Hello all,
27 years olded, i am not a famous blogger nor a web entrepreneur i
am just Loïc [lo-e-k], i managed several e-commerce development teams
at France Telecom and i manage now a technical team on the shared
(corporate) mobile services platforms department within the same
company, responsible for operations & impact management, exploit tools
development and processes optimization. I am not naturally a great
social animal but i use computers since i am ten years old and the web
since 1996. I discovered the web2.0 paradigm very late due to my
social skills lacks but i am now convinced to do evangelization and
help to define the next step.
When talking about data portability we're generally talking about
the detailed data transportability between two or more services but i
think the first topic to discuss is about the universality of the end
user social network allowing him to really have only one acccount per
service. Perhaps have we to consider the whole before the detail
because the detail conception will determine what it will be possible
to do into the whole ?
I will try to explain my vision of the data portability taking in
consideration my current knowledge level of the objectives of this
workgroup.
What we are thinking about is the transposition of the real whithin
the virtual : in the real world i am one physical person but i have
several lifes. As known i am not really the same person between
school, house, office, holidays, this google group and so on.. The
question to solve is not really how to duplicate my ID but how to
store more than one ID into my wallet and don't be afraid by police
controls (well.. goal is not to give tools to bypass the law.. ;) ),
it's the ability to have only one fidelity card for all supermarkets,
perhaps shared with my wife but where my really private stuff can't be
viewed by her. Because the life is not as simple ;)
I think the first work on data portability subject could be the
determination of all use cases that can be considerated, then their
prioritization according to available technos and industry actors
views, and then only after that fundamental work the other subjects
could be discussed.
To achieve this work some namings could be found as a first step
for :
- user types
- services
- identities
- ...
As I have not yet a general view on all the work done, i wanted your
opinion on the following scenario, illustrated below :
As Alice, Bob and Charlie are now old school people, i renamed them
to be more actual :)
[Insert picture here :
http://picasaweb.google.com/mglcel/Public/photo#5222897023971396434]
In some years, when
dataportability.org group will have done all the
work needed :
Alina have three digital identities :
- First : she had an account on MySpace when she was teenager
- Second : she have created another Myspace identity (not account)
when she began to work, a LinkedIn account as she has not used this
service before, and a facebook account.
- Third : she want to share some personal stuff with its closest
social network on facebook
Brayden is a complicated and hyperactive guy, he needs four
different digital identities :
1) its private life, shared with its family, closest friends and his
girlfriend, Alina, on facebook, twitter and flickr
2) its extended private life for his extended friends (blog and office
friends for instance). He uses facebook, twitter, flickr and LinkedIn
3) its hidden private life, well.. he have an hidden french
girlfriend, Claire.
He shares some stuff with her and the friends of its french second
life on facebook and flickr
4) Brayden is a paparazzi, he want to only share its people photos
with its colleagues on flickr
Claire is afraid by the web and data sharing, data visibility and
security considerations.., she only have two distinct identities :
1) Public on LinkedIn because she is searching for a job
2) Really private on facebook as some friends have invited her. But
she uses a fake name, it could have a possibility that companies try
to search her on this social service, she don't believe the computer
security experts..
All these profiles could exists, it is not our job to make any
judgment no ? Now some considerations :
a) how each actor could manage these different identities from one
unique place, OpenID extension, Higgins ?
b) when creating multiple identities we create multiple social
networks which are closed each other
c) Alina don't want that its old social network of teenager interferes
with its current working girl one
d) her facebook account have to manage two identities and so two
different networks, perhaps she want or not that its extended network
can view the members of its family network but she don't want to share
her daughter pictures to them.
e) as twitter messages are used to update the facebook status of
Brayden, he must have the possibility to have two different status,
one visible only for family and the other to its extended network ;
twitter and facebook communication must take this aspect into
consideration. When he submit a message on twitter he must have the
choice to submit it to one or more identities, as an extension he
perhaps want to have two different visible nicknames on twitter.
f) Brayden must be sure that Alina can't learn the existence of Claire
with a normal web manipulation : its secret french network must not
view any information that he don't want to share with us, on facebook,
twitter, flickr or LinkedIn.
g) On Flickr he don't have only family, friends, public and guests, he
can manage each information and media according to the visibility he
want to give it.
h) For all these reasons, Alina, Brayden and Claire don't want that
their personal information could exists in search engines or at least
they want to give them a time of validity. Using extended semantic web
concepts they have the ability to indicate or modify at any moment the
validity date of their informations. Alina have the ability to delete
the cached pages of its old teenager blog as the content is clearly
identified to be hers.
i) On LinkedIn, perhaps they want to share some details only with some
contacts (multiple points of presence are not shown on the diagram)
j) Identification is unique on all services, authentication is
identity-aware but digital identities are service-agnostic
Each identity corresponds to a different social network, to a
different personality.
All relevant aspects could be translated for very different services :
google reader share list, friendfeed, youtube,
del.icio.us bookmarks
and so on..
This design is very far to be perfect and really incomplete but do you
think that all highlighted aspects have been taken into consideration
within the dataportability project ?
Thanks for your time.
I manage to act as an evangelist for the french people, who is in its
large majority so far of this vision of the web, i am waiting now for
one year the facebook invitation validation of my best friend, data
sharing is the main fear, we are paranoiac people..