On Wed, 2 Jan 2008, James M Snell wrote:
Peter Keane wrote:
[snip]
Feed-level elements are of two types: metadata about the feed, and items
in a "set" of things. Throwing a new type in the mix is really a huge
change -- if it's about the feed, OK (but clearly a tombstone is an
I disagree. Feeds already contain several sets of things... entries,
authors, contributors, links, and categories. Adding a new set of
things is not a "huge" change by any stretch of the imagination.
But those are all clearly "metadata about the feed" i.e. they have a
well-defined relationship with the feed. Same with entries -- the
conceptual model (although not spelled out in the spec) is easy to grok.
Being able to abstract out a conceptual model is *really* important and
it's what makes Atom so useful. What sort of things are these tombstones?
Well, their most obvious relationship is to an entry not the feed itself.
How would I extend one of these tombstones in a similar but
not-exactly-the-original-purpose way? Atom is an elegant set of
*interfaces* and here are these *implementation details* (tombstones)
smack dab in the middle.
Also, what sort of slippery slope is this? Will this be the start of more
bags of stuff being added to a feed? Right now a feed has a 1:1
relationship with it's set of contained items. Suddenly that's 1:2
(entries and tombstones) which effectively means 1:n. Note I still
contend that metadata about the feed itself is a different sort of beast.
a deleted entry like that -- a new type seems to me to presuppose an
understanding of the possible use cases which is simply impossible at
this point.
The key challenge with these kinds of views is that there are existing
applications and use cases out there that simply do not share this point
of view.
I would highly recommend that you take a look at the Open Archives
Initiative Protocol for Metadata Harvesting because Atom shows great
promise as newer/simpler protocol for the exact same purpose (to the
extent that OAI-ORE could replace or at least extend OAI-PMH, it's a done
deal since OAI-ORE is expressed in Atom). This is THE widely implemented
standard in the archives area and work on issues like tombstones and feed
archives with Atom really requires (it seems to me) that folks understand
how this is being done right now -- OAI-PMH is a truly impressive piece of
work and has proven incredibly useful.
--peter
