[texhax] using larger units to determine breaks?

[Morten Høgholm]

On Wed, May 6, 2009 at 2:53 PM, William Adams wrote:
> On May 6, 2009, at 8:06 AM, Morten Høgholm wrote:
>
>> There is of course Michael Plass' thesis on the matter "Optimal
>> pagination techniques for automatic typesetting systems"
>
> Gotta love the beginning of the last sentence of the abstract, ``For
> certain simple badness functions, the pagination problem is NP-
> complete...''
>
> And for complex badness functions? What's more complex than
> Nondetermininstic polynomial time?

If you have a more complex badness function you may wind up with fewer
possibilities, non? But I must admit I have not read that thesis.

>> and also
>> Stefan Wohlfeil's "On the pagination of complex documents".
>
> Available as a .pdf!
>
> http://www.pi6.fernuni-hagen.de/publ/tr234.pdf

Sorry about that - that is just a cut-down article. The thesis itself
is considerably longer and bears the title "On the Pagination of
Complex, Book-Like Documents"

>
> Ironic though that there's what I'd consider a bad page break on pg. 7
> (the numbered list at the top of pg. 8 has lost its introductory
> paragraph which is at the bottom of the preceding page).

Yes, I never liked those ones either. Happens frequently though due to
the penalty settings (in article.cls, this \@beginparpenalty is -51,
so a break is encouraged).

>> The latter
>> came with a program - XFORMATTER I believe - to make the pagination
>> choices.
>
>
> I didn't see a place to get this program from.

Unfortunately, I don't know if it is available anywhere.

-- 
Morten