Data Portability and Privacy Policies
Bob Janacek
Email addresses and privacy policies often intersect. So when taking my social graph from one network to another, it should also include the email addresses of my contacts. But after some inquiries with a few social networks, the email addresses of members, even those in my social graph, seem to be restricted.
Any thoughts on this portability dilemma?
--bj
Jonathan Vanasco
should also include the email addresses of my contacts."""
You're not privileged to see/search by email address on many social
networks.
Just because I add you as a friend / contact on a Social Network
doesn't mean that I've decided to allow you access to my personal
information or email address.
On some networks I do, on many/most I do not.
I've actually been working on a long essay covering this. I should
finish it up this weekend.
Christian Scholz / Tao Takashi (SL)
<jona...@findmeon.com> wrote:
>
> """So when taking my social graph from one network to another, it
> should also include the email addresses of my contacts."""
>
> No, it absolutely shouldn't.
Indeed, it might be included if I allow the service to give it out to
you. I would like to be able to define which service gets to see it
and which user of which service (applies to every data I put
somewhere), even on the service itself.
-- Christian
--
Christian Scholz
Tao Takashi (Second Life name)
taota...@gmail.com
Blog/Podcast: http://mrtopf.de/blog
Planet: http://worldofsl.com
Company: http://comlounge.net
Tech Video Blog: http://comlounge.tv
IRC: MrTopf/Tao_T
Elias Bizannes
This is is very much at the heart of the debate between Scoble and
Arrington a month ago. The answer is more about philosophy than
technology. I've previously written on the importance of protecting
your e-mail[1], but I still don't know what the right answer is. I
would be interested to see more discussion about this simple but
complicated point.
[1]http://liako.biz/2007/11/facebooks-privacy-is-smart-on-technology-
but-stupid-in-thought/
On Jun 28, 7:23 am, "Christian Scholz / Tao Takashi (SL)"
>
> > """So when taking my social graph from one network to another, it
> > should also include the email addresses of my contacts."""
>
> > No, it absolutely shouldn't.
>
> Indeed, it might be included if I allow the service to give it out to
> you. I would like to be able to define which service gets to see it
> and which user of which service (applies to every data I put
> somewhere), even on the service itself.
>
> -- Christian
>
> --
> Christian Scholz
> Tao Takashi (Second Life name)
Bob J
Scenario 1: Mary migrates her social graph from social network LAST to
social network NEXT, and friends in her social graph are not already
members of NEXT.
Although Mary could easily communicate with friends on LAST, using
NEXT, she would only be able to invite those friends that have email
addresses listed in her NEXT social graph. That seems like a data
portability hurdle. It creates a big barrier to switching between
social networks, but is essential to the privacy of a social network's
members.
Scenario 2. In the past, Mary invited non-members via their email
addresses to join social network LAST.
When those non-members joined LAST, would their email addresses
automatically become part of Mary's social graph? After all, Mary was
the one that provided those email addresses to the social network.
Bye for now. Time to get a bagel and coffee, and ponder some more.
-bj
Elias Bizannes
Isn't that a form of spam? I think it would be more effective if Mary
could automatically add her friends - no notification, no requests, no
e-mails being sent. The identifiers are behind the scenes.
Scenario 2: How can you enforce something when you have no control
over it? For example, I "own" my name - but that's not going to stop
you from using my name in a way that is a misrepresentation of my
opinions. This is different from how I "own" my blogs domain name - I
can enforce how representations of my opinions occur on that ie, I can
edit your comments.
So going back to e-mails, we can't enforce what people do with them
(addresses that is, not accounts). However, just like you wouldn't go
around calling me a Nazi which is contrary to my political views, you
wouldn't use my e-mail (if you had it) in a way that is just 'wrong'.
Not as to the point whether applications should give you data that you
submitted, I believe for DataPortability to be practical, it's not
about "physically" porting your (ahem, virtual) data; rather it's
about porting the context of that data's usage. So yes, Facebook has e-
mails of 200 people I submitted when looking for friends when I joined
most of whom are now members; but because not even the owner of that e-
mail address can't control how its used, neither should you dictate
that you own it. However, you should still get the benefit of using
that data which is access - and in this scenario, it's facebook
talking to another application that you indeed have a relationship
with som...@email.com - you don't move the data, but you get the
benefits of it because it helps rebuild your social graph on another
application.
