[General] Interop 2: Content Syndication / Exchange

[Gregor J. Rothfuss]

here is a proposal for our second interop
session, "Content Syndication / Exchange"

rss for syndication / discovery
-------------------------------

my conclusion so far is that we should leverage
rss as much as possible, because it is

a) already widely deployed
b) has a lot of sexy uses besides "boring" interop [8]

fully leveraging rss would allow us to discover 
the necessary resources to then start syndication, 
content exchange over web dav or initiate
xml-rpc / SOAP calls.

we can probably put a lot of the discovery stuff
that roger mentioned into the rss header, using
rss modules. this would keep the number of new
formats to a minimum and should be transparent
to (well written) rss readers. [9] is an especially
interesting one.

relevant for our purposes seem the following rss modules:

mod_admin
This module provides administrative properties that can 
be used to help improve the robustness and reliability 
of broad RSS usage between providers, aggregators, clients, 
and other users. is a URI (typically a 'mailto:' URL) for 
contacting the person or source of the particular instance 
of RSS/RDF information.

mod_cc
This module aims to give metadata regarding the copyright 
license under which the RSS feed, and the objects it 
points to, are released under.

mod_link 
The mod_link module supports the syndication of site 
link information along with RSS 1.0 XML feeds. The W3C 
HTML link mechanism provides the original inspiration 
for this module. 

mod_search 
Provides addional information about items resulting 
from search engine query results. 

mod_servicestatus 
This module extends RSS to include elements which 
allow the description of the status and current 
availability of services and servers. Some data, 
such as whether a server is up or not, would 
normally be generated automatically, whilst 
other data, such as explanatory text for humans, 
might be the result of further processing or direct 
human input.

mod_taxonomy 
The taxonomy module is a RSS 1.0 module and a 
RDF application enabling the identification of 
topics covered by a RSS channel or item.

plus, we would probably come up with additional modules for:

- supported APIs (links to WSDL etc)
- last changes on the site (not just feeds) [14]
- site metadata
 - name
 - responsible
 - supported languages [16]
- supported formats for content exchange [15]

the problem of discovery of this rss file can be tackled in two
ways:

a) a well-known location [5] such as /w3c/rss.xml
b) a LINK [7], [10] element in the HEAD of every page, such as
<LINK rel="rss-feed" href="..." />

as for the format of the discovery file, there is some
prior art [6] as well.

how does rss fit in?
--------------------
the way i imagine it, both web guis like xopus or bitflux
and other cms systems [11] would first query this rss file,
and use it to adjust their services / functionality to
the cms they are talking to.

what does this NOT solve?
-------------------------
we still need a format for the content itself. there are several
approaches to describe cms content, but most are focussing on
blogging. [12] it may be feasible to use a subset of jsr 170 once
it emerges publically in december 2002. this depends on whether
jsr 170 will be API-centric, or defines some XML formats as well.


resources
---------
rss 1.0 modules [1]
http://web.resource.org/rss/1.0/modules/proposed/subscription/

rss for search engine syndication [2]
http://groups.yahoo.com/group/syndication/message/192

NewsML [3]
http://www.iptc.org/NewsML/DTD/NewsMLv1.0.dtd

Article about XML in news syndication [4]
http://www.xml.com/pub/a/2000/07/17/syndication/newsindustry.html

well-known location [5]
http://www.w3.org/TR/P3P/#Well_Known_Location

discovery.xml [6]
http://groups.yahoo.com/group/syndication/message/2462

LINK HTML element[7]
http://groups.yahoo.com/group/syndication/message/2522

rss book [8]
http://rss.benhammersley.com/

P2P blog net [9]
http://www.peerfear.org/blognet/

rel=META [10]
http://lists.w3.org/Archives/Public/www-rdf-interest/2002Jan/0074.html

OSCOM Interop Approach [11]
http://cvs.oscom.org/snapshots/projects/interop/interop-latest.tar.gz

weblog specs prior art [12]
http://www.truthlaidbear.com/blogmd/priorart.html

jsr 170 [13]
http://www.jcp.org/jsr/detail/170.jsp

recently changed pages [14]
http://davenet.userland.com/1997/07/26/JITSEs

exposing standard formats [15]
http://internetalchemy.org/ocs/directory/0.5/#stdformats

IETF languages standard [16]
http://www.ietf.org/rfc/rfc1766.txt