Roller and JavaBB devs:
I'd like to take a moment to let you guys know about a project I'm
just launching. I'm posting this here since the project is currently
using substantial amounts of code from both Roller and JavaBB (as well
as JA-SIG CAS). Having spoken to both Dave Johnson and Dalton Camargo
already, I believe that there is much potential for good relationships
between my project (OpenQabal) and the Roller and JavaBB communities.
The long-term goal for OQ is to abstract everything out to the point
that all of the functionality is pluggable, with various pieces composed
into an overall system using standard interfaces (probably XML based,
but that isn't necessarily a given). If that happens, our hope is that
we can get whatever functionality is required to do this composition
merged upstream with Roller and JavaBB so that we are not maintaining
a fork per-se of either project. However, in the interim, the coupling
between the two codebases in OQ is such that we are shipping slightly
modified copies of both Roller and JavaBB codebases in our project.
Of course my intention is to share back with Roller and JavaBB and
any bug-fixes, enhancements, or anything else that is developed
for OQ is certainly made available back to you guys. I have already
signed an Apache "contributor license" thing so that I can contribute
code to ASF projects, and everything we do for OQ will be Apache License
v2.0.
Anyway, to the project itself:
The project is called OpenQabal, and in short, is all about creating an
open-source platform for social-networking and collaboration. We are
focusing on the principles of federation, composition, and openness;
with a special emphasis on enabling "distributed conversations" and the
"federated social graph."
Specifically some of the goals of OpenQabal are:
* Provide a fully functional and powerful social networking &
collaboration platform and default suite of applications.
* Provide a snap-in architecture to allow for easy insertion of
additional applications (or replacement of default applications) into an
OQ instance
* Leverage federated identity protocols such as OpenID, SAML,
WS-Federation, XACML, Liberty ID-FF, Liberty ID-WSF, OAuth, etc. to
eliminate the "profile fatique" problem and allow federation of a user's
"social graph" across domains
* Provide open access through standard protocols and APIs to all of
the functionality and data associated with an OpenQabal instance
* Interoperate readily with other applications which are built on
open standards such as RSS, XHTML, Web Services, WS-Trust, OpenSearch,
OpenGIS, WSRP, Atom, FOAF, XOXO, XFN, SyncML, LDAP, DSML, OPML, XBEL,
SPML, APML, RDF, RDF Schema, OWL, Dublin Core, ODM, SPARQL, SRU, SKOS,
RDDL, POWDER, GRDDL, XLink, XInclude, XPath, XQuery, RIF, KIF, XTM, MOF,
XMI, SWRL, SWSF, FIPA, AIML, XMPP, etc., so that OpenQabal instances can
both host, and become components of, mashed-up / composed internet
applications.
* Leverage emerging Semantic Web technologies to build a richer
model for linking blogs, to allow conversations to span multiple domains
and providers (eg, we want to build something that goes beyond simple
Trackbacks).
* Support Microformats where appropriate.
* Support the Google OpenSocial APIs
* Provide a richer model (and associated tools) to enable real
dynamic conversations, as opposed to contemporary blogging platforms
which mostly just provide a flat view of a post and it's associated
comments. Think Slash with it's powerful comments system.
* Introduce the model of a Conversation as an abstraction with many
views (Forum style, Blog style, NNTP, e-mail thread, etc.) into the same
underlying conversation
* More to come...
Development is just getting started and the initial version is a long
way from the overall vision of a truly federated social-networking
system. But substantial functionality exists already and the project
can be built and deployed. If you are interested in this sort of thing,
surf over to the project site at <http://openqabal.dev.jav.net> and
check it out.
TTYL,
Phil