DataPortability should be about controlling the benefits of the usage
of that data. Who cares about ownership when all we want is control
(I'm willing to not have to fight over legal title, as it's control
that's ultimately appealing about ownership). And who cares about
controlling how data is used, when ultimately, all we want are the
benefits surrounding that data. As long as another application gives
me the benefits of using that data for a purpose I nominate (ie,
finding new friends by using an e-mail as an identifier) then that is
all I need.
Bob Ngu
use case between Scoble and Arrington
http://ungeekdapo.wordpress.com/2008/05/19/my-data-and-having-it-my-way/
My take on it, user privacy is a personal and individual thing and is
also dependent on the context of usage.
Brady Brim-DeForest
Christian Scholz / Tao Takashi (SL)
On Sat, Jun 28, 2008 at 6:43 PM, Brady Brim-DeForest <bra...@gmail.com> wrote:
> I see it this way: Once I 'confirm' you as a friend, I am granting you
> access to any data in my profile (within the context of social networking).
> That means that that relationship is owned by both you and me, and that the
> communication methods (like email addresses) needed to facilitate that
> relationship, essentially now belong to us both.
I see it that way:
When I confirm you as a friend (or something like a friend ;-) ) on a
social network then I want to define what data is available to you
(Xing allows much of that). This does mean that I give a (revokable)
license to you to access or maybe also export that data.
When I sign up to a new social network I also would like to grant a
(revokable) license to that service to use my data (which I either
enter or let it subscribe to from another service). The license will
include what the service is allowed to do with it (use it for ads,
give out to third parties, let people export it, let people
annotate/comment on it, etc.)
An interesting part might be annotation/comments, here I might give a
license when annotating an asset by another user which is maybe not
revokable and allows him to move that data with the object.
BTW, some people at the re:publica in Berlin and at IdentityCamp in
Bremen were discussing an expiration date for data. So after amount X
it should disappear again (think of all the photos of you when you
were drunk). I don't think something like this will ever work
technically but also don't think it's a good idea anyway ;-) Just
wanted to mention it.
-- Christian
> -Brady
> --
> Brady Brim-DeForest
> www.brimdeforest.com
>
> On Sat, Jun 28, 2008 at 9:25 AM, Bob Ngu <bob...@yahoo.com> wrote:
>>
>> Good question, I blogged about this exact scenario, highlighting the
>> use case between Scoble and Arrington
>> http://ungeekdapo.wordpress.com/2008/05/19/my-data-and-having-it-my-way/
>>
>> My take on it, user privacy is a personal and individual thing and is
>> also dependent on the context of usage.
>>
>> On Jun 26, 10:16 am, "Bob Janacek" <b...@certifiedmail.com> wrote:
>> > Email addresses and privacy policies often intersect. So when taking my
>> > social graph from one network to another, it should also include the
>> > email addresses of my contacts. But after some inquiries with a few
>> > social networks, the email addresses of members, even those in my social
>> > graph, seem to be restricted.
>> >
>> > Any thoughts on this portability dilemma?
>> >
>> > --bj
>>
>
>
>
>
> >
>
--
Christian Scholz
Tao Takashi (Second Life name)
Bob Ngu
> (revokable) license to that service to use my data (which I either
> enter or let it subscribe to from another service). The license will
> include what the service is allowed to do with it (use it for ads,
> give out to third parties, let people export it, let people
> annotate/comment on it, etc.)
the Scoble/Arrington problem, details extracted from my post
Let’s take the hotly debated example between Scoble and Arrington.
