Re: more on the simplified feed format

[Seairth Jacobs <seairth@xxxxxxxxxxx>]

"Dare Obasanjo" wrote:

--- Seairth Jacobs <seairth@xxxxxxxxxxx> wrote:
  

Note that lack of the etag attribute is because I
realized you could 
handle that bit by doing a conditional GET on the
entry document itself 
(see my response to Randy).  This would also reduce
the number of times 
the feed itself needs to be updated (kept
synchronized with the entry 
documents).
    

Do you realize how ridiculously expensive this is? I
have users with hundreds of feeds and the thought of
performing a few dozen GET requests for a single feed
(e.g. if you are subscribed to Robert Scoble's feed)
is impractical. 
  

As long as the client and server support conditional GETs, I think it's still better than the current practice of repeatedly fetching a feed with everything crammed into it (even the stuff you already have).  Note that in http://www.intertwingly.net/wiki/pie/SuperSimpleFeedFormat, I have made the attribute available again, but optional.

As it is I already have gotten less and less use from
feeds that don't provide full content (e.g. I've
almost stopped reading Tim Bray's feed bacuase he only
includes the first paragraph in the feed and its
rarely interesting enough to hook me into launching
the browser to go to his page and I'm not the only one
I know who's mentioned this) let alone going to the
extreme of providing no content. I don't see how this
is beneficial to either content producers or
consumers. 
  

With this approach, that's wouldn't necessarily be an issue anymore, at least as long as the <content> element contains the entire entry.  You would only perform GETs on new or updated entries, which should be relatively quick except for the very first time you get a feed or if you don't get a feed for a long time.  Besides, this seems like an issue that doesn't really have anything to do with the feed format.

Seairth