On 21.11.2008 (14:34), Bertalan Fodor wrote:
> Also there is a quick insert mode work in progress for jEdit /
> LilyPondTool.
That's the second reason why I found out I had to make something myself
(third, actually: after my sound contempt for emacs, and the desire to do
some learning-by-doing): I had high hopes for the LPT's quick insert mode,
but in the current version, it didn't do much for me (that's also why I
nagged you for an update in the other thread :)
> However, my quick insert mode will be really a virtual piano made from the
> keyboard. I found that approach much better. It will sport automatic
> accidental handling. For example if you want to write
>
> c es c bes, b c
>
> in quick lily you type:
> d g e d l , e d
>
> In LilyPondTool you'll be able to type:
>
> c g c s x c (try it!)
Hm... where in that sequence is the ',' information? If 's' is 'bes,', why
then is'nt 'x' = 'b,'?
As far as I understand your layout, the two bottom rows mimic a piano
keyboard, like so:
-------------------------------------------------------------------------
| | | | | | | | | | | |
| Q | W | E | R | T | Y | U | I | O | P | | |
---------------------------------------------------------------------------
| as | bes | | cis | dis | | fis | gis | | | | |
| A | S | D | F | G | H | J | K | L | | | |
-------------------------------------------------------------------------
| a | b | c | d | e | f | g | | | | |
| Z | X | C | V | B | N | M | | | | |
-------------------------------------------------------------------
That, to me, has the great disadvantage of moving too far away from the
ergonomics of the computer keyboard, i.e. disregarding the difference
between a piano keyboard and a computer keyboard. With this layout, (a) the
home row is taken up by little-used accidentals, and (b) one function ("add
a note") will either have to be done with two hands, or by letting one hand
jump all over the keyboard. That's why I went for the home-row based
layout with the pitches in one hand and the rhythms in the other:
-------------------------------------------------------------------------
| s | g | a | b |times| | | ' |16/64|32/128 | |
| Q | W | E | R | T | Y | U | I | O | P | | |
---------------------------------------------------------------------------
| c | d | e | f | r/R | 1 | 2 | 4 | 8 | | | |
| A | S | D | F | G | H | J | K | L | | | |
-------------------------------------------------------------------------
|undo | del |flat |sharp|breve| dot | , | | | | |
| Z | X | C | V | B | N | M | | | | |
-------------------------------------------------------------------
In practice, I find this much quicker, even though there are a few extra
key strokes. I would type
a dc a rmc rv a
> Much more natural. However, it makes it complicated, because the software
> must know the key. So for example is c minor it will be rendered as
> c es c bes, b c
> but in h major, it will be rendered as
> c dis c ais, b c
I assume/home you will also have a key-agnostic mode, e.g. for transcribing
music; I wouldn't want to have to think "which key am I in?" all the time,
if all I want is to enter a 'g'.
In practice, at least for most of the music I type, the "remember the
accidental modification" model does more or less what you wish (after the
initial switch from 'b' to 'bes', every 'r' press will give a 'bes'), but
with greater simplicity and -- I think -- flexibility.
I like the idea of setting the key, though. It could either be done
explicitly, or by reading back to the previous \key command if one exists.
I think I'm going to shamelessly steal that idea...
Eyolf
--
Neurotics build castles in the sky,
Psychotics live in them,
And psychiatrists collect the rent.
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user
