Re: [GLISS] differentiating pre/post/neutral commands

[David Kastrup]

Graham Percival <address@hidden> writes:

> On Sat, Sep 01, 2012 at 09:18:17PM +0200, David Kastrup wrote:
>> Graham Percival <address@hidden> writes:
>> 
>> > Continuing to brainstorm on the problem of it not being obvious to
>> > which note a particular \command refers to, what if we used:
>> >
>> > \postfix:      c2 d\p  is unchanged
>> > /prefix: for music functions like   c2 /parenthesize d
>> > .neutral: for commands which aren't attached to notes, such
>> >   as .clef or .times.
>> 
>> It adds no new functionality and makes it harder to write and read
>> things.  How should LilyPond behave if you get your slashes wrong?
>
> Give an error?  The "new functionality" is "it becomes obvious
> whether a command applies to the note before or the note after".

That's what whitespace is quite good for.

> Since we wouldn't have any definition for /p, writing that would
> produce an error.
> (each prefix character would have its own namespace)

So LilyPond would pretend that it does not know what you are talking
about, instead of saying "I know this one, and this is the wrong place
to put it".

>> Frankly, the current syntax discussions are leading nowhere.
>> Brainstorming is fine, but pretty useless if there is no target or
>> topic other than "let's make things different".
>
> The target is "let's make it easier to learn/write/read .ly files".

.I -don't -see that +affixing .words .with .various .letters +indicating
.particular .categories -will +help -understanding .much.

> The meta-target is "after spending 5 years very publicly telling
> people *not* to talk about changing the syntax because we would do
> so 'in a year or two', I think I should encourage such
> discussions.".  I mean, people trusted me when I said that there
> would be a time for discussing syntax changes.  It was advertized
> on our "help us" page for something like 3 years until Janek
> commented it out in 884194 on 2012-02-14 because "noone knows when
> GLISS will happen".  If I didn't fight to give people an
> *opportunity* to have a meaningful *discussion* about syntax
> changes before we stabilize things, I would be abusing that trust.

Well, the music function work has removed a lot of pressure of the "I
want my favorite construct xxx in the grammar" kind since there are
quite a number of things people can put in themselves if they want to.
Yes, the process of making music functions more versatile was quite
undemocratic.  But it was not as much that I was sole decision-maker,
but rather that my skills and areas of interest were shaping the work I
was doing, with a focus of making currently existing constructs
reimplementable in Scheme.

The direction that LilyPond's parser is taking is increasingly
supporting _constructs_ as contrasted to _elements_.  So there is much
more inclination nowadays to ask oneself the question "how can I express
this with existing tools?" rather than "how can I fit this into the
parser as well?".

That does not mean that everything one had wished for can now become a
reality, but often the old wishes are so much more close to what is
already possible that the smaller differences are not much of a bother.

> I'm quite open to trying to find a way to structure these discussions
> so that people who are interested can participate, yet people who
> aren't interested don't need to worry because they'll get an
> opportunity to shoot down bad proposals before they're accepted.  But
> we've barely *started* having a discussion about changes, so I don't
> think that we can claim that they're going nowhere.

It is not as much "bad proposals" rather than "arbitrary" ones that I am
worried about.  Stuff that is no substantial improvement or
deterioration, yet causes a lot of energy and possible
incompatibilities.  We are not starting from scratch but an existing
language, and incompatibilities should not be gratuitous.

> At Waltrop, you only heard about one quarter of the ideas that
> Janek had.  I've got about half of the number of ideas that he
> had.  And that's just two people; who knows how many ideas our
> users have.  Or what about people like Francisco -- IIRC he's been
> teaching lilypond to composition students for the past 5 years.
> How many people per class... maybe 30?  So he might have taught
> 150 students how to use lilypond?  I'm sure that he has some
> thoughts about what his students found difficult to understand.
>
> Of course many of our ideas will not be good.  That's fine!
> That's how creative thinking works!  So let's create a place --
> either on this list, or a different list -- where people can talk
> about their ideas, point out flaws and make suggestions, look for
> common themes, etc., and see if there's anything worth adopting.

Locking people away in a syntax discussion list, then ignoring them or
having to deal with the consequences does not sound like a strategy fair
to either those interested in changes, nor to those not interested but
affected by such changes.

-- 
David Kastrup