lilypond-auto
[Top][All Lists]
Advanced

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

Re: [Lilypond-auto] Issue 1377 in lilypond: Enhancement: make the last-p


From: lilypond
Subject: Re: [Lilypond-auto] Issue 1377 in lilypond: Enhancement: make the last-page spacing more consistent and less tight
Date: Mon, 27 May 2013 07:42:02 +0000


Comment #22 on issue 1377 by address@hidden: Enhancement: make the last-page spacing more consistent and less tight
http://code.google.com/p/lilypond/issues/detail?id=1377

Well, I was in the context of "optimal page breaking". I am not claiming that this patch in particular caused issue 3341. And indeed, it reverts cleanly (apart from a merge conflict in a comment which is irrelevant) and the reverted version still shows the same effect. So it is clear that this patch is not responsible for issue 3341.

But we still have a global page spacing engine that is basically out of control. We have no actual tests for checking that it does anything anywhere close to what it claims to do with regard to global optimization: all fixes to the algorithms have been to exclude excruciatingly bad decisions.

But we have no positive evidence that we don't just get a random (or even the worst) choice among tolerable choices. And the decisions are made independently from the actual spacing, based on a rectangular rather than a skyline model. And the rectangular model does not even share the same code with other positioning mechanisms, so discrepancies can (and will) occur.

We've had anecdotal reports that people continued using 2.12 (or 2.14) since the overall page breaking got worse since then. They may or may not be incorrect, but we don't really have tests that would give us sufficient confidence to state "we are pretty certain you are wrong about that".

--
You received this message because this project is configured to send all issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings



reply via email to

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