mode and \resetRelativeOctave (hopefully) clear?
> >>
> >> It originated when I was (iirc) transcribing Chattanooga Choo-Choo, and
> >> there's a repeated phrase, so I thought I'd define it as a variable.
> >> OOOPPSS! The starting and ending notes are a fifth or more apart, and
> >> the phrase repeats with nothing else in-between. The resulting staircase
> >> was spectacular!
> >>
> >> If somebody would care to take that as hint for putting an example in
> >> the docu, that's fine by me! :-)
> > 
> > Just use \relative early.
> > 
> I was thinking more along the lines of (note this is NOT TESTED)
> 
> phrase = { c f d g }
> 
> \relative { \phrase \phrase \phrase }

Yes, that's exactly the problem with putting \relative around
constructions rather than the variables themselves, illustrated
by "mover".

> \relative { \resetRelativeOctave \phrase \resetRelativeOctave \phrase
> \resetRelativeOctave \phrase }

Each \resetRelativeOctave needs a pitch to set the octave, but, yes,
that is one way of dealing with the issue. A benefit is that you get
the chance to modify the octavation at each repetition, but that
benefit could also be achieved with \transpose c c' additions instead.

But the solution shown by "stayer" is simpler and more elegant, and
will also make any deliberate octavation changes more obvious in
the source.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

RE: Auto-transposition

[mogens]

> Hi List,
>
> I have used Lilypond for a few years now.  I (almost) always use
> \relative and I have never had trouble with it.  Indeed, as a lay
> person, I have no idea which octave
>   c'4 d' e' f' g'1
> is – I know I could look it up, but it is certainly not in my head.

So what reference pitch do you put after \relative ?  Or do you just do
it by trial and error?

Something like: when writing in treble clef, the first note should likely have 
a single apostrophe.  Errors are obvious and trivial to fix (and therefore not 
distressing as all).

Regards,
Mogens

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

Flaming Hakama by Elaine  writes:

> David Kastrup  writes:
>
>> Flaming Hakama by Elaine  writes:
>>
>>> And the behavior of \relative that is weirdest:
>>>
>>> * The octave of \relative is lowered when used with bass clef.
>>>
>>> The fix for the clef issue is to use a \transpose when you use it
>>> with bass clef.
>>
>> That one is nonsense.  \relative does not interact in any way with the
>> bass clef.
>
>> Best guess: choral music with tenor (injudiciously) written in \clef
>> "treble" rather than \clef "treble_8" confused your sense of octave.
>
>
> Actually, I think it had to do with midi output for bass clarinet.   The
> sample I used was not transposed the octave, and I had to compensate by
> transposing the music down.

According to Wikipedia, bass clarinet is written in treble clef but
played one octave lower than written (after subtracting the usual B♭
transposition).  So, uh, I have no idea?

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Wols Lists]

On 15/12/17 13:45, David Wright wrote:
> On Fri 15 Dec 2017 at 10:02:19 (+), Wols Lists wrote:
>> On 15/12/17 06:20, Saul Tobin wrote:
>>> Relative mode makes perfect sense if you're entering music that cares
>>> mainly about the relationship between notes within a phrase (i.e. most
>>> music). IMO absolute mode might be easier from the perspective of the
>>> software, but it's not how most musicians think, and that's
>>> important. Maybe the documentation could do a better job explaining the
>>> semantics of relative mode and when to use \resetRelativeOctave?
>>>
>>> I take exception to the idea that relative mode ought to be deprecated.
>>> I've been using exclusively relative mode to compose for almost ten
>>> years, and I think it's great.
>>
>> I think Han-Wen actually wrote \resetRelativeOctave for me :-)
>>
>> But if you don't understand relative then it will mess you up.
>>
>> Does anybody (not me :-) want to write a little update for the docu that
>> will make both relative mode and \resetRelativeOctave (hopefully) clear?
>>
>> It originated when I was (iirc) transcribing Chattanooga Choo-Choo, and
>> there's a repeated phrase, so I thought I'd define it as a variable.
>> OOOPPSS! The starting and ending notes are a fifth or more apart, and
>> the phrase repeats with nothing else in-between. The resulting staircase
>> was spectacular!
>>
>> If somebody would care to take that as hint for putting an example in
>> the docu, that's fine by me! :-)
> 
> Just use \relative early.
> 
I was thinking more along the lines of (note this is NOT TESTED)

phrase = { c f d g }

\relative { \phrase \phrase \phrase }

\relative { \resetRelativeOctave \phrase \resetRelativeOctave \phrase
\resetRelativeOctave \phrase }

Cheers,
Wol


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Flaming Hakama by Elaine]

On Dec 15, 2017 12:59 AM, "David Kastrup"  wrote:

David Kastrup  writes:

> Flaming Hakama by Elaine  writes:
>
>> And the behavior of \relative that is weirdest:
>>
>> * The octave of \relative is lowered when used with bass clef.
>>
>> The fix for the clef issue is to use a \transpose when you use it with
bass
>> clef.
>
> That one is nonsense.  \relative does not interact in any way with the
> bass clef.