Arrington’s position is that his data is his data and he should have
the ultimate control over how that data is used which includes the
ability to stop someone like Scoble with whom he has shared his email
with to not use it on 3rd party services like Plaxo without
Arrington’s explicit permission. Scoble’s counter argument is that
once Arrington shares his email that he has given implicit permission
for Scoble to use it elsewhere. He cited the examples of adding
Arrington’s email to his Gmail and Yahoo mail accounts so he can email
Arrington from either email account and Arrington has no issue with
that. However, Arrington has an issue if Scoble wanted to expose
Arrington’s email to a 3rd party service like Plaxo which according to
Arrington has a tendency to spam people. If Scoble doesn’t care that
Arrington use his email on Plaxo, then that’s Scoble’s right to not
care, but if Arrington has an issue with Scoble using his email on
Plaxo, then that’s Arrington right to care. Note that Arrington’s
reaction is context based because he has no problem with Scoble adding
his email to Gmail and Yahoo accounts, which brings me back to my
point about user privacy being a personal and individual thing and is
potentially limitless examples, so it wouldn't make sense to define
privacy controls for limitless one-offs. I don't have a solution
though, hence why I am suggesting that its time to define open privacy
standards much like the approach taken by OpenID for identity and
OAuth for authentication/authorization. Let’s start with something
small and tangible for 1.0 version and build on it based on real-world
usage on what works and what doesn’t work, essentially going it the
agile way.
Is anyone else interested in such an effort? If so, let me know here
or drop me a comment on my post.
http://ungeekdapo.wordpress.com/2008/06/23/time-to-define-open-privacy-standards/
On Jun 28, 11:51 am, "Christian Scholz / Tao Takashi (SL)"
<tao.taka...@googlemail.com> wrote:
> Hi!
>
> >> Good question, I blogged about this exact scenario, highlighting the
> >> use case between Scoble and Arrington
> >>http://ungeekdapo.wordpress.com/2008/05/19/my-data-and-having-it-my-way/
>
> >> My take on it, user privacy is a personal and individual thing and is
> >> also dependent on the context of usage.
>
> >> On Jun 26, 10:16 am, "Bob Janacek" <b...@certifiedmail.com> wrote:
> >> > Email addresses and privacy policies often intersect. So when taking my
> >> > social graph from one network to another, it should also include the
> >> > email addresses of my contacts. But after some inquiries with a few
> >> > social networks, the email addresses of members, even those in my social
> >> > graph, seem to be restricted.
>
> >> > Any thoughts on this portability dilemma?
>
> >> > --bj
>
> --
> Christian Scholz
> Tao Takashi (Second Life name)
subsy...@mac.com
> offs.
Starting with a use case is extremely valuable, even if it is one of
an infinite number. Most important note for me "user privacy [is] a
personal and individual thing and is also dependent on the context of
usage."
If everything operated from the perspective of the individual (part of
the work we are attempting with VRM), then these personal privacy
settings and control of data in context would be entirely possible.
While we are discussing layering control on top of existing networks,
we should also be discussing building a network (or a blueprint for
networks) that supports social networks and user privacy/sharing
concerns at from underneath -- kind of like social network plumbing.
Wouldn't it be nice if:
* users who were more promiscuous could set their base profile to wide
open - in other words, yeah I am interested in going with you to other
networks, sure send me an email
* users who want to be notified and make a choice when certain uses of
their information happen could have this
* users who wanted to unlink with a social network, or change privacy
settings with another individual or network, or across the board -- np
* users who were more privacy minded - explicitly set relationship
parameters down to the attribute shared level.
* share it all here, share none of it there, then change your mind
tomorrow
Your choice baby!
asa
On Jun 28, 2008, at 5:21 PM, Bob Ngu wrote:
>> details extracted from my post:
>
> Let’s take the hotly debated example between Scoble and Arrington.
> Arrington’s position is that his data is his data and he should have
> the ultimate control over how that data is used which includes the
> ability to stop someone like Scoble with whom he has shared his email
> with to not use it on 3rd party services like Plaxo without
> Arrington’s explicit permission. Scoble’s counter argument is that
> once Arrington shares his email that he has given implicit permission
> for Scoble to use it elsewhere. He cited the examples of adding
> Arrington’s email to his Gmail and Yahoo mail accounts so he can email
> Arrington from either email account and Arrington has no issue with
> that. However, Arrington has an issue if Scoble wanted to expose
> Arrington’s email to a 3rd party service like Plaxo which according to
> Arrington has a tendency to spam people. If Scoble doesn’t care that
> Arrington use his email on Plaxo, then that’s Scoble’s right to not
> care, but if Arrington has an issue with Scoble using his email on
> Plaxo, then that’s Arrington right to care. Note that Arrington’s
> reaction is context based because he has no problem with Scoble adding
> his email to Gmail and Yahoo accounts, which brings me back to my
> point about user privacy being a personal and individual thing and is
> also dependent on the context of usage.
