Re: [Rosegarden-user] insert rest in inside segment

[D. M. M. <mic...@ro...>]

On Thursday 23 July 2009, Dan Muresan wrote:

> Not that I want to push the macro idea, but since this should not have
> appeared so complicated, let me explain.

It appears complicated to me because I have no clue off-hand how to go about 
recording this event buffer.  I'd have to go on a shopping expedition and pore 
over API docs looking for ideas and inspiration.

The other idea is something where I understand the gist of how it should all 
be strung together out of off-the-shelf components, and it's only a question 
of details.  Some big details, and some big fiddling, but I'd know exactly 
where to begin.  It makes a difference.

> I understand. But even just enabling the user to enter melody lines
> (without chords) using a MIDI keyboard would be a huge improvement
> over mouse input.

There's step recording too, incidentally.

But still, yes, I get where you're going with all this, and I'm largely 
agreeing with the underlying ideas here.  Which is not the same as saying 
"sure, I'll code that right away."  Honestly, I'm not sure where I'd begin on 
such a beastie, because this is very firmly in Chris's usual territory.  He's 
the closest we have to a dedicate music math hacker, and he's on vacation at 
the moment.

> Of course :) Maybe I should enter some of these things in the
> Sourceforge Tracker / feature requests.

Indeed, and I hope I don't have to put them at "send us the patch" priority 
five years from now.  I really do hope that.  Right now, with so many more 
mundane things to grind through standing in the way of that release, this 
really isn't the time for big thinking stuff, but we have a tendency as a 
project to never get around to the big thinking stuff for one reason or 
another, and this is one areas where I'd like to hope the release after next 
might finally be able to address some of these niggling problems.

Up to now, "not impossible" has been the bar, and a lot of things are "not 
impossible" but an incredible pain in the ass to accomplish.  I want to raise 
that bar badly, but my work in that direction got interrupted by having to 
waste what will have been close to two years on this Qt4 port.

Some great things are coming out of this port, and it's more than a mere 
exercise in moving sideways, but we are not moving very far forward in 
proportion to the number of hours we've had to burn to get there.  This has 
been a very expensive endeavor.  Very expensive.

I'm glad you're taking all of this with a good attitude.  I know it can be 
very frustrating when we have to say "yes, these are great ideas, but hurry up 
and wait."  Unfortunately, that's all we can do at the moment.  I really can't 
get sidetracked just now, or an October release is completely impossible, 
instead of merely improbable.

I think on balance though, Rosegarden should start to make a lot of real 
progress on these issues in 2010 and beyond.  I have traditionally tried to 
avoid as much development work as possible, and do anything else, but this 
port has forced me to just bulldoze my way through the obstacles, and get it 
done, because nobody else will.  I'm a vastly better developer than I was a 
year ago, and I'm going to be taking on more and more challenging problems in 
the years to come.

In the meantime, we appreciate your patience.
-- 
D. Michael McIntyre