Re: [oll-user] Finally, the revised instructions - please review!

[Marc S. <ma...@ou...>]

On 1/29/2014 2:20 AM, Janek Warchoł wrote:
> 2014-01-29 Marc Sabatella <ma...@ou... 
> <mailto:ma...@ou...>>
>
>     Most of the below is meta-discussion not really relevant to the
>     challenges, but still, I think, interesting to think about, as it
>     illustrates some of the difficulties inherent in making meaningful
>     comparisons.
>
>
> Yes, this is a valuable discussion!

Thanks for taking my comments in the spirit in which they were 
intended.  I just have "one" thing to add:

> You're right, i'm sorry!  Obviously, i'm biased towards LilyPond 
> here.  I'll try to be more careful.

No apology necessary!  There's no shame, of course, in having a 
hypothesis going in to an experiment.  And don't get me wrong, either.  
I might be a MuseScore fanboy in some respects, but I have no illusions 
here.  I am quite sure that overall, for the sort of tasks these 
challenges are setting forth (taking existing music and creating a new 
edition of it), LilyPond as well other similar well-implemented 
command-line programs (I'm partial to abcm2ps, and I hope someone 
volunteers to champion it here!) are going to do very well in comparison 
to most WYSIWYG programs.  Not that having a command line versus WYSIWYG 
interface makes an inherent difference in how one implements one's 
layout algorithms, but I do think there is likely to be correlation in 
philosophies and this is going to affect where one spends one's energies 
in implementation.  That is, the developers of a WYSIWYG program are 
likely to have priorities *other* than layout, and it's likely layout 
will suffer for this. So I would say I share your hypothesis here.  But 
perhaps not the value judgement you inadvertently associated with it.

In my mind there are really multiple sets of tradeoffs involved when 
comparing notation programs.  One is the tradeoff we are focusing on 
here: between having fewer manual adjustments necessary versus having 
those manual adjustments easier to perform.  Also, by incorporating 
version control - which I think is a good thing, because it *is* 
important in a collaborative project - the challenges also touch on 
possible tradeoffs between having text-based formats versus binary 
formats.  Except of course, if there *are* advantages to binary, nothing 
in these challenges is likely to expose it.

But there are other tradeoffs that might be relevant in other sorts of 
comparisons.  And the ones that interest me most come into play when 
using a notation program as a compositional tool as opposed to 
reproducing existing music.  My hypothesis is that in the same sense 
that command-line programs will tend to have the upper hand in 
layout-based comparisons, WYSIWYG programs will tend to have the upper 
hand in compositional-process-based comparisons. Unfortunately, the 
compositional-process comparisons would, I think, be much, much, harder 
to perform in even the most minimally objective sense than layout 
comparisons.

Still, the next interesting question to me is, to what extent are the 
layout advantages of one given program offset by the 
compositional-process advantages of another?  And, recognizing that 
layout is, in the end at least, also potentially an important part of 
the compositional process, to what extent are further improvements in 
layout for a given program more important than further improvements 
elsewhere in the compositional process, even for someone more focused on 
composition than on reproduction?

These are the issues that MuseScore has been grappling with and that 
will increase in importance going forward.  So to me, it is interesting 
to get a sense for how *big* the gap in layout facilities is between 
MuseScore and other programs including LilyPond.  Perhaps even more 
interesting is the matter of identifying, if possible, the biggest 
bang-for-the-buck areas for improvement.  But the goal I would have in 
mind wouldn't be to beat LilyPond at its own game - just to get a better 
sense of where MuseScore stands with respect to the tradeoff between 
layout and compositional process, to help MuseScore improve *its* own game.

Marc

-------------- next part --------------
An HTML attachment was scrubbed...