>
> Is anyone else interested in such an effort? If so, let me know here
> or drop me a comment on my post.
> http://ungeekdapo.wordpress.com/2008/06/23/time-to-define-open-privacy-standards/
--
Asa Hardcastle, Technical Lead, openLiberty ID-WSF ClientLib
Tel: +1.413.429.1044 Skype: subsystem7
Bob Ngu
>
> * users who were more promiscuous could set their base profile to wide
> open - in other words, yeah I am interested in going with you to other
> networks, sure send me an email
> * users who want to be notified and make a choice when certain uses of
> their information happen could have this
> * users who wanted to unlink with a social network, or change privacy
> settings with another individual or network, or across the board -- np
> * users who were more privacy minded - explicitly set relationship
> parameters down to the attribute shared level.
> * share it all here, share none of it there, then change your mind
> tomorrow
>
excerpt from my post...
A service should implement privacy controls ranging from air tight to
I honestly don’t care who sees and uses my data along with a set of
sensible defaults, e.g.,
1. I am a private person and I want to explicitly approve every use
of my data, WARNING: this might result in excessive permission
requests but it’s what you asked for, think Microsoft Vista User
Account Control feature.
2. I am ok with sharing some but not all of my data with my friends
and the world. Here’s where the sensible defaults come in, the
challenge is to define a set of sensible defaults that makes sense for
a particular service but needs to be contextual aware or requires
contextual approval.
3. I love attention and frankly don’t care who sees and uses my
data. WARNING: you might regret this later on when you run for the
post of US President or mayor when there are bikini pictures of you on
the web but that’s a risk you accept.
I suspect most users will pick option 2, and that’s where the real fun
begins, defining sensible defaults for each service while allowing for
contextual awareness or approval - that's as far as I got with a
proposal.
The work with VRM sounds it might be applicable, can you provide more
details, a succinct summary will be great ^_^
jtfmulder
online friends. I also import my contacts from my address book to
Gmail.
I like it when I join a NEW Social Network and they ask if you want to
"find your friends that already are on the NEW Social Network?" I
always say "yes" to this. I have no problem with adding an online
friend(s) from MySpace, Twitter, Flickr, Facebook, etc. to yet another
new service that I've decided to join. When this option is available I
think that it is a perfect example of Data Portability across Social
Networks.
I have a lot of online friends from various Social Networks that I
have no idea what their email address is. Another problem that I run
into is I have various interests. A friend that likes 1980s music may
not be at all interested in Web 2.0, for example. If I have their
email address in my contacts list, I can invite them to join a new
service, but they may not be interested. Even worse, they may consider
invitations to new services or social media apps to be SPAM
Jonathan Vanasco
a social network doesn't mean that I'm privileged to see their email
address, or will allow them to see mine.
This isn't a privacy policy or corporate concern, this is a user based
concern and one of culture. There is a large difference between an
addressbook with emails and a social network.
People in the Data Portability movement - who are consummate social
networkers and often public figures - need to remove themselves from
examples and start looking at the usage patterns and concerns of the
general public.
MySpace is the top social network in the world. It gained popularity
because people could quickly add bands and make new friends across
groups and comments pages. People actively befriend hundreds of
accounts, some known, many unknown, creating new relationships.
Do you think this activity would go on if it meant sharing your email
address and personal contact information with every person you add?
Absolutely not.
Being a friend or contact on one social network doesn't equate to
being a friend/contact in real life, wanting to share your email
address/universal identifier, or wanting your identity exposed on
other websites.
I'd love the ability to add my MySpace friends onto Facebook and vice-
versa -- but there is no way in hell I could ( or anyone else should )
support any sort of technology that automatically exposes my Facebook
ID to the entirety of my MySpace friends, or shows me how to find the
definitive matches to my Facebook list on MySpace.
Exposing your social network IDs is a common-sense user privacy issue
-- not a corporate privacy policy one.
Loïc DIAS DA SILVA
Facebook and others are like towns, your facebook profile is as your home.
You can have a primary home and several second ones and perhaps sometimes you need to change of primary town, building another, dataportability is made for this in my point of view.
As your have homes, you move during the day to school, office, swimming-pool or sports, which are some flickr, del.icio.us and others social graph consumers.