Best guess: choral music with tenor (injudiciously) written in \clef
"treble" rather than \clef "treble_8" confused your sense of octave.

--
David Kastrup


Actually, I think it had to do with midi output for bass clarinet.   The
sample I used was not transposed the octave, and I had to compensate by
transposing the music down.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

 writes:

> Hi List,
>
> I have used Lilypond for a few years now.  I (almost) always use
> \relative and I have never had trouble with it.  Indeed, as a lay
> person, I have no idea which octave
>   c'4 d' e' f' g'1
> is – I know I could look it up, but it is certainly not in my head.

So what reference pitch do you put after \relative ?  Or do you just do
it by trial and error?

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

Kieren MacMillan  writes:

> Hi all,
>
> Mats wrote:
>> By the way, whoever is working on the Changes document for upcoming
>> 2.20, don't forget to mention about \fixed!
>
> +1

2.19.22 apparently.  I'd have pegged it at 2.18 already.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

RE: Auto-transposition

[mogens]

Hi List,

I have used Lilypond for a few years now.  I (almost) always use \relative and 
I have never had trouble with it.  Indeed, as a lay person, I have no idea 
which octave
  c'4 d' e' f' g'1
is – I know I could look it up, but it is certainly not in my head.  So for 
someone with little or no formal musical training, I think \relative is the 
best way to go – and I selfishly think that the documentation should be written 
for people such as me.

Regards,
Mogens



From: Kieren MacMillan
Sent: December 15, 2017 7:28
To: Lilypond-User Mailing List
Cc: Mats Bengtsson
Subject: Re: Auto-transposition

Hi all,

Mats wrote:
> By the way, whoever is working on the Changes document for upcoming 2.20, 
> don't forget to mention about \fixed!

+1

David W wrote:
> Just use \relative early.

The fact that this works like it does is further evidence that \relative is not 
sufficiently intuitive. I'm pretty sure if you asked 100 newbies — or even 
not-so-newbies — what the output of your code would be, there would be a 
significant subset (maybe even a majority?!) that wouldn't answer correctly.

Saul wrote:
> IMO absolute mode might be easier from the perspective of the software, but 
> it's not how most musicians think

I disagree. In fact, I would offer that the situation is essentially the 
opposite of what you suggest: to wit, if I extract a line out of the middle of 
a relative block

   c4 d e f g1

and ask what pitches (including octave!) those are, no human musician can tell 
me (better than random chance), whereas software *can*. On the other hand, if I 
pull the same line out of the middle of an absolute block

  c'4 d' e' f' g'1

both human and software can get it (i.e., the octave) correct 100% of the time.

> Maybe the documentation could do a better job explaining the semantics of 
> relative mode and when to use \resetRelativeOctave?

That's definitely true.

Gianmaria wrote:
> I don't think there is a clear advantage to use relative vs absolute.

I think there are clear advantages and disadvantages to both. The main problem 
I see is the insistence (implicit in the documentation) that relative mode is 
the best for newbies to start with, and the volume of list posts from newbies 
having problems with relative mode makes it clear that that is not true. 
Perhaps if the documentation were make crystal clear, with \resetRelativeOctave 
used in every example, etc., then I could feel comfortable backing down from my 
crusade to try to save newbies from the pitfalls we have [unintentionally] laid 
for them.

Best,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi all,

Mats wrote:
> By the way, whoever is working on the Changes document for upcoming 2.20, 
> don't forget to mention about \fixed!

+1

David W wrote:
> Just use \relative early.

The fact that this works like it does is further evidence that \relative is not 
sufficiently intuitive. I'm pretty sure if you asked 100 newbies — or even 
not-so-newbies — what the output of your code would be, there would be a 
significant subset (maybe even a majority?!) that wouldn't answer correctly.

Saul wrote:
> IMO absolute mode might be easier from the perspective of the software, but 
> it's not how most musicians think

I disagree. In fact, I would offer that the situation is essentially the 
opposite of what you suggest: to wit, if I extract a line out of the middle of 
a relative block

   c4 d e f g1

and ask what pitches (including octave!) those are, no human musician can tell 
me (better than random chance), whereas software *can*. On the other hand, if I 
pull the same line out of the middle of an absolute block

  c'4 d' e' f' g'1

both human and software can get it (i.e., the octave) correct 100% of the time.

