On Wed, Nov 16, 2005 at 03:07:20AM +0100, Daniel A. Nagy wrote:
As for the subject of our discussion, I think that we all agree that the spec for 0x80 should be stripped of "a note from one person to another..." bit., because one major implementation does not treat it that way.
The only disagreement seems to be whether "a note from one person to another" should be retained as an interoperable feature or should it be delegated to private notation namespace.
The disadvantage of the latter approach would be that various implementers would (possibly) implement this same semantics with a host of different notation names and won't interoperate.
Now, I can see that implementing the former using a type flag also causes problems. Maybe, it should be a common, ITEF-namespace notation? Or an entirely separate subpacket type akin to "reason for revocation"?
My view - it has to be outside the spec. "a note from one person to another" is a very high level thing and we'll never be able to nail down what it means. And if we do, we then have to add all the other variants like "a note from one person to two persons..." "a note from the program to a person..." "a note in XML..."
I am not sure. But in either case, as far as immediate modifications to the standard text are concerned, this "a note..." part should be removed from the definition of 0x80, because it means something that 0x80 definitely doesn't. Whether or not to add that text someplace else is an entirely different question.
Is this rough consensus?
PS: point of order here - as we are in last call, what are we "permitted" to change? Only completely wrong things? Or is this flexible?