[rdfweb-dev] foaf in slash-clones
marc at broadbandmechanics.com
Sun Oct 5 15:29:16 UTC 2003
This brings up some really REALLY interesting work and discussions I've
been having lately with folks - as we've gone over the realities of
inter-change between disparate social network systems.
We start off talking about mapping one set of existing 'social net
graphs' from one world to another, which is fairly straight forward.....
.... which leads us to the standard topic of "how do we know is someone
WANTS to get moved over to some strange social net - if you're a member
of one's friend's list?"
OK - so we have an answer for that concern. By putting an explicit
checkbox in participating systems which says "Export as FOAF?" we think
we can solve this issue. If the checkbox is NOT clicked, then simple,
stubbed out accounts are generated at the destination system, somewhat
similar to what TypePad does when generating lists of foaf:knows from
We call that 'dead data'.
But if one DOES click on the checkbox and give permission to "Export as
FOAF" themselves - then a hot link is established between originating
and destination systems - so any changes or additions in one systems is
mimicked on the other. This is the oft desired hot-link of live data
that (some of us) have always dreamed of.
OK - so that's the easy part. Associated control over one's FOAF file
is also desired, so folks can lock-out ANY association to their FOAF
file, or demand consensual permission over ANY association or just let
"catcher catch can" - and let what happens happen - via an 'open' FOAF
file. This control may seem absurd to some, but answers concerns many
have about security and privacy vis a vis FOAF.
Now what about systems that don't have such clean foaf:knows lists?
LinkedIN is one example. I met some folks on Friday (NetModular) - who
have a rather interesting approach as well. They build social nets for
enterprise and have a social net called "Sullivan Executive" and
"NetPlaya" (for Burning Man types....)
They've been playing around with assigning various granularities to
relationships and qualitative rankings as well. Tags such as Friend,
Co-Worker, Associate, Supplier, Client, Affilate, etc. are mixed with
Highly recommend, Recommend, Reliable and Unknown. They also plan on
moving into more "meaningful" meta-data as well (as they nebulously
Anyway the point is: "there are lots of different systems of
relationships out there."
As an inter-change play - we imagine being able to "map" top all these
different systems and facilitate movement of entire social net graphs
between these systems. I know Ecademy has it's own internal "rankings"
From: rdfweb-dev-bounces at vapours.rdfweb.org
[mailto:rdfweb-dev-bounces at vapours.rdfweb.org] On Behalf Of Julian Bond
Sent: Sunday, October 05, 2003 1:22 AM
To: rdfweb-dev at vapours.rdfweb.org
Subject: [rdfweb-dev] foaf in slash-clones
I've kicked off or re-activated a discussion in the Drupal dev group
about supporting foaf. This raises a couple of questions.
Most "Slash-clone" code maintains more or less complete member profile
records which probably map quite well onto <foaf:Person>. But few of
them have any function that looks much like <foaf:knows>. Does this mean
it's worth squeezing something into this mould the way typepad did with
blogrolls, or is there some neat function that slashclones ought to have
that would incidentally generate foaf:knows data.
Has anyone got an insider's link with Slash, Scoop, *-Nuke, etc
developer's groups or any of the other slashclone applications and could
encourage the building of a foaf module.
Julian Bond Email&MSM: julian.bond at voidstar.com
Personal WebLog: http://www.voidstar.com/
M: +44 (0)77 5907 2173 T: +44 (0)192 0412 433
rdfweb-dev mailing list
rdfweb-dev at vapours.rdfweb.org
More information about the foaf-dev