[KI-LC] Fwd: [Wg-consumer-identity] Consumer Identity WG: Next Steps and other Miscellanea

Bob Pinheiro kantara at bobpinheiro.com
Mon Aug 31 10:36:50 PDT 2009


I wanted to respond to your suggestion about the need for complex use 
cases, and pass it around to the other WG Chairs for comment, because I 
think it raises an important issue regarding interactions between the 
WGs, and how the WGs can jointly work together toward common goals.

Complex use cases such as the two examples you suggest (a car-buying 
scenario, and visiting a hospital emergency room in a distant city) 
would, as you have noted, bring into play issues that extend beyond a 
strict focus on consumer identity.  While consumer identity issues may 
focus primarily on identification or authentication (such as asserting 
an individual's identity to a relying party with high assurance), 
complex use cases would involve other issues that are being addressed by 
other WGs.  For instance, there are a host of privacy issues that may be 
relevant when sharing one's personally identifiable information with 
several service providers involved with buying a car (registering the 
car, getting insurance, applying for a car loan, etc).  So these privacy 
issues may be of interest to the Privacy and Public Policy WG.  And to 
the extent that a consumer is able to control the flow of their own 
personal information to various service providers, that may be of 
interest to the User Driven Volunteered Personal Information WG.

The hospital emergency room example raises a further issue: how can 
consumers authorize a hospital or healthcare provider to have access to 
relevant subsets of their online health information?  A piece of that 
problem may lie in consumer identity, to the extent that the consumer 
must first authenticate to some type of system prior to designating the 
appropriate authorizations for how their medical information can be 
accessed, and by whom.  But I believe that the way in which such 
authorizations could be implemented then becomes of interest to the User 
Managed Access WG, as well as the Healthcare Identity Assurance WG 
(which is also interested in authentication for access to medical records).

So I think the issue here is, do we strive to create complex use cases 
that extend across multiple WGs, and parcel out individual pieces of the 
use case to the individual WGs?  Or do we keep it simple and create use 
cases within each WG that are more focused and specifically address 
issues relevant to each WG?

The advantage of the first option is that the use cases may be more 
realistic, since the real world is untidy and more realistic scenarios 
and use cases will probably involve issues of interest to multiple WGs.  
On the other hand, this approach could become unwieldy and difficult to 
manage.  The advantage of the second approach is that the issues will be 
more focused on something specific to a given WG, making it perhaps 
easier for the WG to propose a solution.  On the other hand, there may 
be a danger of "stove-piping", in that the more narrow focus may mean 
that important issues at the boundaries of privacy and consumer identity 
(for example), or user managed access and volunteered personal 
information, may be missed.

Each of the WG Charters has a section on Related Work and Liaisons, 
where other WGs doing overlapping work are listed.   But how these 
liaisons would occur has not been addressed as yet.



