Some further reading: Designing Semantic Publish/Subscribe Networks using Super-Peers<http://www.kbs.uni-hannover.de/Arbeiten/Publikationen/2005/chirita04designing.pdf> Chirita, Paul-Alexandru and Idreos, Stratos and Koubarakis, Manolis and Nejdl, Wolfgang
My notes available here: http://iss.im/node/93 On Mon, May 12, 2008 at 5:42 PM, Nick Vidal <[EMAIL PROTECTED]> wrote: > Hi Dan, > > Really interesting work you got! I read about it a while back and wanted to > get in touch with you to discuss some ideas I have about searching. I have > already discussed it privately with a few friends but I think it would be > nice to have some feedback from this list. It's half-baked, but this is what > I have in mind: > > I've been working on ISS (Instant Syndicating Standards) <http://iss.im/>. > It was part of Google's Summer of Code in 2006 and I had great help from the > XSF and the Psi Community. The key concepts are: > > a) we have a format that associates each individual with their own > broadcasting channels (called *tagcloud*); > b) and a format that describes how these channels are connected through a > trusted network of people (called *taglink*). > > The next step that I'm exploring to include in ISS is Search. This is the > basic workflow: > > a) each individual generates a social graph beforehand consulting the > cascading taglinks; > b) a query can be sent to friends up to *x* degrees apart, where *x* is > define by the user; > c) this query is published on the users' *searched* node with an ID; > d) each query has a TTL (e.g. TTL = 1 month); > e) friends may accept the query and see if they have entries that match. If > so, they send the IDs of the matched entries and publish that to their * > matched* node. The query is kept until the TTL expires or according to the > policies. > f) users receive the matched entries in their aggregator. These appear > associated with the original query and separated from the main flow of the > aggregator. > > The idea is to have a *syndicated search* that is totally decentralized > and served by friends (and friends of friends) for an extended period of > time. > > I still need to work on it to cover more specific details, but feedback is > welcome. > > Best regards, > Nick Vidal > > On Mon, May 12, 2008 at 3:53 PM, Dan Brickley <[EMAIL PROTECTED]> wrote: > >> Peter Saint-Andre wrote: >> >>> BTW, I had to tweak something in the Mailman admin interface so I >>> decided to see how many people are subscribed to this list. Turns out >>> there are ~250 of you. I didn't think so many people were interested in >>> the intersection of XMPP and social networking, but I was wrong. :) >>> >> >> Which is as good a nudge as any to finally de-lurk. Um yup, XMPP is one of >> the quiet success stories of social networking interop. It really seems to >> have turned a corner these last few years and all that hard work is paying >> off. >> >> I have various interests here, ... most generally, in looking for >> integration opportunities across the various 'social web' technologies that >> have recently matured. In the FOAF project for example, we're interested in >> use of XMPP for representing groups, buddylists, and sharing of user >> profiles. More generally I've been hacking around with the use of XMPP as a >> data bus for RDF querying using SPARQL (as Peter well knows, being my XMPP >> helpline). Some notes on that at [1]. I'm also interested in ways of >> describing XMPP group chats in ways that make them more findable, so the >> recent work on using HTML 'link' for autodiscovery is rather promising. >> >> Also I'd like to note that at both Social Graph Foo Camp a couple months >> ago, and at XTech in Dublin last week, a good few people noted "oh, I didn't >> realise how important XMPP was before seeing these talks". >> >> cheers, >> >> Dan >> >> >> [1] http://danbri.org/words/2008/02/11/278 >> >
