Facebook data portability
Kai Hendry
I'm trying to ensure that I have my own data. I can't figure it out
with Facebook. My searches seem to say it's not possible with a lot of
ToS that I couldn't give a toss about.
There is even a dataportability group on Facebook... so how do I get
my data out? I want a dump of all my friends emails at the very least!
Elias Bizannes
Kai Hendry
> Another good start is to challenge your assumptions. You don't want to dump
> your friends e-mails per se, you just want to get the benefits of using that
> data in other applications. You can get data portability through data
> accessibility, not just exportability.
the emails of my friends to a mailing list and leave it that for now.
I think the real problem is how centralised Facebook is. There should
be technologies in place where people can easily create a home
(profile) page and have reasonably the same amount of privacy Facebook
gives.
Paul Massey
forget to mention that social media has complicated this term) may not
want you "dumping" or "listing" or "using" their email addresses and
its unlikely that current Privacy Policies permit data controllers to
allow such processing of your friend's data in this way.
Also - what do you mean by "high level"? To me its a non-detailed
bird's eye view. If you take such a view, your going to miss the
detail that make data portability challenging (but not impossible)
from a data protection and privacy perspective.
Kai Hendry
On Nov 10, 11:54 am, Paul Massey <paul.mas...@klgates.com> wrote:
> Of course there is also the issue that you're "friends" (lets not
> forget to mention that social media has complicated this term) may not
> want you "dumping" or "listing" or "using" their email addresses and
> its unlikely that current Privacy Policies permit data controllers to
> allow such processing of your friend's data in this way.
email from me. Don't be ridiculous. :)
> Also - what do you mean by "high level"? To me its a non-detailed
> bird's eye view. If you take such a view, your going to miss the
> detail that make data portability challenging (but not impossible)
> from a data protection and privacy perspective.
identify my friends. If facebook refuse to divulge that "sensitive"
information, that just having the names would be nice.
average
> DataPortability is a new approach to approaching the economics of
> information. It's a slow, painful way that's evolving as it requires lots of
> discussions, arguments and thinking with no clear proof we are being
> 'successful'. But the best contribution you can do, is
> a) come up with an economic argument why data portability is in the
> interests of companies like facebook
> b) identify courses of action to get them to transition. One over hyped area
> is open standards; one underdeveloped area are the terms of service
> agreements
> c) Have people join the conversation.
I've wanted to say it for a long time,and mostly this is directed to
Elias and Chris, but this whole idea of getting companies to be
interested in data portability seems way too Old Paradigm. And I
find that what is becoming of data portability is getting too narrow to
the point of mundanity. The most obvious point to me is that
data portability and social networking is ultimately *peer-to-peer*,
just like real life, so why cater to companies? Regarding your point a)
above, for example, I'm convinced there is no economic advantage
to companies like Facebook, but that's a bonus :).
What happened to APML? Google Chrome, for example, offers some very
interesting possibilities to gather "attention data" and could
automatically build a rich "personality profile" of each individual.
The ability to store a user's social graph would offer some interesting search
attenuation and should be built into it. And such a complex and rich profile
could be used as a sort of authentication mechanism when entering various
web domains, much like we use a person's dress or hair style to identify the
"tribes" they belong to. By keeping this data "*locally*, a site could know
what kind of conversation to have with each individual as they enter, and
the individual wouldn't have to copy it anywhere or give their data
assets to anyone. And that's just one SMALL example.
Chrome, for example, being nearly a complete open-source webOS in
itself, could offer p2p services that would bring the social network
right where it belongs--to the person, not facebook, flickr, or some
other bullshit commercial service. COMPANIES SHOULD BE CATERING TO
YOU, not the other way around. Cater to the venture capitalists and
economists who, if they're smart, will see the potential for a new
kind of economy, both in terms of marketable products and skill
trading; perhaps without even realizing it, the world is clamoring for
it. I know I am.
And I know it hasn't come back into vogue, but micropayments ARE the
future, despite Clay Shirky's premature declaration of its death years ago.
What was missing before was a sophisticated enough personality profile
to gauge the user's interests and a platform on which users could
offer their own data and services so they could feel like equal
partners in a community of trust. Take all the raw personal data that
individuals (not corporations mind you) are creating: journal
entries, music, pictures, etc, etc, and put it into a p2p platform
with abilities to trade credits of some sort and watch NEW WEALTH
being created. On top of the data, imagine scores of open,
loosely-coupled, modular applets that act as go-betweens between
peers, finding new and innovative ways to build on top of that data.
