Frank slipped these into his advertising splash, I see:
> - allows for changing the position algorithm
> - will soon do spanning floats
Which prompts me to publicly say:
I forgive Sebastian Rahtz!! (Well for one thing, anyway
... there's still hyperref:-) )
for never having supplied us humble coders with what he claims is `the
obvious algorithm for float positioning (including spanning floats)'.
Having tried quite hard over the last few weeks to look inside his
mind and reconstruct it, I now realise that it does require a somewhat
weird intellect (something Dr Rahtz refers to in a phrase involving
`planets', `brains' and `like you ****ing mathematicians') to find an
algorithm that is time- and space-efficient but also gets a reasonably
good result most of the time (and maybe even warns you when it is not
doing so well).
The reason why I am going pubic in this particular forum is to ask for
Not necessarily help in designing this wonderful algorithm but in
formulating, or finding in the literature, the rules and heuristics for
judging whether a float placement is acceptable and/or good.
Frank has already mentioned that the rules of classic LaTeX can
perhaps be relaxed somewhat:
> - handle different float strategies, eg float after callout, float after or
> in the same column, float on the same page
but these are only suggestions and the need to deal with spreads
and with spanning floats will certainly call for further strategies.
Thanks in advance, and please pass this request on to anyone who is
interested or has useful knowledge.