* Eric Scheid <[EMAIL PROTECTED]> [2005-10-21 17:05]: > It was thought that the 'self' link in an archive would point > to the archive document itself, which meant a different > relationship was needed for the purpose of locating the URI > which is the one that contains the most recent updates/entries.
I admit I intentionally kept out of that discussion because I don’t have a lot of personal interest in the subject, and the thread has been so fast and furious I didn’t bother staying on top of it just because. But I’d like to offer some thoughts on this particular point; apologies in advance if these points have already been raised: First, rel="self" is going to be implemented by most everything that groks Atom 1.0 in order to support one-click subscription, if applicable, right? Whereas this new relationship might not find such wide-spread support. Second, although less important – even applications which do support rel="subscribe" will have to implement a fallback behaviour. Arguably, they already do in some fashion, because a feed may omit the rel="self" link, so I don’t know if this is such a pressing issue. For these two (similar) reasons I think it might be wise to keep rel="self" in the role that this new rel="subscribe" thing is supposed to fulfill, and invent a new relationship that can point to the canonical location of the archive feed document. Third, and in some ways most importantly, I see overlap with Bob Wyman’s “preferred feed” agenda from way back when: http://www.imc.org/atom-syntax/mail-archive/msg15001.html Does anyone remember that? Maybe it would be wise to opt for something slightly more generic that can express semantics useful for feed scenarios other than archives? Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/>
