FoFTN: Friends of Friends and Topical Networks. Any implementations?

 It is virtually impossible to use the Internet nowadays:

 1) most Web sites are censored and/or "curated" (many of them
self-censor themselves!)
 2) trolls, shills and all kinds of state-sponsored sock-puppetry flood fora
 3) many users have an ad hominem approach to relate to other users
 4) most documents are unnecessarily repeated
 5) search engines dump on you millions of "hits"
 6) most pages have too many distracting ads (which could be parsed
off the html encoding anyway)
 ...

 Ideally I would like to not deal with the problems of textual
sequencing (there is no other way to write a text than character after
character) and segmenting (it would be nice if what seems to matter to
me is parsed and presented in a way that I can partially drill down on
the rest of comment if I so choose).

 There should be some way to segment, sort out, categorize and
stratify comments, pointing me to what I need to pay attention to even
within a paragraph; bypassing all the blah blah, set ups and all kinds
of verbosity which make like 80% of all texts

 This is something that should be implemented client side and among
clients themselves in parallel, because hosts may, <b>-will-</b>
easily abused such thing if we keep that information on the same
content server providing us with the data. In fact, there could be
"trust" servers in parallel to content servers for each
site/domain/path/page; users exercising that freedom themselves.

 I wonder how do you model trust, which to me is kind of a forest of
DAGs with some sort of relative Bayesian weightings.

 There are still some people whose criteria I would like to follow. I
trust A so I would like to be notified on what she has to say
regarding particular themes even while contributing to another thread.
A in turn trusts B, but B trusts C whom I don't trust ... There are
many such cases which may appear in people's regular interactions, in
fact that might be more of a topological problem

 I want for it to associate, both, based on people and the content of
previous messages I or other people I trust have posted and I want to
exclude from my views people's comments I don't care about. This way
you don't have to care about moderating/chaperoning fora. People would
take care of that for themselves

 Any implementation of such things? A RFC at least?

 thanks,
 lbrtchx

Received on Sunday, 28 June 2015 15:55:18 UTC