PhotoSynth being one cool example of how thousands of photos from
random users posted around the internet can recreate *in 3-d* popular
destination sites like the Notre Dame. Also consider a web-crawl app
that aggregates personal journal entries in the mesh, looking for similar
conversations happening elsewhere and grouping them into a tunable,
on-the-fly, interest-specific, news roll. Forget the hierarchical domain
name system as a tool for aggregating a domain-specific interests,
abstract it out into a richer set of meta-info -- imagine a domain cloud
of individual content enriched and combined with tag-filters, reputation
systems, etc. and providing app-templates to slice and re-form all that
data to generating new and richer meanings.
The possibilities for data portability are way too big and interesting
and corporations simply will NOT be the innovators in this area, isn't
this obvious? So don't sell your soul just because you get a little
attention from the rich and famous. Though Google still seems like
the best partner so far. Anyone hiring out there?.
Still dreaming....
marcos
Elias Bizannes
> Regarding your point a)
> above, for example, I'm convinced there is no economic advantage
> to companies like Facebook, but that's a bonus :).
It's a shifting economic reality that I am talking about; and it's a
zero sum gain. However the advantage Facebook could get, is
repositioning itself to become a very important part of our lives
more so than now. Have a read of my value chain for information[1]
> What happened to APML? Google Chrome, for example, offers some very
> interesting possibilities to gather "attention data" and could
> automatically build a rich "personality profile" of each individual.
> The ability to store a user's social graph would offer some interesting search
> attenuation and should be built into it. And such a complex and rich profile
> could be used as a sort of authentication mechanism when entering various
> web domains, much like we use a person's dress or hair style to identify the
> "tribes" they belong to. By keeping this data "*locally*, a site could know
> what kind of conversation to have with each individual as they enter, and
> the individual wouldn't have to copy it anywhere or give their data
> assets to anyone. And that's just one SMALL example.
APML is just one of the many standards we have broadly encouraged.
However as I will announce soon, we are about to begin a process of
"evaluating" open standards. I think before we can commit to relying
on these standards (as what you describe is more the place of an
entrepreneur taking advantage of the market trend that is data
portability), the actual DataPortability.org needs to make it clear
what are the standards that are safe to use from our perspective.
> Chrome, for example, being nearly a complete open-source webOS in
> itself, could offer p2p services that would bring the social network
> right where it belongs--to the person, not facebook, flickr, or some
> other bullshit commercial service. COMPANIES SHOULD BE CATERING TO
> YOU, not the other way around. Cater to the venture capitalists and
> economists who, if they're smart, will see the potential for a new
> kind of economy, both in terms of marketable products and skill
> trading; perhaps without even realizing it, the world is clamoring for
> it. I know I am.
Some solid ideas there. It's still Google though, and as much as they
love to tout "Open", they shouldn't be embraced blindly. The DiSo
Project is doing some great work in creating a decentralised social
network, by using a blog as the centre of our your existence. I think
your idea is similar but much more interesting and open to the broader
population - would be interested to see more of your thoughts.
But the reality is, companies are a stakeholder, just like the
standards communities are, users, governments, and the rest. You can't
ignore them.
> The possibilities for data portability are way too big and interesting
> and corporations simply will NOT be the innovators in this area, isn't
> this obvious? So don't sell your soul just because you get a little
> attention from the rich and famous. Though Google still seems like
> the best partner so far. Anyone hiring out there?.
I totally agree. But for different reasons - having seen it first hand
myself, companies by design struggle to innovate. Despite all that is
touted by Google, they suffer they same corporate bullshit the rest of
the world suffered by their internal entrepreneurs. I would argue the
reason Google has been so innovative, as a lot to do with the fact
they've made some smart acquisitions and brought fresh blood in to run
new product lines which gives them some freedom.
You might be interested to read a blog post I did on why big
organisations suck [2].
> Still dreaming....
Thank you for contributing. Happy to chat with you off this mailing
list on your thoughts. We need more people like you to voice their
thoughts - whether on this mailing list, in the blogopshere, even on
Twitter. That's how stuff gets done - you start with a conversation,
you identify, and then you work towards.
Elias
[1] http://liako.biz/2008/05/the-value-chain-for-information/
[2] http://liako.biz/2008/06/organisations-need-to-be-a-size-12/
Kai Hendry
> The possibilities for data portability are way too big and interesting
> and corporations simply will NOT be the innovators in this area, isn't
> this obvious? So don't sell your soul just because you get a little
> attention from the rich and famous. Though Google still seems like
> the best partner so far. Anyone hiring out there?.
from my experiences with Google checkout.
