The main thing that caught me off guard was the lack of real consideration of the HTTP headers for determining the converter to be used. Though I may be incorrect in this belief, as I have not gone through the Restlet code base enough to know for sure that it is not being considered for the purposes of choosing the Representation type to be passed in.
> Wow, scoring converters is a delicate business! I understand a new approach > is in the works, though. > > I replace the out-of-the-box Jackson and Freemarker converters with my own > custom versions for other reasons, but a side benefit was that I can > prevent scoring surprises like the one you encountered. > > --tim > ------------------------------------------------------ http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=2938395
