Will data portability take as long as free software?
Aaron E. Klemm
It occurred to me that data portability is a movement as interesting as
free and open source software. I'm asking on my blog whether success
will take more or less time than what we see today in FOSS:
http://unripped.org/blog/2008/04/13/will-data-portability-be-a-battle-like-free-software/
I'm anxious to here your opinions.
Cheers,
Aaron Klemm
http://unripped.com
Kaliya *
Well since this movement is building on the hard work of people who started working on this from several angles since the beginning of 'the web' - talk to Doc Searls and Judith Meskill about their thinking/seeing an identity layer emerging relatively soon after the 'page web' emerge.
The Identity Commons was founded in 2001 (http://wiki.idcommons.net/index.php/History) specifically around the issues of people and their data and control of their information - it was a direct response to the 'liberty allicnace' that was a response to MSFT passport.
The Identity Commons today has a diverse community of technical, and social issues groups working on these problems.... the OSIS Interop being the most recent example with 50+ corporations and projects making user-centric identity and data sharing tools that are in relatively advanced prototypes. http://osis.idcommons.net - they just lead this Third major Interop at RSA a major security conference in SF http://www.rsaconference.com/security_topics/identity_and_access_management/blog.aspx?blogId=14645
It could be 'only 10 years off' and since things were obviously 'accelerating' and we know more about how to form these kinds of movement - maybe it is 5 years off.
I know from a conversation after the last SF meet-up there were some cautionary conversations had with "leadership of DP" about believing DP was right around the corner (like in the next few months) as opposed on the 18 months at the very earliest to very likely 5 years out. Not sure if this was actually "heard".
The conversation has been going on a long time - I know personally I have been working towards vision since I first read about it in 2003 and fully time and lucky enough to get paid to do so for much of the last 4 years - it will be a ways more - I have learned to be patient and invite others to dive in but also find the pace of the evolution happening and to engage with the longer term conversations that come 'before' this one along with understanding what this particular manifestation of the conversation can bring to and share with others.
There is not 'one answer' (no one blueprint) there is only a conversation - and in that relationship and through that implementation and collaboration. Fostering technical communities that thrive and work together is my passion and I know some amazing things are happening now that will get us closer to the vision of people being empowered with the ability to control their own data on the web.
if you want to have the conversation about this I invite you to think about coming to the data sharing events that build on the practice and traditions that we have in the identity community (personal identifiers and management of them is key to managing personal data in a distributed web - not just "porting it around") http://www.datasharingsummit.com
=Kaliya
Julian Bond
>the OSIS Interop
>being the most recent example with 50+ corporations and projects making
>user-centric identity and data sharing tools that are in relatively
>advanced prototypes.
Just reading the wiki, it seems like there's a lot more about InfoCard
than about OpenID and relatively few OpenID Relying Parties being
tested. And I'm curious to know how much of that testing is about data
sharing. "OpenID support for Simple Registration" and
"OpenID support for Attribute Exchange" do get a mention.
>I know from a conversation after the last SF meet-up there were some
>cautionary conversations had with "leadership of DP" about believing DP
>was right around the corner (like in the next few months) as opposed on
>the 18 months at the very earliest to very likely 5 years out. Not
>sure if this was actually "heard".
This depends so much on what measure you use to decide that we've got
there. There's a huge amount that can be done right now, and is being
done right now, to further the cause of data portability. From
implementing OpenID to marking up pages with microformats to
participating in the emerging standards. Just look at the progress
between last summer and now and the amount of code that has been written
and implemented in that time that's relevant to data portability.
--
Julian Bond E&MSN: julian_bond at voidstar.com M: +44 (0)77 5907 2173
Webmaster: http://www.ecademy.com/ T: +44 (0)192 0412 433
Personal WebLog: http://www.voidstar.com/ skype:julian.bond?chat
Made In Mexico
Elias Bizannes
Yes in the sense we may have some announcements soon (ie, this year)
of a technical implementation that could enable "DataPortability" in a
simple form that can be expanded on later.
No in the sense for a more sophisticated future of DataPortability,
which has since expanded this year due to the broader involvement of
the industry, it could take years.
The battle we are fighting is perception, culture, and attitudes - not
technological challenges. And I am specifically referring to the
original intent of DataPortability which was to promote open
standards, not some of the other issues that have since come under the
banner of DataPortability.
Cheers,
Elias
Kaliya *
technological challenges. And I am specifically referring to the
original intent of DataPortability which was to promote open
standards,
So why not stick with that.