Of course it's in Google's best interests if Facebook et. al. open up,
so Google can spider through their valuable data and offer it to you
in search results with advertising.
I'm still not convinced you can advocate to big companies being open
is in their best interests. I think the user deserves some basic
rights. It's my bloody data and I want it back.
Elias Bizannes
> is in their best interests. I think the user deserves some basic
> rights. It's my bloody data and I want it back.
This isn't a holy war, but merely an awakening to a better way of
doing things. You might be interested to read a blog post I wrote
recently, which I think will make you realise the user still has the
power and its in the interests of these companies to maintain a
relationship with you, not a once-off data grab:
http://liako.biz/2008/11/you-dont-nor-need-to-own-your-data/
Also read Chris's recent post about data lock-in:
http://chrissaad.wordpress.com/2008/09/04/data-lockin/
average
<nor...@googlegroups.com> wrote:
> DataPortability.General
>> Regarding your point a)
>> above, for example, I'm convinced there is no economic advantage
>> to companies like Facebook, but that's a bonus :).
>
> It's a shifting economic reality that I am talking about; and it's a
> zero sum gain. However the advantage Facebook could get, is
> repositioning itself to become a very important part of our lives
> more so than now. Have a read of my value chain for information[1]
I'm not really interested in having Facebook being a bigger part of my
life any more than I'm interested in having GE (General Electric) be a
bigger part of my life. I'm interested in what will facilitate the
generation of new value, novelty, creativity. Obviously Facebook
would like to be a bigger part of my life, but not for any reason that
would last even a generation. Ultimately the only ideas which sustain
themselves are those which engage the human spirit.
Facebook isn't really creating vibrant communities, for example, which
is what social media SHOULD be doing. At best, Facebook is like a
geeky club; MySpace is even worse and more like a dive bar.
>> What happened to APML? Google Chrome, for example, offers some very
>> interesting possibilities to gather "attention data" and could
>> automatically build a rich "personality profile" of each individual.
>> The ability to store a user's social graph would offer some interesting search
>> attenuation and should be built into it. And such a complex and rich profile
>> could be used as a sort of authentication mechanism when entering various
>> web domains, much like we use a person's dress or hair style to identify the
>> "tribes" they belong to. By keeping this data "*locally*, a site could know
>> what kind of conversation to have with each individual as they enter, and
>> the individual wouldn't have to copy it anywhere or give their data
>> assets to anyone. And that's just one SMALL example.
>
> APML is just one of the many standards we have broadly encouraged.
> However as I will announce soon, we are about to begin a process of
> "evaluating" open standards. I think before we can commit to relying
> on these standards (as what you describe is more the place of an
> entrepreneur taking advantage of the market trend that is data
> portability), the actual DataPortability.org needs to make it clear
> what are the standards that are safe to use from our perspective.
What I see is more than a market trend, these are things that
ultimately drive an individual's passions. Matching up people around
the globe in more creative and efficient ways is something beyond a
trend; these are interpersonal, fundamental needs that transform and
can radically re-shape societies--how they unite and group themselves,
govern themselves and share power. It's a bit outdated by now, but
Marshal McLuhan _Understanding Media_ is still a great read a highly
relevant to the discussion. I view it as compulsory reading for
anything relating to understanding the ramifications of the internet.
>> Chrome, for example, being nearly a complete open-source webOS in
>> itself, could offer p2p services that would bring the social network
>> right where it belongs--to the person, not facebook, flickr, or some
>> other bullshit commercial service. COMPANIES SHOULD BE CATERING TO
>> YOU, not the other way around. Cater to the venture capitalists and
>> economists who, if they're smart, will see the potential for a new
>> kind of economy, both in terms of marketable products and skill
>> trading; perhaps without even realizing it, the world is clamoring for
>> it. I know I am.
>
> Some solid ideas there. It's still Google though, and as much as they
> love to tout "Open", they shouldn't be embraced blindly.
I wasn't touting Google, per se, so much as lauding the trajectory and
vision they are sharing and projecting with the the Chrome
project--not to mention the risk they're taking. Cyberspace (does
anyone even use that word anymore?) needs a better, more versatile and
dynamic window into the interconnected landscape.
> The DiSo
> Project is doing some great work in creating a decentralised social
> network, by using a blog as the centre of our your existence. I think
> your idea is similar but much more interesting and open to the broader
> population - would be interested to see more of your thoughts.
>
> But the reality is, companies are a stakeholder, just like the
> standards communities are, users, governments, and the rest. You can't
> ignore them.
