Open heuer opened 12 years ago
Well, basically, the limitation to paged feeds is because paged feeds are what we want. :-)
Archived feeds are basically bogus, as far as I can tell. If you do things right you won't lose entries, even with paged feeds, so the whole thing seems completely pointless.
We could of course allow archived feeds but I can't really see any reason to. Even if your feed doesn't change, you can still use the "next" relation.
I'm happy to keep this the way it is.
We could of course allow archived feeds but I can't really see any reason to. Even if your feed doesn't change, you can still use the "next" relation.
Sure. My point was that the server don't have to provide "next" relation acc. to RFC 5005 if it uses Archived Feeds. Acc. to my understanding "next-archive" links would be sufficient. And therefore I wondered why a client shouldn't follow these links. The result would be the same. Why should standard-conform Archived Feeds forbidden implicitly? Why shouldn't the feed be more lenient and MUST contain either a "next" or "next-archive" link?
Anyway, this isn't an important issue I just wondered about the rationale.
Überpicky sidenote: """[...] each page MUST contain one link element with the "next" relation type [...]""" To which page should the last page point to? ;)
You're making sense, but it's simpler to have only "next". I think the support for archived feeds is mostly imaginary, since in practice I guess all SDShare implementations will be written to be just that.
Graham, what do you think?
Ooops. Missed the side note. Touché. Making a separate issue for it, fixing it later.
I'm happy just to stick to 'next' and pretend we dont know about these Archived Feeds smeeds.
5.1.5.:
Acc. to my understanding of RFC 5005 paged feeds don't need a link with "next" relation type. If an implementation uses "RFC 5005 -- 4. Archived Feeds", the relation types are called "prev-archive" and "next-archive".
What's the reason to limit pagination to "RFC 5005 -- 3. Paged Feeds"?