[graphviz-interest] viewport feature enanchement proposal

Hi,
given the fact that ps2 output is of very high quality, I'm finding that
pairing statically zoomed parts of a knowledge map with the entire map
itself can be very very valuable in (self-)learning context (which context
is not a learing context?).

It seems to me that a viewport 'mode', specified at graph level with, say,
viewport=node and at node level with viewport/viewported=true could be more
easy to use as the bounding box computation would be made by the code (eh
eh).

The viewport should include each viewport=true node and each node falling
in between.

It's not my fault if you have added world class desktop publishing features
to layout algoritms:

http://professosaurus.altervista.org/lezioni-2a-completo-000.pdf
http://professosaurus.altervista.org/pieno-expose_each_part_in_order-000.pdf
http://professosaurus.altervista.org/evoluzione-completo-000.pdf

The sense of this email, surely, is a big thank to you all, not a vanity
fair of my pdfs, of course.

Many thanks,
Federico

_______________________________________________
[email protected]
http://lists.research.att.com/mailman/listinfo/graphviz-interest