> Maybe the documentation could do a better job explaining the semantics of 
> relative mode and when to use \resetRelativeOctave?

That's definitely true.

Gianmaria wrote:
> I don't think there is a clear advantage to use relative vs absolute.

I think there are clear advantages and disadvantages to both. The main problem 
I see is the insistence (implicit in the documentation) that relative mode is 
the best for newbies to start with, and the volume of list posts from newbies 
having problems with relative mode makes it clear that that is not true. 
Perhaps if the documentation were make crystal clear, with \resetRelativeOctave 
used in every example, etc., then I could feel comfortable backing down from my 
crusade to try to save newbies from the pitfalls we have [unintentionally] laid 
for them.

Best,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Re: Auto-transposition

[Mats Bengtsson]

On 2017-12-15 13:30, Ralph Palmer wrote:


Greetings -

Just a reminder and note:

I have been using \relative for years and have been quite happy with 
it. Perhaps my situation is fairly unique. I transcribe a lot of 
fiddle tunes, then transpose them down an octave and into alto clef. 
Using \relative is easier for me than using absolute pitches and 
\transpose, especially when I need to move an A part or B part up or 
down an octave because the fiddle tune won't all fit on the viola down 
an octave.


My point being that I understand the arguments for absolute pitch 
entry and I prefer \relative pitch entry. I think there's room for both.


All the best,

Ralph


Agreed!

By the way, whoever is working on the Changes document for upcoming 
2.20, don't forget to mention about \fixed!


   /Mats

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Fri 15 Dec 2017 at 10:02:19 (+), Wols Lists wrote:
> On 15/12/17 06:20, Saul Tobin wrote:
> > Relative mode makes perfect sense if you're entering music that cares
> > mainly about the relationship between notes within a phrase (i.e. most
> > music). IMO absolute mode might be easier from the perspective of the
> > software, but it's not how most musicians think, and that's
> > important. Maybe the documentation could do a better job explaining the
> > semantics of relative mode and when to use \resetRelativeOctave?
> > 
> > I take exception to the idea that relative mode ought to be deprecated.
> > I've been using exclusively relative mode to compose for almost ten
> > years, and I think it's great.
> 
> I think Han-Wen actually wrote \resetRelativeOctave for me :-)
> 
> But if you don't understand relative then it will mess you up.
> 
> Does anybody (not me :-) want to write a little update for the docu that
> will make both relative mode and \resetRelativeOctave (hopefully) clear?
> 
> It originated when I was (iirc) transcribing Chattanooga Choo-Choo, and
> there's a repeated phrase, so I thought I'd define it as a variable.
> OOOPPSS! The starting and ending notes are a fifth or more apart, and
> the phrase repeats with nothing else in-between. The resulting staircase
> was spectacular!
> 
> If somebody would care to take that as hint for putting an example in
> the docu, that's fine by me! :-)

Just use \relative early.

Cheers,
David.
mover = { c' c' d' e' f' g' a' a' }
stayer = \relative { c' c d e f g a a }

mresult = \relative { \mover\mover\mover\mover }
sresult = \relative { \stayer\stayer\stayer\stayer }

\mresult
\sresult


wot-u-dun.pdf
Description: Adobe PDF document
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Ralph Palmer]

On Thu, Dec 14, 2017 at 5:04 PM, Neo Anderson  wrote:

> Dear All,
>
> Anyway, I'm thankful you've taken your time to solve my problem.
>
> On Thursday, December 14, 2017, 1:29:22 AM GMT+1, Kieren MacMillan <
> kieren_macmil...@sympatico.ca> wrote:
>
>
> Hi David,
>
> >> Because my only goal was to clarify for future readers of the list that
> "manipulations on stuff" are not the only thing that can cause headaches.
> > I failed to see how cut and paste differed from manipulations on stuff.
>
>
> >> For future list-readers: Using absolute and fixed entry modes allows
> you to avoid the issues/effort described by David.
> > If people would only cease bringing any questions about \relative
> > to the list, we'd be able to avoid this particular drum beat.
>
> Agreed. But I would offer that the reason people (esp. newbies)
> continually bring questions about \relative to the list is because it's not
> as intuitive and "idiot-proof" as our documentation would suggest — which
> is precisely why I keep beating the drum, to try to keep people from making
> the same decade-long mistake that I did (which was to use \relative, as
> suggested/implied in the official documentation).
>
> Cheers,
> Kieren.
>

Greetings -

Just a reminder and note:

I have been using \relative for years and have been quite happy with it.
Perhaps my situation is fairly unique. I transcribe a lot of fiddle tunes,
then transpose them down an octave and into alto clef. Using \relative is
easier for me than using absolute pitches and \transpose, especially when I
need to move an A part or B part up or down an octave because the fiddle
tune won't all fit on the viola down an octave.

