If I may add: it's worth to try the shell version (where you would have to replace "." by "->", that is, for instance, LP->MAXIMAL_VALUE) since there you have tab-completion to tell you the properties available, even with the correct spelling.
The confusion comes from mixing two (very different) kinds of objects. Things like Polytope<Rational>, SimplicialComplex, Graph, ... are big objects which live on the Perl side with properties and rules and such. But there are also small objects which live on the C++ side, e.g., Matrix<Rational>, Ve...
Actually, -O1 is quite harsh (in terms of performance). Another option (if performance does matter) build a vanilla gcc 4.3 or 4.4 from scratch and use that one.
As it seems there is something terribly wrong with the installation at TU Berlin. Similar problems from there have been reported before. The current version 2.9.9 is extremely simple to install, even locally. You should try to get a correct installation.
It's true, currently the jreality interface seems to treat FacetTransparency as a switch. Facet transparency is switched on no matter what the value is. It does work with the javaview interface though. A value of '0.0' means no transparency, and '1.0' is like clear glass. So a typical (working) exam...
There is a new effort (from the Berlin group) to improve polymake's documentation. This is very welcome! Here are some comments in sequential order on the wiki page at http://www.polymake.org/doku.php/tutorial/introduction : Basic constructions: Actually, it is sort of a bad idea to use non-rational...