2010/6/8 Hellekin O. Wolf
<address@hidden>
On Tue, Jun 08, 2010 at 12:20:23AM +0200, Melvin Carvalho wrote:
> 2010/6/7 Rob Myers <
address@hidden>
>
> I'm not the biggest fan of facebook, but perhaps not 100% fair to call them
> completely centralized (though I absolutely agree centralization can and
> does cause problems).
>
> But credit where credit is due, the opengraphprotocol which they introduced
> this year at f8, is probably the leading decentralized system out there, and
> by some distance imho.
>
*** Short version: so yeah, they designed an ontology for mass
marketed products and interests. Can you expand on the distance?
They could easily have chosen to invent their own thing, but opted for standards based HTML5 / RDFa.
What they are doing is decentralization at web scale, so that existing sites like espn, imdb, rotten tomatoes and at least 50 other partners, are interoperable in a distributed way.
I just dont see anything else out there (maybe with the exception of RSS etc.), that has that kind of reach / adoption, at web scale. Though please correct me if I've missed something.
RDFa 1.0 and 1.1 have had several RFC as part of HTML5. Anyone is entitled to build 1) vocabularies using it 2) applications using it ... that's what the web is all about. It's just that facebook have chosen to be an early adpoter, and respect to them for that. As the ecosystem grows, everyone will benefit. No one is obliged to use facebook widgets, but anyone can use the rich data. And the schemaless property of the data allows 100% data freedom, rather than hiding/restricting data behind an API.
Facebook are within their rights to make their own vocabulary. As it happens they did seek advice from Dan Brickley (author of FOAF) and also had a hands on face to face session, with Tim Berners-Lee ... and also openly discussed next steps on their mailing list.
Why did they do this? Well they wanted to further their business model ... that's fair enough. But in the process have made a bit step forward to creating a more open distributed web.
Long, rantish version:
*** I spent 10 minutes looking up OpenGraph. I couldn't see an RFC,
nor the API, as its access requires a Facebook login. Is there any
peer review involved? But I saw a very handy way for FB to "hire"
users to refine their commercial semantic dataview of the world.
I'm not saying the approach doesn't have a technical interest,
although it doesn't strike me as particularly innovative, but it is
clearly based on the assumption of the Consumer-User, sharing consumer
market information. Broad categories a l'emporte-piece, but only for
very broad cultural products. No dance, no poetry, no plastic, video
nor media art; no journalist, no activist (public figure? non profit?
Euphemisms); nothing out of website or blog?; etc.
Certainly not an example of a social network for human beings. It
seems to me that the important part relates to loved ones, relatives,
people with whom you share trust and intimacy. That part of the
Social Graph that should remain hidden from the public. But even
considering the public side: isn't it shameful to encourage people to
languishly build useless knowledge about mass media and entertainment?
(because, statistically, that's "what the users like", when bombarded
with such contents) One has to be pretty deeply swamped into the tiny
perspective of the glimpse of Civilization that we call the
consumerist society to ignore the consequences of presenting such a
limited worldview of and to his customers.
The great thing is that we now have an indentikit of the average
consumer. Not very encouraging, I must say. Fortunately, this
average consumer is a bit more complex than that and will eventually
opt for a system that allows her to expand and improve it openly
according to the sacred principles of the Church of St Ignacius.
An important point of OpenGraph is that it allows already public
content to fit the data model of Facebook. It's another version of
the Mechanical Turk. In a world of freedom, free work goes to the
king!
All right, I can live with good restaurant ratings too. But in the
future, there are a bit more important things to give attention to
than marketed products. Things like culture, education, poverty,
health, food, that are not following "the charts".
==
hk