You can extend Atom any way you want (with foreign namespaced elements), as long as the following conditions are met:
1- if all those elements are removed, you end up with a valid Atom document X
2- the meaning of the document X is not incompatible with the unabbreviated document
On 11 Nov 2004, at 12:13, Bill de h�ra wrote:
[...] I have no idea what is meant by 2.
The wording of this can be improved clearly. But let me start with an example. Take the document with the following sentence:
Doc 1. A. Jack has a house
You can add a sentences to Doc 1 to get Doc 2
A. Jack has a house B. Jack a du travail. C. Jack est mari�.
B and C are in French. When you understand French you understand something more than when you understand english. But they have not said anything incompatible with Doc 1. They have just added information.
B says: Jack has work and C says: Jack is married
You could have stripped Doc 2 of B and C and had still had something usable.
So the point is: whatever extensions you add to an Atom document, a parser that parses it as a pure Atom document cannot be misunderstanding the document. Ie: he is not liable to anything.
I am sure that with a little effort we can work out how to put this in one clear crisp sentence.
Henry Story wrote:
(2) need hardly be mentioned, because it will be assumed that your document needs to be parsable by simple Atom parsers, and so it would be foolish to try to say something that most parsers would misunderstand.
That seems much more simple and much clearer.
Anything more requires serious justification.
I see nothing here to back that claim up.
We are working on a standard where reaching consensus is the goal. Adding any type of limitation requires serious justification. I don't see any at all at present for the PaceExtendingAtom proposal. I think this is just a simple matter of methodology: don't close options you don't have to.
cheers Bill
