denemo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Denemo-devel] Documentation of Scheme interface


From: Richard Shann
Subject: Re: [Denemo-devel] Documentation of Scheme interface
Date: Sat, 04 Oct 2008 20:45:29 +0100

On Sat, 2008-10-04 at 12:34 -0500, Jeremiah Benham wrote:
> On Sat, 2008-10-04 at 17:00 +0100, Richard Shann wrote:
> > On Sat, 2008-10-04 at 07:41 -0500, Jeremiah Benham wrote:
> > > 
> > > You have been making many references about C and lack of object
> > > orientation. Have you ever used Objective-C. My iphone has an
> > > Objective-C available for it. I was tempted to install it and play
> > > around with it during some commutes to work. I was wondering if we
> > > could
> > > use that to convert some structures to Objective-C classes or
> > > something. 
> > 
> > It's not C that is to blame (witness gtk as an object oriented system
> > written in C). And Denemo is not devoid of object orientation, its just
> > that there are some of the really nitty gritty things which I find
> > inpenetrable and would have benefitted from more OO design. 
> 
> Do you know of any good tutorials and or books on OO design?
I don't really - in fact, I am not at all sure that I would be good at
designing something from the bottom up. It needs sustained work and I
only get half hours normally. I think that when we have finished pushing
the current Denemo program as far as it will go someone will take it as
a basis for writing such a program from scratch - we have a bunch of
good ideas here. 
>  I did not
> get this from K and R.
> 
> > We just have
> > to write more functions like find_prevailing_clef() and retire stuff
> > that relies on side-effects of drawing the music.
> > 
> > I hope I haven't sounded too pessimistic a note - I am genuinely
> > surprised by the possibilities opened up by attaching a scripting engine
> > to the Denemo command set, it is very exciting. I wonder when we should
> > release though - early gives people a chance to join in more easily, but
> > might put people off because of snags we haven't ironed out - later
> > makes folks wait.
> 
> I think early and often is the best philosophy.
I agree - there is a fatal attraction to delaying while just one more
improvement is added, I feel that attraction, but we need to fight it.
We have a significant enhancement to release, scripting, so let's do it.
If we create a branch for 0.8.0 and look for showstopping bugs in that,
meanwhile change the version number in the master branch to be 0.8.1
(using the convention that odd numbers are for unstable) and start
working towards 0.8.2
What could be in the 0.8.2?
        1) Easier scripting (no abort on syntax errors)
        2) Being able to delete unwanted menu items.
        3) Rationalization of the menus (using 2. above we could retire some
denemo commands and make standard some scripts as menu items to replace
them)

Another stream of work would be the midi ideas, hopefully someone will
offer to help with that before too long.

Richard






reply via email to

[Prev in Thread] Current Thread [Next in Thread]