One concept could be introduced to achieve this, a kind of virtual profile.
If i want to manage my social graph from plaxo instead of facebook the operation could looks like this :
0/ I have a facebook account
1/ I create a Plaxo account
2/ Using facebook apps and API, Plaxo can retrieve my friends facebook ids
3/ Plaxo try to identify if some of my friends have yet an account on its database reading name, nickname or previoulsy created Plaxo<->Facebook attachment.
4/ If a link could be found, a Plaxo social attachment is created between me and the others existing Plaxo accounts,
5/ If a link could not be found, a 'virtual profile', only readable by me is created inside Plaxo database who play a role of proxy for this user, this account uses my facebook account to proxy requests.
6/ dataportability ensure then that if i delete the link from Plaxo it will also be deleted on facebook. fb account is not used anymore for my social graph management, it becomes a 'virtual profile' only used as a proxy.
My 'dead' account on facebook is used by my friends as before to transparently see my activity, perhaps my social graph is now managed by Plaxo, my photos by Flickr, my feed by friendfeed and my status by Twitter but my friends can't see this, they only see a normal facebook account. Yes i can't share with us the latest facebook killer app, but if i am not in the same town as my friends i can't go with us to their newly created attraction parc...
Their 'non existent' accounts on Plaxo are used by me to see their activity on facebook but i have not to manage this, i only see a normal Plaxo account.
So i have not to bring their emails with me to move from a network, a town, to another network, my destination town. I have not to ask the question of the email owning...
There must be a common framework to share common services like notes, medias, messages, presence, feed, etc.. But there are yet many initiatives in this way aggregated by DP.org.
Instead of Plaxo if i wanted to move to littlesocialnetwork.com, this kind of features can be not implemented, but if i want to move from NY to a little town without airport and roads the only way of communication we have to talk together is the phone or perhaps only postal mails.. DP works to have a common air backbone.. :)
What's the bad trick about this view ?
(Sorry for my english if i can't be understood, i can make schemas.. ;) )
Email addresses and privacy policies often intersect. So when taking my social graph from one network to another, it should also include the email addresses of my contacts. But after some inquiries with a few social networks, the email addresses of members, even those in my social graph, seem to be restricted.
Any thoughts on this portability dilemma?
--bj
Christian Scholz / Tao Takashi (SL)
On Mon, Jun 30, 2008 at 8:58 AM, Jonathan Vanasco <jona...@findmeon.com> wrote:
>
> Let me reiterate my point - just because I'm contacts with someone on
> a social network doesn't mean that I'm privileged to see their email
> address, or will allow them to see mine.
that should apply to any data of course. I would like to define either
profiles or filters or whatever you want to call it and define for
either a service or individual users what they are allowed to see.
> This isn't a privacy policy or corporate concern, this is a user based
> concern and one of culture. There is a large difference between an
> addressbook with emails and a social network.
Yes, esp as the concept of a friend or contact is a rather complex
one. There are many shades between all those terms and I usually want
to let certain people let know more about me (or contact me) than
others.
> People in the Data Portability movement - who are consummate social
> networkers and often public figures - need to remove themselves from
> examples and start looking at the usage patterns and concerns of the
> general public.
Well, I see this every day in Second Life (I know you don't seem to
like it but it's nevertheless some amount of a general public out
there to which this should apply). Many people in SL go a long way to
prevent people from matching their SL account with their RL identity.
Except of course maybe for their closest friends. Unfortunately right
now they have no means to specify this in their profiles. Everything
can be seen by everybody. This needs to be changed (and I hope to
influence the Architecture Working Group a little to make this better
in the upcoming version of the protocol).
> MySpace is the top social network in the world. It gained popularity
> because people could quickly add bands and make new friends across
> groups and comments pages. People actively befriend hundreds of
> accounts, some known, many unknown, creating new relationships.
>
> Do you think this activity would go on if it meant sharing your email
> address and personal contact information with every person you add?
> Absolutely not.
Right. Or maybe even your realname.
> Being a friend or contact on one social network doesn't equate to
> being a friend/contact in real life, wanting to share your email
> address/universal identifier, or wanting your identity exposed on
> other websites.