My point being that I understand the arguments for absolute pitch entry and
I prefer \relative pitch entry. I think there's room for both.

All the best,

Ralph

-- 
Ralph Palmer
Brattleboro, VT
USA
palmer.r.vio...@gmail.com
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Wols Lists]

On 15/12/17 06:20, Saul Tobin wrote:
> Relative mode makes perfect sense if you're entering music that cares
> mainly about the relationship between notes within a phrase (i.e. most
> music). IMO absolute mode might be easier from the perspective of the
> software, but it's not how most musicians think, and that's
> important. Maybe the documentation could do a better job explaining the
> semantics of relative mode and when to use \resetRelativeOctave?
> 
> I take exception to the idea that relative mode ought to be deprecated.
> I've been using exclusively relative mode to compose for almost ten
> years, and I think it's great.

I think Han-Wen actually wrote \resetRelativeOctave for me :-)

But if you don't understand relative then it will mess you up.

Does anybody (not me :-) want to write a little update for the docu that
will make both relative mode and \resetRelativeOctave (hopefully) clear?

It originated when I was (iirc) transcribing Chattanooga Choo-Choo, and
there's a repeated phrase, so I thought I'd define it as a variable.
OOOPPSS! The starting and ending notes are a fifth or more apart, and
the phrase repeats with nothing else in-between. The resulting staircase
was spectacular!

If somebody would care to take that as hint for putting an example in
the docu, that's fine by me! :-)

Cheers,
Wol

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

David Kastrup  writes:

> Flaming Hakama by Elaine  writes:
>
>> And the behavior of \relative that is weirdest:
>>
>> * The octave of \relative is lowered when used with bass clef.
>>
>> The fix for the clef issue is to use a \transpose when you use it with bass
>> clef.
>
> That one is nonsense.  \relative does not interact in any way with the
> bass clef.

Best guess: choral music with tenor (injudiciously) written in \clef
"treble" rather than \clef "treble_8" confused your sense of octave.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

Flaming Hakama by Elaine  writes:

> And the behavior of \relative that is weirdest:
>
> * The octave of \relative is lowered when used with bass clef.
>
> The fix for the clef issue is to use a \transpose when you use it with bass
> clef.

That one is nonsense.  \relative does not interact in any way with the
bass clef.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Flaming Hakama by Elaine]

From: Flaming Hakama by Elaine <ela...@flaminghakama.com>
> Subject: Re: Auto-transposition



It was pointed out to me that this claim is untrue:


> And the behavior of \relative that is weirdest:
>
> * The octave of \relative is lowered when used with bass clef.
>
> The fix for the clef issue is to use a \transpose when you use it with
> bass clef.
>


I think I was remembering an issue from some time ago which I probably
misdiagnosed.

Sorry for the noise.



David Elaine Alt
415 . 341 .4954   "*Confusion is
highly underrated*"
ela...@flaminghakama.com
skype: flaming_hakama
Producer ~ Composer ~ Instrumentalist
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Gianmaria Lari]

On 15 December 2017 at 07:20, Saul Tobin  wrote:

> Reading your example, it seems to me that the "FAIL" is caused by the
> hypothetical user misunderstanding the semantics of relative mode. Using
> relative mode without explicitly defining what pitch the phrase is meant to
> be relative to is unreadable and prone to break. If it were written
> properly, it would be:
>
> \relative {
>   \resetRelativeOctave c'
>   c4 d e f g1  % a rising diatonic scale, starting on middle C
>  % if the next line is meant to start on the G above middle C, it should
> have a new \resetRelativeOctave
>   g'4 f e d c1  % written relatively, this means we care about the melodic
> leap upward from the previous note
>   %  now I want to reuse the rising diatonic scale… so I cut and paste the
> first two lines:
>\resetRelativeOctave c'
>c4 d e f g1  % a rising diatonic scale, starting on middle C — no
> problem!
> }
>
> Relative mode makes perfect sense if you're entering music that cares
> mainly about the relationship between notes within a phrase (i.e. most
> music). IMO absolute mode might be easier from the perspective of the
> software, but it's not how most musicians think, and that's important. Maybe
> the documentation could do a better job explaining the semantics of
> relative mode and when to use \resetRelativeOctave?
>
> I take exception to the idea that relative mode ought to be deprecated.
> I've been using exclusively relative mode to compose for almost ten years,
> and I think it's great.
>

In short: I don't think there is a clear advantage to use relative vs
absolute. I say this because of my (small) lilypond experience (I started
with relative and now I only use absolute) and more important, because I
periodically see this type of thread on this mailing list. But please
correct me if my previous sentence is not true.

