[Cubicweb] Decentralization & Multi-source
arthur.lutz at logilab.fr
Mon Oct 11 12:30:09 CEST 2010
The broad question : can the multi-source feature promote and render
functional the decentralized internet ?
The web2.0 model right now offers centralized servers which offer
services on a proprietary software which community functionality. The
open source movement has alternatives for most of these online services
but often you loose the community/social side of things (unless you host
a multiuser instance which goes back to the centralized model).
Let's take an example : http://www.delicious.com
<http://www.delicious.com/> bookmark with a social twist. The added
value of using the central/community approach is that it suggests tags
for the URL you are bookmarking, you can see the number of times it's
been bookmarked and when you search your bookmarks you'll get
suggestions from other user's bookmarks. When you try to leave delicious
you will find scuttle  or semantic scuttle  which has basic
functionality. How difficult would it be for two (or more) cubicweb
instances to register each other and aggregate these informations ?
The exception of the "isolated web application" that is emerging is
status.net (twitter open source clone). With status.net you can host
your own microblog and subscribe to other self-hosted microblogs. That
way, you control your personal data but are still connected to the web
of other services.
Would this kind of registration involve multi-source ? Or some other
technology is preferred ? Linked Data ? I know that right now
multi-source is configured in the files, how far are from a modification
of that directly in the web app ?
The bonus question would be how feasible would it be to get a small
CubicWeb application to run on a low-CPU low-energy embedded device such
as a plug computer ? For inspiration see http://www.plugcomputer.org/
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Cubicweb