I tend to add bits and pieces to test specific parts of the report that I am working on at the time such as SVG graphic, Pedigree Chart, adoptions or Google Maps for example. It is not exactly well structured and certainly not documented, so it would be difficult for anyone else to know which particular bit to look for. In addition, some of it is based on my own family tree, which I am not ready to release at present.

For grammar and translation issues, I find it better to use a real tree that I know well, e.g. my own family tree, so that in reviewing the report it is easier to spot sentences that are in accurate or wrongly worded, because you already know what the result should be.

But I agree it would be nice to have a common set of test genomaps, but they would need to be designed with translation in mind and be very well documented so that the results can be checked against the expected outcome. Each language seems to have its own difficulties in translation, such as gender of places etc. , inflections, possessive nouns and so on. These might need their own specific test data.

Is there anyone out there prepared to take up the challenge?