Like other people I stopped using relative mode after having more harm than
good. Using "fixed" all started working always, it is more clear and
honestly I don't have the impression you have to write/work more.

I especially think that for novice user this should be the suggested way to
work: at beginning you have a lot of problem to make things working. Even
if you could make errors in both absolute and relative mode, with absolute
also a beginner is immediately able to see where is the problem and to fix
it. With relative I remember getting frustrated wasting a lot of time
trying to understand where was the problem, fixing it and discover I was
creating another problem in another part of the music.

Now to be more clear:
- it is sure when I start using lilypond I didn't use relative correctly
(at that time didn't know \resetRelativeOctave) and
- I probably don't write that much music so that maybe I'm not the best
user to say what's the best mode.

But this is the standard beginning for everybody. So adding even more
complexity for something that maybe (maybe) give you some advantages in
future I don't think it's a good choice.
And yes, I'm not a lilypond expert but I'm an excellent  novice :) with a
lot of problem. For this reason I think the manual should:

- avoid to suggest using relative
- avoid to show the majority of examples using relative (for the large
majority of examples it is really not need any fixed/relative)

I'm sure there are other people like you Saul that can have great advantage
using \relative. But because (to me) it is not that clear if it is
generally advantageous and because (to me) it is more complex I don't think
it should be the suggested way to go. This is just my two cents.
g.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Saul Tobin]

Reading your example, it seems to me that the "FAIL" is caused by the
hypothetical user misunderstanding the semantics of relative mode. Using
relative mode without explicitly defining what pitch the phrase is meant to
be relative to is unreadable and prone to break. If it were written
properly, it would be:

\relative {
  \resetRelativeOctave c'
  c4 d e f g1  % a rising diatonic scale, starting on middle C
 % if the next line is meant to start on the G above middle C, it should
have a new \resetRelativeOctave
  g'4 f e d c1  % written relatively, this means we care about the melodic
leap upward from the previous note
  %  now I want to reuse the rising diatonic scale… so I cut and paste the
first two lines:
   \resetRelativeOctave c'
   c4 d e f g1  % a rising diatonic scale, starting on middle C — no
problem!
}

Relative mode makes perfect sense if you're entering music that cares
mainly about the relationship between notes within a phrase (i.e. most
music). IMO absolute mode might be easier from the perspective of the
software, but it's not how most musicians think, and that's important. Maybe
the documentation could do a better job explaining the semantics of
relative mode and when to use \resetRelativeOctave?

I take exception to the idea that relative mode ought to be deprecated.
I've been using exclusively relative mode to compose for almost ten years,
and I think it's great.


On Wed, Dec 13, 2017 at 4:29 PM, Kieren MacMillan <
kieren_macmil...@sympatico.ca> wrote:

> Hi David,
>
> >> Because my only goal was to clarify for future readers of the list that
> "manipulations on stuff" are not the only thing that can cause headaches.
> > I failed to see how cut and paste differed from manipulations on stuff.
>
> I think of "manipulations on stuff" as being
>
> foo = { stuff }
> ...
> \transpose c d \foo
> \foo \foo \foo
> etc.
>
> Cut and paste on the other hand, is this:
>
> \relative {
>   c'4 d e f g1  % a rising diatonic scale, starting on middle C
>   g'4 f e d c1  % a falling diatonic scale, starting on the G a 12th above
> middle C
>   %  now I want to reuse the rising diatonic scale… so I cut and paste the
> first line:
>   c'4 d e f g1  % a rising diatonic scale, starting on middle C <-- FAIL!
> }
>
> >> For future list-readers: Using absolute and fixed entry modes allows
> you to avoid the issues/effort described by David.
> > If people would only cease bringing any questions about \relative
> > to the list, we'd be able to avoid this particular drum beat.
>
> Agreed. But I would offer that the reason people (esp. newbies)
> continually bring questions about \relative to the list is because it's not
> as intuitive and "idiot-proof" as our documentation would suggest — which
> is precisely why I keep beating the drum, to try to keep people from making
> the same decade-long mistake that I did (which was to use \relative, as
> suggested/implied in the official documentation).
>
> Cheers,
> Kieren.
>
> 
>
> Kieren MacMillan, composer
> ‣ website: www.kierenmacmillan.info
> ‣ email: i...@kierenmacmillan.info
>
>
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user
>
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Flaming Hakama by Elaine]

> -- Forwarded message --
> From: Neo Anderson <sebulb...@yahoo.com>
> To: Lilypond-User Mailing List <lilypond-user@gnu.org>
> Subject: Re: Auto-transposition
>
 ...

