On Wed, Feb 6, 2013 at 5:46 PM, flawer <[email protected]> wrote:
I would hope that granular and modulable metadata assembled into contextualized descriptions,
i guess you refer to data explicit ontologies from resources types () and from more abstract or meta things (i.e. human values, courage, lazyness, etc. i.e. blue, red, etc)
yes :) describing objects with ontologies, or eventually with tags ( and enable a combination of a folksonomy and ontologies to describe objects and relations between them )
implicit data (i.e. logs in at same time on fridays, etc) would be even more interesting to network with.....
for then stablishing preferences of investments or for simplier exchanging material things as a excuse for getting in a better human relation with certain peers (the machine recommends me to invest/exchange with x related maybe unknown peers)
yes, manual queries using the system, or algorithms for the system to automatically make suggestions
if instead of preferences we say: requirements (to deal *just* with "couragous blue" mates), we have a kind of currency..
yes
(flow network is perhaps more accurate :) and brings the overlap issue:
Somehow, I see a variety of [clauses] , amongst other ontologies many seem to be taking for granted , of which perhaps certain could overlap ?
sure. in (i.e. human values), you shouldn't flow/convert couragous points into coward points, or you shouldn't use a shareful coward thing if you don't have coward points... or you should be able to clear your coward points with courageous points? haha.
:) yes... note : or perhaps even before any interpretation into "coward points" ( whatever the objective algorithm / interpretation of such points would be ) , the visualization of past transactions as a form of reputation , or of currently described contexts and suggestions, can speak for itself :) I guess, very much like on e-bay or couchsurfing
i guess this depends on the owner of the ontology, the relations he allowed that concept to be transferable with.
yes, ideally ontologies would be free to use... one would need to convene to use the same ontologies... though perhaps some kind of combination with a folksonomical approach could bridge some of the stiffness of ontological meaning giving ? Or natural language processing... but perhaps that becomes more complex, and I do not know much about it...
i imagine i am trying subscribing to "courage" and "courage2" flows in network x, which are incompatible between them.. owned by different authors).
i have overseen some more specifications in this direction, but i am... aa..
:)
_______________________________________________ P2P Foundation - Mailing list http://www.p2pfoundation.net https://lists.ourproject.org/cgi-bin/mailman/listinfo/p2p-foundation