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

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

On 1/26/2014 5:35 PM, Janek Warchoł wrote:
> I agree with you that a good engraving program should not need manual 
> breaking.

This makes no sense to me.  I thought I gave a pretty clear explanation 
of why manual line break decisions are inherently subjective and why 
good human engravers often *need* to take responsibility for these 
according to the specific requirements of the situation.  I suppose in 
some contexts it might be OK to just let line breaks fall where the 
program defaults say they should, and of course in those case, it is 
reasonable to expect that the results would look "good" in some sort of 
general sense.  But I don't find merely looking vaguely "good" 
acceptable for my own work, nor do the publishers I have worked with 
accept that.  Both I and the publishers I have worked with have pretty 
specific - and mutually exclusive - ideas over how we want our line 
breaks to fall.

For instance, there are two different publishers I have worked with 
where one can find examples of the exact same piece of music engraved by 
both of them, but they have different line breaks.  This difference is 
not because of any accident in how their software laid things out - it's 
because the two publishers have conflicting house standards governing 
these things.  Any software that happened to produce results that fit 
one publisher's rules would automatically fail to meet the other's.  In 
fact, one of these publishers has different set of standards that apply 
to one series of books versus another series - I imagine that's actually 
very common in the publishing world.  So regardless of what software one 
chooses, one *will* often need to overrule default line breaks if one is 
trying to meet some particular set of standards (one's own, or those of 
the publisher one is working for).

> I was thinking about asking participants to prepare two versions of 
> the engraving - one with default line breaking of their program, and 
> one with the line breaking the same as in the original engraving. 
> However, as Marc wrote, this would be very inconvenient, as some 
> participants would have to make all adjustments in two versions (well, 
> this actually demonstrates that these programs have a serious weakness).

Again, I don't really follow.  In the examples I gave, it isn't anything 
about the program that makes it necessary to perform manual adjustments 
twice - it's inherent in the nature of the types of adjustments that are 
affected.  That is to say, command line programs would be just as 
affected as any WYSIWYG program (if not more so) - assuming they need 
manual adjustments at all, of course. So indeed, a program that requires 
fewer adjustments has the advantage.  But assuming a program requires N 
adjustments, performing those adjustments for two different sets of line 
breaks means potentially N*2 adjustments.

> I think that we can judge the line-breaking aspect well enough even 
> before the engraving is beautified. I.e. if the line-breaks after note 
> entry are not the same as line-breaks of the original, then we should 
> compare them at this stage, draw conclusions about the software, and 
> then force line-breaks and continue working on other aspects.

Yes, I agree here.

> Oh, and by the way: i think that "adjusting global layout settings" 
> should be moved before adjusting line breaks - it wouldn't make sense 
> otherwise.

Not necessarily.  In many contexts - such these challenges where we are 
explicitly supposed to copy line breaks - what may make most sense is to 
first decide on the desired line breaks, then choose global layout 
settings that achieve the optimum balance between readability and use of 
space within that constraint.  And in my experience, there is usually 
some give and take here.  That is, one chooses the line breaks one 
thinks one wants, one chooses global settings to produce optimal results 
within those constraints, then one re-evaluates one's line break 
decision based on whether the optimal results for those constraints are 
reasonable or not.  Again, this is not a software-specific observation.  
It applies equally well to WYSIWYG and command-line programs as well as 
hand engraving.

Of course, for pretty much any software, once one arrives at the line 
breaks and global layout settings one desires, it should be perfectly 
simple go back to the defaults and then actually apply the changes in 
either order.

Marc