Hello, I realise that what I have currently placed in the SVN repository for making the python harness tests work for both python2 and python3 has lead to some duplication. After some discussion with Attila off list we think we have a solution which satisfies both our concerns with suggested solutions up to now. The short answer is that as the harness test files only store data, then this could be shifted to data files rather than python source code and so escape the issues of python2 and python3 having different names for the types which are unicode aware. One such format might be XML, however XML I appreciate is verbose and so I might look at other formats such as JSON, although XML might be the most flexible. Any thoughts from others on this? I know it still leaves the doctests as an issue to solve. Also would it be appreciated to have such work done before a release on 1 June? If so then let me know as I would need to do some keyboard bashing over the weekend as next week will not be so convenient for me to do code changes as I will be travelling. Michael WWapplesFor a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com