by the way, I agree with your approach of separating the serializers
from the objects they serialize. Long experience has taught me that
giving objects serializer member functions, although superficially
attractive from an object-oriented perspective, causes problems of its
own. Having the serializers separate means:
i) the objects don't have to be extensible to support the addition of
extra serialization methods
ii) tends to lead to better documentation of the serialized form of the object
iii) funnily enough also tends to better object encapsulation, since
the serializers cannot access the private members of the object and so
cannot inadvertently expose the internal structure of the object in
the serialized form
iv) tends to result in using a standard rather than something
proprietary for the serialized form of the object
Martin
2008/8/19 cd...@peermore.com <chris...@gmail.com>: