p2p plus server-centric approaches?

Here's a piece I wrote in response to an email from Daniel Harris of Kendra. His words are indented (">").

============================================

Ok, let me take your points one by one, Daniel.

BTW, I use "brand" in this mail for convenience to describe individuals, organisations, businesses etc. and their services, products etc., as I'm interested in understanding all that as part of a seamless spectrum of identity in terms of the online world.

> Define "content and social networking"?

An online platform that allows individuals and groups to:
1) publish digital content and information about that content and about themselves, and
2) locate other people (including groups, organisations etc.) with whom they share affinities or who offer a particular product or service and interact with them and their content

> What do you mean by "users can connect directly to one another"? What is going on here? Users cannot connect. Their machines connect. Never directly though. It's not hard wired. It routed. What are you really talking about here? What layer are you talking about?

Excuse my lack of clarity—a bad habit of forgetting I haven't explained the context of an assertion sufficiently. I take your point in terms of the technical aspect of the issue. So yes, users' online interactions are invariably mediated by UIs, meta-data schemas, data-exchange protocols, IP routing etc. etc. What I was getting at, however, was the difference between users interacting:
1) Using a shared meta-language to describe their "brand" (themselves, the services or products they offer etc.) and to locate other brands according to the match between their search requirements and how those brands describe themselves. The meta-language is flexible in allowing users to construe their brand according to the ontology of their choice, yet facilitates the emergence of a common über-ontology from individual ontologies, hence allowing the matching of interests to brands (how I understand the Kendra concept—right?)
2) As (1), but with the added context of one of many possible semi-closed sub-networks which provides an integrated and consistent set of tracking mechanisms for the particular interactive modalities and dynamics of that sub-network—for instance, i-together's Closeness function that indexes the degree of separation and directionality of commenting, linking, content sharing etc. So now, rather than users' brands being represented solely by the users' own definitions, they are also contextualised by the web of interactive connections between all users of the given sub-network. This provides a feedback mechanism—much like real-world community—where users' brands exist in a context of multi-layered relationship to others, presumably encouraging open, honest and transparent behaviour.

Ok, I just knocked up this diagram to try to clarify what I mean [click]:



And, BTW, I have no idea if this integration of self-organised sub networks within a P2P über-network is remotely viable! Guys..?

> Is this all IM?

> I am not really interested in IM.

No, I threw IM out of the concept a while back. Take a look at a six-minute Flash presentation for an earlier version of the concept to get a feel of where the idea for i-together's been evolving from if you like. We still had IM in the concept then, but, like you, I'm interested in more enduring content/data.

> I am interested in publishing and querying structured data. Did you take a look at Kendra Base?

Yes, as you will have hopefully realised after reading the above (either that or I've completely missed the Kendra point! : ). I realise, though, that I'm not touching here on many of Kendra's other important aspects, such as the content distribution system, the revenue generation angle or the specific tie-ins with the entertainment industry. But none of that would seem to be contrary to the gist of the above. It's just I have no idea about those areas! : )

> I guess what would be nice is an idea of what are the problems you're trying to solve. I read the mission but then it all seems to jump to technology. Maybe I'm missing the in-between bit.

To provide a really simple way for people to express and share their creativity, build community and find people and stuff that excite them.

> I guess what I'm interested in is being able to ask my application: "Find me all organic restaurants within 30 mins cycle ride from N4 3LZ, UK that are open today till 8pm".

Wouldn't that be fantastic to be able to do?! And yet presumably, when you ask a question, you not only want particular kinds of information, but also a sense of the context of the information source—is it trusted and valued by others, is it trusted and valued by others that you yourself trust and value etc. That's what I'm talking about above.

> It doesn't have to be written in natural language. I'm quite happy to enter it as in a form type system.

That makes sense.

> I read a lot about "content sharing" but nothing about sharing ideas. Care to comment? I want to ask the system "how many people want to go to war?".

You like to set yourself big challenges, huh! I don't know the answer to that one, either the question or the question about the question. How about we talk some more and try to work it out?