On Mon, May 07, 2001 at 10:48:58AM -0700, Ian Clarke wrote: > On Mon, May 07, 2001 at 09:06:18AM -0500, Scott Gregory Miller wrote: > > > > > > If the CHK data contains the key for the metadata, then it won't coalesce > > > between two different sets of metadata anyways... > > > > Good point. Alright then, strict subset is the best way. For most > > purposes, Content-Type is sufficient. > > I am not sure what you mean.... my view is that all of this separating > out the metadata stuff is going to cause a lot of pain for questionable > benefits. I really don't think that making two files with different > metadata have the same CHK is worth the ugliness that would be fixing in > stone the metadata permitted with documents. The reason is that if we > don't allow metadata, client authors will be forced to place that > information in the document itself (to avoid the expense of requesting > an additional file), and we are back to the exact same situation.
That is what I meant. All I'm saying is that we should add a section in the metadata spec containing an optional strict subset of the spec (in other words, fully readable by clients that don't implement it) so that clients that want to try to make sure that CHK data coalesces as much as possible have the option of following a specification that will lead to bytewise identical metadata most of the time... -- 'DeCSS would be fine. Where is it?' 'Here,' Montag touched his head. 'Ah,' Granger smiled and nodded. Oskar Sandberg md98-osa at nada.kth.se _______________________________________________ Devl mailing list Devl at freenetproject.org http://lists.freenetproject.org/mailman/listinfo/devl
