Hi all, Over the past ten days, there have been 7 new context releases, of which probably only three would have been necessary if there would not have been silly bugs embedded in the new code. In fact, we are still awaiting an eight upload (for that extra space after \quote and \quotation). When there is a bug found or a request made on the mailing list, usually Hans (and sometimes me) will post a bit of code that *should* remove a bug or implement a new feature. We both expect the original requestor to test that bit of code for side-effects (since we usually do not have enough time to do thorough testing ourselves). And that's where it goes wrong, of course. Beginners at ConTeXt are often incapable to test anything except the immediate intended effect (and not always even that). I wanted to bring this up because I hope that some of you can find a way to help testing those new bits (or find other ideas to solve the problem without placing too much of a burden on Hans' time). A possible side-effect could be the organic growth of a set of regression test files, an idea that has been talked about before. Greetings, Taco