I wouldn't suggest ignoring them; however, most seem too slow,
misguided, or too narrowly focused, so I see a solution to come more
likely from the bottom-upwards rather than from the top. Mostly I
suspect they'll become irrelevant as more compelling ideas are hacked
into the internet from below.
>> The possibilities for data portability are way too big and interesting
>> and corporations simply will NOT be the innovators in this area, isn't
>> this obvious? So don't sell your soul just because you get a little
>> attention from the rich and famous. Though Google still seems like
>> the best partner so far. Anyone hiring out there?.
>
> I totally agree. But for different reasons - having seen it first hand
> myself, companies by design struggle to innovate. Despite all that is
> touted by Google, they suffer they same corporate bullshit the rest of
> the world suffered by their internal entrepreneurs. I would argue the
> reason Google has been so innovative, as a lot to do with the fact
> they've made some smart acquisitions and brought fresh blood in to run
> new product lines which gives them some freedom.
...which is why I believe an open-source/GPL p2p app for social
networking is likely to produce the most innovation. Fresh blood is
much easier to come by there. What's missing is compelling vision in
which the hackers can organize around and believe in so that their
skills and creations aren't just going to be sold out. I feel like
I'm 2/3s of the way there, but am lacking a community around me to
make that happen. I was disappointed in the direction that
data-portability took in response to Aaron Cheung's (sp?)
hyper-rational insistence on establishing order and "chains of
responsibility". The field is too new. Order shouldn't be
established prematurely and sometimes anarchy is actually the best
transitional [non-]governance model, in order to allow sufficient
freedom so that all the different ideas and trajectories can fall into
THEIR OWN natural order and balance--not unlike how water undergoes a
chaotic phase-transition when it turns from water to ice--forming and
dissolving patterns of order in a sort of limbo-transitional period
involving millions of interactions until an actual crystal of order is
formed. Like all those bouncing, chaotic water molecules, we don't
know in advance whether the final result will be hexagonal, cubic, or
something else entirely.
> Thank you for contributing. Happy to chat with you off this mailing
> list on your thoughts. We need more people like you to voice their
> thoughts - whether on this mailing list, in the blogopshere, even on
> Twitter. That's how stuff gets done - you start with a conversation,
> you identify, and then you work towards.
Your encouragement is appreciated. It communicates to me that it's
still worthwhile to monitor this project and discussion board.
Thanks,
marcos
Elias Bizannes
> data-portability took in response to Aaron Cheung's (sp?)
> hyper-rational insistence on establishing order and "chains of
> responsibility". The field is too new. Order shouldn't be
> established prematurely and sometimes anarchy is actually the best
> transitional [non-]governance model, in order to allow sufficient
> freedom so that all the different ideas and trajectories can fall into
> THEIR OWN natural order and balance--not unlike how water undergoes a
> chaotic phase-transition when it turns from water to ice--forming and
> dissolving patterns of order in a sort of limbo-transitional period
> involving millions of interactions until an actual crystal of order is
> formed. Like all those bouncing, chaotic water molecules, we don't
> know in advance whether the final result will be hexagonal, cubic, or
> something else entirely.
*All* the active contributors of the DataPortability Project in the
earlier days demanded a governance framework after our anarchy caused
too much frustration. We thought (the founders) we could experiment
with a new social organisation that was decentralised and
non-heirarchical, but it caused problems - as simple as it sounds
"making a decision" could not be done in the earlier days.
I remember once someone in London decided on something on a thing we
were working on; then when the person in San Francisco got into gear,
made a different decision. They contradicted each other, they confused
external parties they contacted, and there was no way to resolve what
to do. People in Australia did something with the Americans whilst the
Europeans where sleeping - and then the American's would hit bed time
and the Europeans woke up - only to be fuming how could that decision
be made without their consultation. So then for a decision to be made,
you had to wait a few days to make sure everyone knew about it - and
seriously, if I want to change a colour in the CSS for the homepage,
do I really need to wait a few days for that?! You had half the group
being a bunch of mavericks saying "just do it" which is a great
attitude but sucks when you have dozen people "just doing it" on their
own accord as its just a bunch of shooting stars. And another dozen
fuming that they were not consulted. When your actions represent a
group of people, you can't "just do it" because it might not reflect
the groups will and which happened several times (requiring some
damage control).
In another example, everyone wanted something but the people who
could ultimately do this, did not. So you had one or two people with
the power (without that power having credibility), deciding on
something another dozen people wanted. People throw eggs at us for
coming up with a bureaucracy, but to be honest, I'd rather have yoke
on my face than go through that painful experience earlier this year
where we could not move forward.
The DataPortability Project has a unique history that I think makes it powerful.
