Hello Jaako, Thank you for your reply. I think it is useful for users to be described this behavior in the specification clearly, even if it is implementation-dependent. I hope this behavior is described in the recommendation too. Regards, Keisuke Tamiya On Mon, 16 Feb 2009 15:51:39 +0200 Jaakko Kangasharju <jkangash@cc.hut.fi> wrote: > Text relating to this case when considering potential future final EXI > versions was indeed missing from the format specification. Currently, > the specification requires an EXI processor to process any final > version that it understands. For preview versions, the specification > already says that the behavior is implementation-dependent. > > As the working group cannot know what potential final EXI versions > will look like in the future, we have decided not to constrain the > processors in any manner in this case. We will therefore add text to > the specification stating that the behavior on seeing a version number > unknown to the processor (either preview or final) depends on the > implementation, and possible behaviors include rejecting the stream.Received on Friday, 27 February 2009 01:24:55 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 February 2009 01:24:56 GMT