Lisa Dusseault wrote: > - response-format should be a comma-separated list of acceptable formats > for the status-response body. This feels a lot like Accept:...but, of course, we can't actually use Accept:, because that would affect what kind of responses come back on success. Do we want an Accept-Error:, maybe? As for the user-text option...is it really that important to be able to turn off the user text? In probably over 90% of the cases, you're going to want to be able to present an error message to the user (or else log something, if you're a robot), and that user text might have extra explanatory information that couldn't be expressed in the standard machine-readable part. -- /==============================================================\ |John Stracke | http://www.ecal.com |My opinions are my own.| |Chief Scientist |=============================================| |eCal Corp. |There are many intelligent species in the | |francis@ecal.com|cosmos. All are owned by cats. | \==============================================================/Received on Tuesday, 19 December 2000 16:34:48 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 12 October 2007 17:53:14 GMT