> >> For future list-readers: Using absolute and fixed entry modes allows
> you to avoid the issues/effort described by David.
> > If people would only cease bringing any questions about \relative
> > to the list, we'd be able to avoid this particular drum beat.
>
> Agreed. But I would offer that the reason people (esp. newbies)
> continually bring questions about \relative to the list is because it's not
> as intuitive and "idiot-proof" as our documentation would suggest — which
> is precisely why I keep beating the drum, to try to keep people from making
> the same decade-long mistake that I did (which was to use \relative, as
> suggested/implied in the official documentation).
>
> Cheers,
> Kieren.
>


I haven't found any problems with \relative that can't be solved by using
another \relative.  Except the clef issue.

Using \fixed to solve or avoid these issues is no less work than using
additional \relative's, and arguably has some down sides:

* needing more ,,, and '''--or needing more \fixed with different
reference octaves to avoid that typing
* making you think in terms of absolute octaves (not something most
musicians are familiar with)
* a single octave mistake in fixed mode will not be as visible, and
will not be as easy to find


In terms of this ongoing discussion, I think it will be most helpful to be
clear about what these common problems with \relative are.  Here are the
ones I've encountered:

* volta alternatives: the octave of the 2nd ending is based on the
previous printed note (the last note of the 1st ending), and and not the
previous musical note (the last note of the common section)
* tags: including or excluding tags that change octaves will change the
octave of what follows
* changing the last note of one phrase may affect the octave of the
subsequent phrase

So, my suggestion is when you find yourself using volta alternatives, tags,
or long pieces, just get in the habit of wrapping each one in a new
\relative.

Does anyone have any other circumstances where \relative is problematic?


And the behavior of \relative that is weirdest:

* The octave of \relative is lowered when used with bass clef.

The fix for the clef issue is to use a \transpose when you use it with bass
clef.



HTH,

David Elaine Alt
415 . 341 .4954   "*Confusion is
highly underrated*"
ela...@flaminghakama.com
skype: flaming_hakama
Producer ~ Composer ~ Instrumentalist
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi,

> it's really hard to "cease bringing any questions about \relative to the 
> list" if one of the very the first
> things you learn about Lilypond is \relative mode - it certainly was mine 
> experience with all the official
> documentation.

I agree 100%. Don't worry about your question(s) — I think David is just 
frustrated with the fact that I regularly warn newbies away from using relative 
mode (because of exactly the type of problems you ran into).

Best wishes,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi David,

>> Because my only goal was to clarify for future readers of the list that 
>> "manipulations on stuff" are not the only thing that can cause headaches.
> I failed to see how cut and paste differed from manipulations on stuff.

I think of "manipulations on stuff" as being

foo = { stuff }
...
\transpose c d \foo
\foo \foo \foo
etc.

Cut and paste on the other hand, is this:

\relative {
  c'4 d e f g1  % a rising diatonic scale, starting on middle C
  g'4 f e d c1  % a falling diatonic scale, starting on the G a 12th above 
middle C
  %  now I want to reuse the rising diatonic scale… so I cut and paste the 
first line:
  c'4 d e f g1  % a rising diatonic scale, starting on middle C <-- FAIL!
}

>> For future list-readers: Using absolute and fixed entry modes allows you to 
>> avoid the issues/effort described by David.
> If people would only cease bringing any questions about \relative
> to the list, we'd be able to avoid this particular drum beat.

Agreed. But I would offer that the reason people (esp. newbies) continually 
bring questions about \relative to the list is because it's not as intuitive 
and "idiot-proof" as our documentation would suggest — which is precisely why I 
keep beating the drum, to try to keep people from making the same decade-long 
mistake that I did (which was to use \relative, as suggested/implied in the 
official documentation).

Cheers,
Kieren.



Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Wed 13 Dec 2017 at 15:15:14 (-0500), Kieren MacMillan wrote:
> Hi David,
> 
> >> … and if you don't take extra care when simply cutting and pasting, etc. 
> >> etc. etc.
> > 
> > Yes, so I'm not sure why you cut the helpful lines above the ones
> > you quoted from my post (that were not my words, however).
> 
> Because my only goal was to clarify for future readers of the list that 
> "manipulations on stuff" are not the only thing that can cause headaches.

