improving LilyPond useability (was: Supporting my work on LilyPond financially)

[Janek Warchoł]

Hi all,

this is quite a different subject from the "promoting LilyPond" stuff,
so i separated this thread.

2013/12/1 David Kastrup <address@hidden>:
> Kieren wrote:
>> Result? Not a single successful convert [to Lily] to date.
>
> I think Frescobaldi with its templates would likely be helpful.
> Possibly also Denemo.  Staring at an empty canvas without any controls
> is a bit disconcerting.  Basically you need to have a printout with the
> basics at hand.  How many pages is our tutorial?

Learning Manual is 200 pages.  10 times too long - noone except the
most nerdy people would read it (no surprise that i'm using Lily - i'm
a nerd ;P).  Even the "Tutorial" part of it is way too long (20 pages
just to get the program running and another 20 pages to get very basic
notation!!).

I've created a "Quick-start" tutorial some time ago - my choir
colleagues used it when crowd-typesetting "Dixit Dominus".  It's only
6 pages long and covers nearly all basic notation elements than a
beginner would need - but it's not just a cheat-sheet: it introduces
and teaches how to use Lily.  Add to that 3 pages explaining how to
write basic structure and we'd have something that gives an easy (but
complete enough) introduction to LilyPond in half an hour (as opposed
to 2 days of reading and heavy thinking for the Learning manual).

I'd be more than happy to share this tutorial and translate it, but i
don't have time to lead an effort to incorporate it in our docs.  So,
if someone wants to take responsibility for this, i'll help, but
without support this will not work out!


2013/12/1 David Kastrup <address@hidden>:
> Henning Hraban Ramm <address@hidden> writes:
>
>> Am 2013-12-01 um 19:15 schrieb David Kastrup <address@hidden>:
>>> I'm always a bit surprised about the low resonance on features like
>>> <URL:http://code.google.com/p/lilypond/issues/detail?id=3648>
>>> Issue 3648: Patch: Isolated durations in music sequences now stand for
>>> unpitched notes
>>
>> I hear you - as a magazine layouter I seldom get feedback at all, and
>> then mostly some nitpicking of the authors.
>>
>> Hey, isolated durations are GREAT! I can remember some pieces where
>> they would have been very handy.
>
> Well, the main reason I'm surprised is that a few years ago there were
> proposals about it and I said "this will have to wait until some other
> parser parts are where they need to be" and there was wailing and
> gnashing of teeth.

Well, i think i know why there is so little resonance: too little
advertising.  You put the patch in the tacker - that's where patches
should be added, but it doesn't make it very visible: issue tracker is
not a newsreader.  I barely manage to look at patches that hit the
countdown - not always, in fact - and i suppose that there are not
many people that regularly swoop the tracker to see what's going on.
And i don't remember this patch being announced in some special way.

What could you do to make sure that outstanding improvements are
noticed?  Good question.  For example, write a post on the blog (or at
least -user) saying "A long-awaited feature is finally implemented",
mention people who requested it (with links), show all things that
will be possible thanks to this patch, throw in a custom-built binary
containing the patch so that power users could test it before it's
actually merged into master.  I bet 10 Euro that you'd get 3 times
more publicity if you do this :-)

(If i win the bet, i'll donate the money to Lily development!)


2013/12/1 David Kastrup <address@hidden>:
>>> If you're looking at a real-world score's input file it's
>>> overwhelmingly daunting.
>>
>> …even for me, and I’m one of Lily’s biggest users in terms of number
>> and size and “real-ness” of scores.
>
> Well, we'll probably need some open discussion of common problems and
> imaginary input that would make it considerably easier for people to
> overcome them.

Well, that's what i'm trying to do for very long time: identify common
problems (for example in articles on LilyPond blog, and in the
analysis that was published in the LilyPond Report a long time ago).

We tried this a year ago during GLISS - it didn't quite work out, but
maybe we could try again.

Anyway, we could do a poll about this.

best,
Janek

PS

2013/12/1 Richard Shann <address@hidden>:
> But the feedback I get about Denemo is almost entirely positive - those
> who find it unusable just quietly switch to something else, out of
> politeness I guess. Most unhelpful!

I've been guilty of this - I'm sorry!  It's been a long time since i
tried Denemo, and i no longer remember what the problem was.  If i get
some time to try again, i'll report.