Bjoern Hoehrmann wrote:
>[snip]
> Perhaps you might to say something like, when used in Atom documents,
> the security considerations for handling links in Atom documents apply.
>
Yeah, this would be better. This draft doesn't introduce any security
concerns of it's own.
> I would personally like to see some note that implementations cannot
> necessarily trust that the publisher has the right to license material
> claimed to be covered by the license, and that care should be taken
> when making decisions based on the license reference, such as
> republishing the content.
>
Good catch.
Another concern that was raised to me by a colleague is that the license
resource being pointed to could change over time, meaning that the
license being referenced today may not be the same license being used
tomorrow even tho URIs may be exactly the same. If the license is
controlled by a different entity than the publisher of the entry, this
could cause problems.
> The third paragraph in section 3 seems overly verbose to me, is there
> no terminology introduced in RFC 4287 that could be used and/or re-
> ferenced instead of the verbose discussion? Or is this assumed to be
> the case already, and the "Implementors should note" non-normative
> note just re-states what is stated elsewhere in the draft or the Atom
> specification?
>
Unfortunately, there is no language in RFC4287 I can draw upon here.
> I do not quite understand feed-level licenses, the draft just says
> what they don't cover, not what they do cover. Say I make a feed with
> the five most insightful blog postings on international politics and
> I license the feed under the most permissive license possible. Can
> you then copy my list of entries over to your top five list? Just the
> IRIs, or also the summary I wrote for those entries? What if I copied
> the summary over from the original feed (assuming, e.g., I may copy
> those, but you may not, for some twisted legal reason)? If I relate
> the license to the entries, would the license cover my summary and
> the third party entries, or just my summary, or just the content of
> the third party entry? Should atom:source be used in such scenarios
> specifically for license reasons?
>
Yes, this should be clarified. Feed level licenses cover the metadata
of the feed (title, subtitle, etc). I've actually found very little use
for feed level licenses and would actually be quite happy to remove them
from the spec completely.
> I think the concept of "informational content" used to explain some
> of these things is too unclear to me to answer these questions.
>
Ok
> I think the reference to "copyright licenses" is a bit unfortunate.
> Why is this reference to "copyright" necessary?
Can you suggest a better term?
- James