I failed to see how cut and paste differed from manipulations on stuff.

> > … so before employing a cut and paste approach, it might be wise to
> > convert the source to absolute. (Some composers disagree.) It just
> > depends on your approach, as every cut involves checking the note
> > following the cut, and every paste involves checking up to two
> > notes. This could be trivial, or it could be a showstopper.
> 
> Correct.
> 
> For future list-readers: Using absolute and fixed entry modes allows you to 
> avoid the issues/effort described by David.

If people would only cease bringing any questions about \relative
to the list, we'd be able to avoid this particular drum beat.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi David,

>> … and if you don't take extra care when simply cutting and pasting, etc. 
>> etc. etc.
> 
> Yes, so I'm not sure why you cut the helpful lines above the ones
> you quoted from my post (that were not my words, however).

Because my only goal was to clarify for future readers of the list that 
"manipulations on stuff" are not the only thing that can cause headaches.

> … so before employing a cut and paste approach, it might be wise to
> convert the source to absolute. (Some composers disagree.) It just
> depends on your approach, as every cut involves checking the note
> following the cut, and every paste involves checking up to two
> notes. This could be trivial, or it could be a showstopper.

Correct.

For future list-readers: Using absolute and fixed entry modes allows you to 
avoid the issues/effort described by David.

Cheers,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Wed 13 Dec 2017 at 12:06:24 (-0500), Kieren MacMillan wrote:
> Hi all,
> 
> > Kieran is correct relative pitch entry can cause headaches
> > if you start doing manipulations on stuff.
> 
> … and if you don't take extra care when simply cutting and pasting, etc. etc. 
> etc.

Yes, so I'm not sure why you cut the helpful lines above the ones
you quoted from my post (that were not my words, however).

The lines were:

✂ ✂ ✂
On Tue 12 Dec 2017 at 22:00:47 (-0500), Shane Brandes wrote:
> Frescobaldi has a neat little tool to convert from relative to
absolute pitch.

… and AIUI the same tool is available as a standalone in ly,
aka python-ly/python3-ly/….
✂ ✂ ✂

… so before employing a cut and paste approach, it might be wise to
convert the source to absolute. (Some composers disagree.) It just
depends on your approach, as every cut involves checking the note
following the cut, and every paste involves checking up to two
notes. This could be trivial, or it could be a showstopper.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi all,

> Kieran is correct relative pitch entry can cause headaches
> if you start doing manipulations on stuff.

… and if you don't take extra care when simply cutting and pasting, etc. etc. 
etc.

Cheers,
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Ben]

On 12/13/2017 11:54 AM, David Wright wrote:



I prefer relative as a pitch entry for ease and speed.
Kieran is correct relative pitch entry can cause headaches if you
start doing manipulations on stuff.

Agreed; if definitions like

*foo = \relative { stuff in notemode ... } *
are used, then \foo can be used in complicated constructions
in safety, because its contents are absolute.




This is what I have found to be most convenient too, and I tend to do 
this in almost every score...so far, haven't had any major issues.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Tue 12 Dec 2017 at 22:00:47 (-0500), Shane Brandes wrote:
> Frescobaldi has a neat little tool to convert from relative to absolute pitch.

… and AIUI the same tool is available as a standalone in ly,
aka python-ly/python3-ly/….

> I prefer relative as a pitch entry for ease and speed.
> Kieran is correct relative pitch entry can cause headaches if you
> start doing manipulations on stuff.

Agreed; if definitions like

foo = \relative { stuff in notemode ... }

are used, then \foo can be used in complicated constructions
in safety, because its contents are absolute.

Putting \relative { } around \score or tagged constructions
containing non-absolute pitches is a recipe for obfuscation.
That's a criticism of the coder, not of LP's syntax.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Shane Brandes]

Frescobaldi has a neat little tool to convert from relative to
absolute pitch. I prefer relative as a pitch entry for ease and speed.
Kieran is correct relative pitch entry can cause headaches if you
start doing manipulations on stuff.

regards,
Shane

On Tue, Dec 12, 2017 at 5:48 PM, David Wright  wrote:
> On Tue 12 Dec 2017 at 20:32:34 (+0100), Simon Albrecht wrote:
>> On 12.12.2017 20:30, Shevek wrote:
>> >The alternative is to use \relative {} but to use \resetRelativeOctave
>> >religiously before every phrase, even the first one in a block. I find that
>> >more natural for composing, personally.
>>
>> I find that this has the disadvantage of not checking the notes
>> immediately preceding it for octave displacements.
>
> Well, nothing checks every note except checking every note.
>
> But for a composer in the heat of inspiration (!),
> \resetRelativeOctave might allow one to work on a later
> section even though previous sections have not yet been
> fully corrected for octavation.
>
> Cheers,
> David.
>
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Tue 12 Dec 2017 at 20:32:34 (+0100), Simon Albrecht wrote:
> On 12.12.2017 20:30, Shevek wrote:
> >The alternative is to use \relative {} but to use \resetRelativeOctave
> >religiously before every phrase, even the first one in a block. I find that
> >more natural for composing, personally.
> 
> I find that this has the disadvantage of not checking the notes
> immediately preceding it for octave displacements.