Focus helps everyone understand you better and know how to interact.
Subsuming the world's problems under one banner makes it actually hard to solve.
not some of the other issues that have since come under the
banner of DataPortability.
For example?
Cheers,
Elias
Kaliya *
Kaliya * <identi...@gmail.com> Mon, 14 Apr 2008 00:09:06
>the OSIS InteropJust reading the wiki, it seems like there's a lot more about InfoCard
>being the most recent example with 50+ corporations and projects making
>user-centric identity and data sharing tools that are in relatively
>advanced prototypes.
than about OpenID and relatively few OpenID Relying Parties being
tested. And I'm curious to know how much of that testing is about data
sharing. "OpenID support for Simple Registration" and
"OpenID support for Attribute Exchange" do get a mention.
i-cards are actually an amazing set of tools that are more advanced then OpenID at doing data sharing about people between entities on the web.
The R-Card format that is evolving in Higgins being the primary one.
OpenID was not a super active participant in this round of interop's this is going to change but lets be clear that Attribute exchange is to simple to get where we need to go with intensive amounts of information and data exchange for people.
It is like there is this whole analysis of the things happening in the existing (4 year old) identity community that is like well it is not "perfect" yet rather then wow - things are a lot farther along then we knew and we could jump in and participate. Like joining a community with actual technical activity with actual major industry players that is actually getting work done.
(I did see Julian on the OSIS list engaging around the last interop - complaining about the lack of OpenID relying parties - a justifiable concern to raise).
This depends so much on what measure you use to decide that we've got
>I know from a conversation after the last SF meet-up there were some
>cautionary conversations had with "leadership of DP" about believing DP
>was right around the corner (like in the next few months) as opposed on
>the 18 months at the very earliest to very likely 5 years out. Not
>sure if this was actually "heard".
there. There's a huge amount that can be done right now, and is being
done right now, to further the cause of data portability.
Well maybe someone will actually listen to Phil's do a project a month idea - cause that is grounded and practical. So far I have heard quite a bit of pushback to this - saying it was not focused on the 'mega-big' problem - and that other groups like microformats should be pushing their own stuff they don't need DP do that or help. Until there is actual messaging clarity about actually 'doing' small 'doable' things then we can all think it is a good idea and DP still looks like it is not doing 'anything' cause they are waiting to solve the metaproblems.
Regards,
=Kaliya
http://www.identitywoman.net
http://wiki.idcommons.net
http://iiw.idcommons.net
http://www.datasharingsummit.com
Julian Bond
>OpenID was not a super active participant in this round of interop's
>this is going to change but lets be clear that Attribute exchange is to
>simple to get where we need to go with intensive amounts of information
>and data exchange for people.
I'd be happy to be proved wrong, but I suspect that OpenID AX is dead in
the water. This doesn't actually matter if some other more or less open
profile description can be bolted into the OpenID Signup/Signin process.
It doesn't even have to be right in OpenID provided it all happens at
the same time. Since XRDS is becoming common to find an OpenID provider,
the same XRDS could also point at the profile data that is transferred
via some other protocol. There are some arguments that the full Service
Catalogue should not be publicly accessible like this particular XRDS
file, but I'm not completely convinced.
>It is like there is this whole analysis of the things happening in the
>existing (4 year old) identity community that is like well it is not
>"perfect" yet rather then wow - things are a lot farther along then we
>knew and we could jump in and participate. Like joining a community
>with actual technical activity with actual major industry players that
>is actually getting work done.
Like, say, the OpenSocial Foundation and the work being done on creating
the OpenSocial RESTful Data API. (he said, deliberately picking an
example outside the Identity Community; Yahoo!, MySpace, Google and
Apache being fairly major players ;) )
>(I did see Julian on the OSIS list engaging around the last interop -
>complaining about the lack of OpenID relying parties - a justifiable
>concern to raise).
MyOpenID has a fairly long list of relying parties. In reality, pretty
much anyone implementing an OpenID consumer is going to check it against
Yahoo.com and MyOpenID. That gets you http, https, OpenID v1/v2 and
sreg. If that all works, you're probably ok. It still would have been
nice to see a bit more formal testing.
--
Julian Bond E&MSN: julian_bond at voidstar.com M: +44 (0)77 5907 2173
Webmaster: http://www.ecademy.com/ T: +44 (0)192 0412 433
Personal WebLog: http://www.voidstar.com/ skype:julian.bond?chat
Makes About 1 1/4 Cups Sauce