Maybe someone will correct me... but I think that the antTask from the SDK
doesn't cache do the iterative compilation, so that's why it may be slower.
This may not be the best technique, but I build with Ant in a way, I set it
as an alternative builder... so, it's the same as running the Ant script
from the command line basically. However, I don't think this will have any
effect on speed...
Generally, I don't think there is a way to make the compiler think that it
should cache data from one compilation to reuse it in another compilation,
if that's not the same project... But, again, I'm not really sure about it.
Well, if the compilation speed matters, I can only think of some general
advises, like, putting more code into SWCs, removing all [Embed] and
@Embed() tags from the code, avoiding MXML in general... Compiling separate
classes for testing rather than rebuilding the entire project (in fact, I
build the entire project only few times a day, the rest of the time I'd work
on a single class, or a small set of classes)...