Right, there are many different relationships you can have with your
contact list. It would be great if even social networks internally
would give me more options on defining certain datasets for certain
contacts. Xing does that quite well IMHO, I can define for every
contact what profile fields are visible, for convenience they are
groupes into personal/business/IM and so on so I can enable/disable a
complete group or can enter it and do it individually.
> I'd love the ability to add my MySpace friends onto Facebook and vice-
> versa -- but there is no way in hell I could ( or anyone else should )
> support any sort of technology that automatically exposes my Facebook
> ID to the entirety of my MySpace friends, or shows me how to find the
> definitive matches to my Facebook list on MySpace.
>
> Exposing your social network IDs is a common-sense user privacy issue
> -- not a corporate privacy policy one.
This also applies to OpenID btw as this sometimes acts as such an ID.
Using this to sync friends list might be easy but defeats the privacy
point. I would like to be able to use different openids on different
systems but still have the possibility to sync contact lists and data
across various networks.
-- Christian
--
Christian Scholz
Tao Takashi (Second Life name)
Christian Scholz / Tao Takashi (SL)
On Tue, Jul 15, 2008 at 4:17 PM, Loïc DIAS DA SILVA <mgl...@gmail.com> wrote:
> But why do we need to move our social graph from one network to another ?
> When you move from a town to another, you don't ask your friends to move
> with you.
That analogy is not the best one IMHO. Asking friends in RL to move
with you will not work because of many constraints they have (work,
family, other friends of them who then also need to move etc. ).
Nevertheless you probably would like them to move with you. Now with
social networks they theoretically can. And many people want that. So
IMHO that's reason enough to work on it.
> Facebook and others are like towns, your facebook profile is as your home.
> You can have a primary home and several second ones and perhaps sometimes
> you need to change of primary town, building another, dataportability is
> made for this in my point of view.
See, even in virtual worlds like Second Life which maybe more are an
analogy to your RL we are working now on making it possible to easily
move between system boundaries in a distributed virtual world where
many people can host servers (both for simulating it and hosting
users). The goal here is not to be a tourist or so on a different
server but really have a one-world experience. This would map to the
actual need to sync data between providers and technically that's not
much different from what we are thinking about in terms of social
networks (after all it's also just profiles, friends, groups and so
on).
> One concept could be introduced to achieve this, a kind of virtual profile.
> If i want to manage my social graph from plaxo instead of facebook the
> operation could looks like this :
> 0/ I have a facebook account
> 1/ I create a Plaxo account
> 2/ Using facebook apps and API, Plaxo can retrieve my friends facebook ids
> 3/ Plaxo try to identify if some of my friends have yet an account on its
> database reading name, nickname or previoulsy created Plaxo<->Facebook
> attachment.
> 4/ If a link could be found, a Plaxo social attachment is created between me
> and the others existing Plaxo accounts,
> 5/ If a link could not be found, a 'virtual profile', only readable by me is
> created inside Plaxo database who play a role of proxy for this user, this
> account uses my facebook account to proxy requests.
> 6/ dataportability ensure then that if i delete the link from Plaxo it will
> also be deleted on facebook. fb account is not used anymore for my social
> graph management, it becomes a 'virtual profile' only used as a proxy.
That sounds better :-) And maybe I misunderstood you. My general idea
about all this would be:
1. I have some "master" server which holds my data. It's one I trust.
Might be Facebook but probably I will host my own
2. I sign up to a new service and give the service a license which
defines how to use my data.
3. When I think that new service sucks I can revoke this license and
the services is then asked to delete all cached data etc.
The same can apply on a per user basis. I simply give a certain
license to that user and also can revoke it again.
The license has the advantage that not the service defines what they
(and myself) are allowed to do with my data but I define. It's how it
should be.
The license should also be easy to understand like Creative Commons is
so that people do not need to understand all those EULAs.
The license could contain such definitions as "is allowed to see my
data", "is allowed to export my data", "is allowed to annotate on my
data (leave comments)".
Something like this was discussed in the realm of dataportability (I
at least remember one discussion about it at one DP meetup in Second
Life with Steve Greenberg) and I hope we can continue on this. :-)
Loïc DIAS DA SILVA
In this same example, could i navigate through my complete social graph with fb, including contacts only present on SL ?
