> As I said before, if the WG can reach consensus, I'm happy > with any old term. I hadn't seen Mark's proposal till a few > days ago, and a mention in an xml.com does not, in my > opinion, a spec-in-stone make. > My only pushback on "next" is that to me, it seems counterintuititive > -- same as your pushback on "prev." *shrug* > > The SixApart people have publicly pointed to FH, so I don't > think they're particularly fussed about any particular > approach other (not to put words in their mouth). I wasn't > able to find a TP feed that uses rel="next"; do you have a > link to one?
I have been holding back until I found the right message to reply to. I think I found it. As it turns out Six Apart's Japanese office recently released something we call "TypePad Mobile" and "TypeCast." Both products are geared at the same problem: making blogs accessible to mobile devices. TypeCast is about making an Atom powered blogging application, whereas TypePad Mobile's purpose is to make all published blogs completely accessible via Atom. It may be the most comprehensive Atom implementation of its kind, but that is a bold statement to make. (FYI: Tatsuhiko Miyagawa was the lead engineer behind the project.) In TypePad Mobile, Tatsuhiko implemented a rel=next and rel=prev in order to facilitate the following problem: * in browsing a blog on a phone, one can only realistically afford to read 1-3 entries at a time. The implementation addresses the need to: ** veiwing the first n posts on the front door, and then paging through entries that go back in time ** viewing category archives (1 or 2 entries at a time) ** viewing daily/weekly/monthly/yearly archives (1 or 2 entries at a time) ** eventually to view tag archives, "most popular" archives, etc * next and prev were also used both in the context of the application: ** show me a list of entries I can edit ** show me a list of the most recent comments ** etc. So there is a real live use case. As for me, I am rarely one to get myself involved in a namenclature debate. More often than not, IMHO, they just go in cicles. Plus once you get involved, it is hard to extricate yourself from the debate. So for *me*, I don't really care what the link relations are labeled provided that their usage is well defined, and that there is concenseus around their intended usage. I have come around to favor: next|prev|first|last|(current or subscribe) because it is similar to (if not identical with) what we have alread implemented in our APP implementatin and in Japan. But also it is sufficiently specific and sufficiently open ended to give me (the implementer) a very reasonable set of ways to apply the principal. Byrne Reese
