Draft: draft-snell-atompub-feed-license-07.txt Reviewer: Robert Sparks [rjsparks@estacado.net] Review Date: Monday 8/28/2006 2:13 PM CST IETF LC Date: 9/11/2006 Summary: This is an AD-sponsored individual submission for consideration as an Experimental RFC. The draft has a few open issues (mostly around clarity) that should be resolved before publication. Overall, I found this document confusing. I had to go through it several times before I felt like I knew what it was trying to accomplish. With that done, I can't figure out why it's trying to do what it's doing. -------- Issue 1: It's not clear why this is Experimental from the current document text (research, limited applicability,...?). Would an applicability statement make sense? -------- Issue 2: What's covered by the license claim is still hard to pull out of the document - the paragraphs added for version 07 help. The following paragraph starts to capture the intent (I think), but the document would benefit from making this a more immediate and central point: > Because entries contained within a feed may originate from other > sources, "license" link relations appearing within a feed apply to > the metadata of the containing feed element only and do not extend > over the metadata or content of the contained entries. I read this to say that the license covers only those bits inside the element that contains the license link, but no bits inside any elements that element might contain. This is the real requirement and should be stated normatively. The following paragraphs are justification for the requirement. The point would be better made if they didn't have the normative text they currently have. (Unless I've missed the point off the document). ------ Issue 3: It would help a lot to describe (even if that means speculate) what a person or automaton will do when consuming this new data. I'm having a hard time guessing what this would be used for (assuming I've understood the requirement as stated in issue 2 correctly). ------ Issue 4: This example is confusing: > Multiple "license" link relations specifying different href > attribute > values are are considered to be mutually exclusive > alternatives. For > instance, if an entry specifies both a Creative Commons License and > the General Public License (GPL), the entry is considered to be > licensed as either Creative Commons OR GPL as opposed to Creative > Commons AND GPL. I think you're trying to say that the format will allow your meta-data to be dual- or multi-licensed and the example is showing how it can express that. If I'm correct, saying it that way might reach a wider audience. Finding a concrete example of something you might dual-license in such a way would help. I'm having a hard time coming up with something that would appear with the dual-license structure you show as an example here. (btw - if the text above survives, there is a duplicated "are" in the first sentence)