1) Started as workgroup. Closed, invite only. Discussions evolved,
with multiple initiatives and directions proposed.
2) Received massive hype, despite just starting out. Flood of new
people joining outside of this workgroup (on this mailing list).
Workgroup disbanded, wikiocracy model tried (which was basically the
anarchy model you describe). Action groups created with specific
functions to achieve goal; groups organised within to execute that
function - evangelism, technical, implementation, policy, steering.
Decentralised functions theoretically were to make the project work.
Decisions to be made by "consensus".
3) Several months later, the core contributors started proposing
better process. Led to community of equals to draft a governance
structure. Over a hundred hours of back and forth went into the fine
tuning - as there was little precedent for our situation: global group
in a score of cities with no previous relationships binding them just
passion, online community not physical, no central authority other
than the founders who guided it. It was difficult to not let this run
the track of typical online communities which have a benelovent
dictator to make it function. We had a group of people making
decisions on how to make decisions despite not having an accepted way
yet on how to make decisions!
So our history has a unique DNA that has created a unique organisation
which may have cost us several months in progress, but in the longer
term makes us a hell of a lot stronger.
We have a steering group elected by the open community, to protect
ultimately the community. DataPortability is a brand and a community;
and steering protect that brand from misuse whilst incubating the
community to achieve our goals. We have a framework on how to achieve
our goals, but we don't have a forced method on what those goals are.
Give it a go - propose an idea and you'll see our model execute. If
it's a bad idea, the community will reject it and it will die
naturally (that's here); if it's a good idea (like the health care
discussions) - the process kicks in like it recently did for health
care, with a clear process to achieve the goal (defined goals,
identified process to execute, fixed time-frame, accountability of
work, support with resources if needed).
Bryan Bishop
> ...which is why I believe an open-source/GPL p2p app for social
> networking is likely to produce the most innovation. Fresh blood is
> much easier to come by there. What's missing is compelling vision in
> which the hackers can organize around and believe in so that their
> skills and creations aren't just going to be sold out. I feel like
> I'm 2/3s of the way there, but am lacking a community around me to
> make that happen. I was disappointed in the direction that
> data-portability took in response to Aaron Cheung's (sp?)
> hyper-rational insistence on establishing order and "chains of
> responsibility". The field is too new. Order shouldn't be
> established prematurely and sometimes anarchy is actually the best
> transitional [non-]governance model, in order to allow sufficient
> freedom so that all the different ideas and trajectories can fall into
> THEIR OWN natural order and balance--not unlike how water undergoes a
> chaotic phase-transition when it turns from water to ice--forming and
> dissolving patterns of order in a sort of limbo-transitional period
> involving millions of interactions until an actual crystal of order is
> formed. Like all those bouncing, chaotic water molecules, we don't
> know in advance whether the final result will be hexagonal, cubic, or
> something else entirely.
Hello, my name is Bryan, first time poster (or I haven't posted in a
good time here), I'm found more and more contributing to discussions
on open source manufacturing and automation discussion groups more
than anything else, but I do still pay attention to the computational
social networking issues of our time. I was approached by a banking
initiative not too long ago that was interested in micropayments and
open monies, and while I'm hardly the one to approach about 'money', I
did propose to them an interesting host-your-own-Apache setup for
distributed p2p social networking as well as loaning. The loaning
stuff is rather insignificant in the scheme of things, especially
since very little code was written overall. But the important concept
of having a low-overhead 'web' server has been realized time and time
again on the web, as well as different formats for expressing content
that people find important. Go forth and take from KDE's address book
system the formats for storing contact information, and from the RSS
model of pushing updates to subscribers that ping a request for news
for family members; also include FOAF XML and so on. This isn't
technically hard at all, what it takes is just a weekend of some
programming and then showing something interesting happening with this
sort of system. It's not quite enough to say "Hey look, I have an
address book" any more (was it ever?), and it's not enough to just
coat it with terms about how it's more social than otherwise kneeling
to the centralized servers. But with initiatives like surfraw and the
other more web2.0 equivalents of it which are unfortunately presently
centralized though have no reason to be (the ones that spider with
your usernames and passwords to aggregate information (this should be
done on your localhost)), maybe that could be interesting to some
people wanting to have more options with their own data. Then, an
exposed API platform might spur off some quick programming and people
might get excited about developing little plugins and extensions, but
that's not much to bank on. If you want to just go for the long haul,
writing the code, releasing the server and a few file readers will be
enough, but nobody will have the software installed at first, making
it ridiculously hard to get any sort of concentrated social value
immediately. Good luck. :-)