Well, nothing checks every note except checking every note.

But for a composer in the heat of inspiration (!),
\resetRelativeOctave might allow one to work on a later
section even though previous sections have not yet been
fully corrected for octavation.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Simon Albrecht]

On 12.12.2017 20:30, Shevek wrote:

The alternative is to use \relative {} but to use \resetRelativeOctave
religiously before every phrase, even the first one in a block. I find that
more natural for composing, personally.


I find that this has the disadvantage of not checking the notes 
immediately preceding it for octave displacements.


Best, Simon

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Shevek]

> But as a user of Lilypond for over fifteen years, I *will* recommend that
you consider avoiding relative entry mode — using absolute mode (and, when
appropriate, \fixed) will like save you headaches (like the one you're
encountering right now) in both the short and long term. 

The alternative is to use \relative {} but to use \resetRelativeOctave
religiously before every phrase, even the first one in a block. I find that
more natural for composing, personally.



--
Sent from: http://lilypond.1069038.n5.nabble.com/User-f3.html

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[Kieren MacMillan]

Hi Neo,

> I'm new to Lilypond, coming over from a commercial program.

Welcome!

> The issue:
> I seem to get measures auto-transposed one by one.
> I have one bar of music, then the second one gets transposed one octave 
> higher, the next one even one octave higher, and so on.
> 
> Without /relative command, the x5 chords get messed up.I use variables but I 
> get the same issue without them too.
> Minimal example auto-transpose-minimal.ly

I can't access your file for some reason…?

But as a user of Lilypond for over fifteen years, I *will* recommend that you 
consider avoiding relative entry mode — using absolute mode (and, when 
appropriate, \fixed) will like save you headaches (like the one you're 
encountering right now) in both the short and long term.

Hope this helps!
Kieren.


Kieren MacMillan, composer
‣ website: www.kierenmacmillan.info
‣ email: i...@kierenmacmillan.info


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Wright]

On Sun 10 Dec 2017 at 20:37:03 (+), Neo Anderson wrote:
> Hi,
> I'm new to Lilypond, coming over from a commercial program. Please, bear with 
> me as I'm still learning.
> The issue:
> I seem to get measures auto-transposed one by one.
> I have one bar of music, then the second one gets transposed one octave 
> higher, the next one even one octave higher, and so on.
>  
> Without /relative command, the x5 chords get messed up.I use variables but I 
> get the same issue without them too.
> Minimal example auto-transpose-minimal.ly

If you write

hita = \relative c { … … … }

then \hita will always yield the same notes. The \relative
construction produces an absolute expression.

Never use \relative { } round the structure that describes your score
because any change you make will change the pitch relationship between
the last note of one variable and the first of the next. LP can manage
it, as you have just shown, but our brains can't keep track.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Auto-transposition

[David Kastrup]

Neo Anderson  writes:

> I'm new to Lilypond, coming over from a commercial program. 
> Please, bear with me as I'm still learning.
>
> The issue:
>
> I seem to get measures auto-transposed one by one.  I have one bar of
> music, then the second one gets transposed one octave higher, the next
> one even one octave higher, and so on.

If you don't know what \relative does, you should look it up or don't
use it.

> Without /relative command, the x5 chords get messed up.

I have no idea what you mean by that.

> I use variables but I get the same issue without them too.

But you might understand more easily what happens when replacing the
variable references with the contained code.

You probably want to define your variables using

myname = \absolute { ... }

so that they don't become part of a surrounding \relative phrase.

> Minimal example
> auto-transpose-minimal.ly
>
>   
>*  auto-transpose-minimal.ly  
>
>  Shared with Dropbox  
> * 
>
> BTW, is it OK to link a minimal example or other LY files via Dropbox?

No.  We archive our mailing list discussions and such files will then
simply become inaccessible.  Dropbox may be appropriate for PDF files or
humongous problem sources which won't be really needed for understanding
the answers (if they are written well).

But minimal examples should be included in the mails to the list so that
they are around for people reading through the archives so that they can
follow the corresponding answers.

-- 
David Kastrup

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user