The
greatest challenge for developers of Social Network Enablement software
will be to allow each of us
to portray our knowledge and our network(s) any way
we want to represent them.
 To let
us represent our knowledge and our networks our way is
consistent with the
principles of a World of Ends.
It is also consistent with the requirement, as I've described earlier,
that social software work with the content of everyone's
weblog/personal knowledge profile regardless of how we choose to
organize that content, i.e. regardless of how our 'mental filing
cabinet' is laid out and indexed. This is a critical design point: We
do not want or need a centralized ontology or taxonomy that each of us
is forced to adapt to. We should be free to organize our own stuff any
way we want to, and the software
should do the heavy lifting. So, for example, when my friend Gary Lawrence Murphy goes
looking for knowledge about 'connector tools for social computing' (his
term), the system needs to ascertain that this is substantially what I
call on my weblog the 'expertise
finder' component of 'Social Network Enablement tools'. Our mental filing
cabinets are organized differently, but that shouldn't prevent the
software from making the equivalency connection.
This is all about communication and accommodation of how our brains
represent things differently. It's about how we personally categorize
and search for information in real life (not the impoverished
keyword-popularity world of Google) and how we listen and interact
orally and in writing with people in real life. In every conversation
we translate from what we perceive the other person's 'mental model' to
be, to our own mental model, a process that learning experts call internalization
of information. To
be truly useful, to enable our weblog/personal knowledge profile to be
a true surrogate/proxy/stand-in for us, social software needs to use AI
to map, assess equivalency, and translate between our different
terminologies and styles of knowledge organization. This would
represent a big advance in software development, and will need to
involve linguists and semantics in the design process as this
company has been doing.
What does this mean for the networking components, the 'connecting
people-to-people' aspects of social software tools? It means that each
of us needs to be able to represent our networks our way, and
let the software draw
the bridges, connect the dots between them. It means, just as there
must be no standard taxonomy to which all our blogs must conform, there
must be no standard, mandatory directory format for our networks. The
Dewey decimal system of knowledge taxonomy sucked. The old hard-copy
Bell phone book sucked (and still does). The last thing we need is to
replace these old, inflexible, restrictive tyrannies with new ones.
Here's a practical example of what's needed. The diagram above shows my
mental model, how I represent my social networks (only in my head, I'm
not that anal). In the
diagram, the position around the circle shows the nature of my
relationship with
those in my network, and it's multivariate: it encompasses the
geographic nature of the relationship (people who are neighbours vs.
those elsewhere in Canada vs. those I've never met in person and have
no idea where they live). It encompasses the intellectual nature of the
relationship (what interests I think we share). It encompasses the
genealogical nature of the relationship (family versus non-family). And
it shows, via proximity to the centre, the strength of the
relationship -- the
closer to me in the middle, the closer I perceive the relationship, and
that's important information too. You'll notice that Gary appears more
than once (in red).
I'm sure Gary would represent his social networks very differently. He
may have a flat alphabetic personal address book/phone directory with
e-mail addresses beside the phone number and a 1-5 star system for
representing nature or strength of relationships, for all I know. The
critical point is that this shouldn't matter. Good social networking
software should accommodate both of our ways of representing our
networks, and help each of us identify useful new contacts, connect
with others and add people to our networks the way we want to.
I can hear the software developers throwing up their hands in
frustration. Let me say it again: When you force people to adapt their
mental models to a standard model (inevitably a complex one to
accommodate a variety of lowest-common-denominator specifications), a
standard model dictated by the technology and its designers, you will
get no usage, or at best reluctant, inefficient usage. Video
conferencing technology is a perfect example. It's too complex, too
counter-intuitive. It doesn't work
the way our eyes work. The webcam, once it can be made
unobtrusive, portable, and self-referential (so the viewer can see me, and
what I see, without my
intervention), it will become ubiquitous where videoconferencing
failed. Eyes open, eyes shut. Camera on, camera off. That's all the
controls it should have. The software should do everything else
automatically. World of Ends.
Same with social networking software. It has to allow each of us to map
our knowledge and our relationships our own way. I'm not proposing my
graphical representation of networks, above, as a standard. On the
contrary, I'm proposing that there be no
standard, but instead a sophisticated mechanism for translation that
sits invisibly (to us) behind each of our weblogs, and behind each of
our network lists, and allows them to be interpreted and used by
everyone else using their
mental models of what things should be called, how knowledge should be
organized, and how networks should be delineated and depicted. And note
that this software needs to reside at the >ends, not in some
massive central
location:
Maybe this is asking too much, to expect technology to accommodate
personal styles of thinking, communicating, and organizing. I think we
can and should ask no less. The technology is there to serve us, damn
it. And imagine what the 'translation' software could teach us about
how we communicate and how we learn, and about the 'signal loss' in
both and how it could be reduced!
So, to the dismay I'm sure of software developers everywhere, it's time
for us all to get selfish and insist on having it our own way.
|
MADE IN CANADA
trust your
instincts 
