How to help with the development?

Any help with the following tasks is greatly appreciated (these issues are sorted by their importance, first is the most important):

• Help FOP development team with the development :-). Or find another open-source cost-free implementation of XSL-FO standard.
• Check if there is partial FO rendering possibility. Suppose you have source XML and a FO output ready. We provide the list of changed (or inserted) nodes in source XML, and the result must be a list of changed nodes (items, whatever) in FO output. We are using FOP as XSL-FO processor however we may switch to another if there is one.
• Investigate the possibility if some open-source XSLT processor is able to perform partial XML transforming only. Suppose you have source XML, XSLT script and a transformed result XML. We provide the list of changed (or inserted) nodes in source XML, and the task is to compute a list of changed nodes in result XML.
• Currently, EuroMath2 is capable of rendering MathML, but only as one image (GENE is able to transform MathML using JEuclid to Image, which is rendered by an Image renderer). This approach disallows backwards element identification. Hence, MathML renderer should be implemented, to render MathML into a tree of IFigures, and a MathML editor should be developed to allow WYSIWYG MathML modification (see the description of EuroMath2 editor/renderer plugin interfaces).
• Modify the Norman Walsh's DocBook XSLT scripts to make them able to be integrated into EuroMath - they must be capable of transferring gene:id attribute and gene-ref processing instruction.
• Investigate the possibility if some open-source XSLT processor (such as Xalan) can be modified somehow, so that there would be no need for special gene:id attribute (automatize the ID tracking system).
• Implement, steal, whatever the XHTML renderer (currently there is a stylesheet present in EuroMath2 able to transform XHTML to FO, but it ignores CSS).