Christian Scholz / Tao Takashi (SL)
On Wed, Jul 16, 2008 at 2:42 PM, Loïc DIAS DA SILVA <mgl...@gmail.com> wrote:
> Thanks for your answer Christian, this is a very interesting way of thinking
> (license), but following my understanding of what you say, it do not explain
> what is exactly the job of this master server but the data storing and
> management. Is this server also forwards the services stuff as messages,
> media and others ?
Well, I think it is hard to build all the control in a decentralized
way, it's easier to have one master server (basically the openid
model) to which services and maybe users then subscribe. There should
be some attributes in place which define how long external services
are allowed to cache it and so on.
But all of this is really just spontaneous thinking. One question
might e.g. be how you handle different personalities. And if you do so
how do you allow certain people to know about them but others not.
> If my data are stored on a personal server for example and i sign up to
> facebook, fb retrieves my data and my social graph which can contains
> contacts from FB, others from SL, emails contact list, etc.. and manages
> with this, allowing data communication or, another solution, it forwards to
> my server all services content.
>
> As a example, is from facebook i want to send a message to one of my contact
> which is only present into SL, is it the job of fb backend to relay the
> message to SL or the route contains a point of forward with my data server ?
> In this same example, could i navigate through my complete social graph with
> fb, including contacts only present on SL ?
Well, the main idea probably is to move the social graph outside those
networks. Networks just have some sort of projection depending on who
of my contacts is actually on that social network. My master server
might know about all of them but only as much as they allow me to see.
It might also know which person is which user on which social network
so it then knows on how to route the message. But you are right, it's
still the question how facebook then knows how to route the message.
So I guess this needs more thinking ;-)
> Finally i am not talking about data synchronization but more about data
> presentation (fb will not own all the content data of my contacts but only
> their services subscriptions, it shows me their activities on the others
> services). I am not sure to be clear.. ;(
So you are talking about aggregation basically? Something what friendfeed does?
Loïc DIAS DA SILVA
> > presentation (fb will not own all the content data of my contacts but only
> > their services subscriptions, it shows me their activities on the others
> > services). I am not sure to be clear.. ;(
On any social service i can have an access to all my contacts... IMHO it is the way to have only one social graph not depending to the site where i am browsing.
Christian Scholz / Tao Takashi (SL)
On Wed, Jul 16, 2008 at 5:20 PM, Loïc DIAS DA SILVA <mgl...@gmail.com> wrote:
>> > Finally i am not talking about data synchronization but more about data
>> > presentation (fb will not own all the content data of my contacts but
>> > only
>> > their services subscriptions, it shows me their activities on the others
>> > services). I am not sure to be clear.. ;(
>
>> So you are talking about aggregation basically? Something what friendfeed
>> does?
>
>
> Yes, exactly, but is is not basic aggregation : i can see my flickr photos
> as the facebook photos, it is my photos !
I guess this should be possible if there would be a generic standard
for accessing this information (as you maybe don't want a special
solution for flickr, one for photobucket, one for ...). The main issue
might of course be political. Does flickr would like that?
And I think FB is also a special case here as it's a social network
which really implements everything (or let other people do that via
applications). In the future I would see it more the way that there
are individual small components, small social applications like maybe
dopplr or twitter. They have a limited functionality but they
implement it well. If the social graph would not be stored anymore on
every of these services but they merely subscribe to your social graph
(or that portion they are allowed to access) then there also wouldn't
be that syncing annoyance anymore.
OpenSocial of course seems to go the other way. But if those social
apps reimplement themselves as OpenSocial gadgets, too then they can
even move around from container to container.
The only question might be that of the business model but I guess it
will stlll be ads (although I also hope for more services offering
premium accounts instead as an option).
> I see all my contacts as a fb account, i am the only one to be able to see
> those who have not a real fb account, in this case their fb page is only an
> aggregation..
Well, they should agree though that they are aggregated by that service.
> In this way i don't only see the contacts of my social graph having a
> facebook account by i can see all my social graph.
I see your point. And with those mentioned social apps it is probably
a need to have some sort of overview over one person's stuff if that's
not stored all in one place anymore like on FB (which is the problem
today already). But I would rather see services like friendfeed taking
that place. In the end, is there a need for such broad social networks
like FB in the end? (not that I expect or wish them to go away but it
seems to be somewhat proven that decentralized approaches work
better).