Re: the question of triplets

[David Kastrup]

Hilary Snaden <address@hidden> writes:

> On 27/04/13 14:41, David Kastrup wrote:
>> Hilary Snaden<address@hidden>  writes:
>> \tuplet has a nicer way to deal with tupletSpannerDuration, by the way.
>>
>> So you can continue using it if you want to, but in a few years,
>> newcomers might no longer recognize what you are doing even though
>> LilyPond most likely will.
>>
>>> \tuplet y/x is more logical, and would have been a better way of doing
>>> it from the outset, but it's really not that difficult mentally to
>>> associate \times x/y with playing y notes in the time of x.
>>
>> It usually takes frequent use to arrive at that mental state, and there
>> are enough real problems with writing music that we don't need to create
>> artificial hurdles just to keep things interesting.
>
> There are 1224 \times in my own compositions, and 14656 in my
> engravings of other composers' work. I've no idea how many of those
> were copied and pasted, but nevertheless that probably counts as
> frequent use.

Sure.

>> That's why in this case we agreed on "disappearing" the "old" way of
>> writing tuplets from the LilyPond code base.  Sure, it will take a
>> few years to go down the pipes.
>
> I can probably cope with that timescale. I'll certainly take a look at
> \tuplet when 2.18 materialises.

convert-ly will make "taking a look" at it likely easier than you want
to.  But as the name of the function is quite different from \times and
its meaning straightforward, I doubt you'll get irritated enough to
implement a scheme for bypassing convert-ly 15880 times (the easiest way
would likely be tampering with convertrules.py before doing an upgrade
of all your scores).

-- 
David Kastrup