Bob Pinheiro
Chair, Consumer Identity WG
consumerid at bobpinheiro.com

j stollman wrote:
> Bob,
> Regarding use cases, I think that we need a use case that is 
> sufficiently complex that it brings into play all of the issues of 
> privacy -- including correlation.  This means that it will require 
> multiple Identity Providers and multiple Relying Parties.  This is a 
> reason that I wanted to learn more about the "buying a car" scenario.  
> Another scenario could be going to an Emergency Room in a city far 
> from where you live (so that they don't have any current medical 
> history).  This would bring in issues like transferring of 
> HIPAA-protected information, insurance, credit card payment, and 
> retention of personal information by the ER that is treating you -- 
> recognizing that they will likely never see you again.
> On Fri, Aug 28, 2009 at 11:14 AM, Bob Pinheiro 
> <consumerid at bobpinheiro.com <mailto:consumerid at bobpinheiro.com>> wrote:
>     I'd like to address a few issues here regarding how the Consumer
>     Identity WG should operate, and solicit your input.*
>     First topic: _Review of the Proposed Use Cases_*
>     I previously put together a set of use cases
>     <http://kantarainitiative.org/confluence/pages/viewpage.action?pageId=23035928>
>     that I believe captures the kinds of problems that we are trying
>     to tackle.  These use cases are grouped into three different
>     scenarios in which high assurance consumer identity is
>     important.   These scenarios are:
>         * Service Provider needs to know the identity of a consumer
>           with high confidence prior to establishing a high-value,
>           identity-dependent service
>         * Consumer wants to prevent others from using his/her identity
>           to establish new accounts
>         * Ongoing access to high-value, online resources
>     However, these use cases are not the final word, and mainly
>     reflect my own thinking.  There are some other potential scenarios
>     and use cases that involve high assurance consumer identity that
>     are missing from this list, and that we may want to consider.  For
>     instance, there's the problem of stolen passwords and other
>     personal information that can enable fraud in online payments and
>     electronic funds transfer/electronic bill payments from personal
>     bank accounts.  Stolen passwords or guessable "challenge
>     questions" can enable a fraudster to impersonate the true owner of
>     a bank account or payment account, and allow the fraudster to
>     transfer money out of the victim's account or make other online
>     payments using the victim's account.  Then there's the whole
>     problem of stolen credit card numbers, resulting in fraudulent
>     credit card charges.  These problems reflect a failure to properly
>     authenticate the person making the payments, or initiating the
>     funds transfer, or using the credit card number.  Identity
>     paradigms such as Information Cards, coupled with stronger
>     authentication technologies, may be able to address these
>     problems.  Do we want to include these kinds of scenarios in our work?
>     The first Consumer Identity WG conference call was held on August
>     7 (call minutes available here
>     <http://kantarainitiative.org/confluence/display/WGCI/Consumer+Identity+WG+Teleconference+-+August+7%2C+2009>). 
>     During the call, it was suggested that some modifications to the
>     use cases might be in order, and a further call was scheduled for
>     August 14.  However, only one other person showed up, so we
>     canceled the call.  We still need to get input on these use cases,
>     or suggestions for additional ones, which leads to the next topic.
>     *Second topic: _Conference Calls and Other WG Interactions_*
>     We need to decide on the best way to interact as a group in order
>     to move forward.  One possibility is regular conference calls. 
>     Although it's often difficult to get a sizable number of people to
>     be available for such calls, until the WG is truly "up and
>     running" it's probably a good idea to have such calls where people
>     can raise issues and discuss other topics of interest.  Once
>     things are moving along, perhaps calls can be scheduled only "as
>     needed" as determined by the group.
>     I would like to get some feedback on the use cases, and discuss
>     any other topics concerning the direction of thisWG that people
>     want to raise.  So I'd like to schedule our next call sometime
>     within the next two weeks.  I'm putting up a new Doodle Poll
>     <http://www.doodle.com/qmyempw9kv3vqtt8> to determine the best day
>     and time.  Please respond asap to the poll if you're available and
>     want to meet, and I'll send out an agenda.  
>     *Third topic: _Kantara Meetings_*
>     As you probably know, Kantara Initiative will hold its first
>     meeting in conjunction with Digital Identity World in Las Vegas on
>     September 14-16.  A few of the other WGs are holding face-to-face
>     meetings during this time.  I'm probably not going to DIDW, and so
>     did not plan any F2F meeting for this WG.  If you're interested in
>     attending DIDW, please register using the priority code 
>     "kantara": https://orders.cxo.com/conferences/enroll.html?conferenceID=51. 
>     *Fourth topic:  _Your Input Is Needed!_*
>     Your comments, suggestions, and ideas are needed on several
>     fronts, including the consumer identity use cases and other WG
>     areas of interest, WG meeting format (calls versus mail lists),
>     etc.  So once again, if you're a participant in the Consumer
>     Identity WG, don't be shy about sharing your comments, thoughts,
>     feedback, ideas, etc. on this mailing list, and/or by
>     participating in the next (and subsequent) conference calls.
>     To stimulate the conversation, here are a few items you may want
>     to think about:
>         * Should we include payment-related use cases, as indicated
>           earlier?
>         * Much of what Kantara is about is the interoperability of
>           identity protocols.  I believe that one of the important
>           considerations needed to enable high assurance consumer
>           identity is the integration of stronger authentication
>           technologies (such as one-time passwords or personal X.509
>           certificates) with emerging identity paradigms such as
>           Information Cards and OpenID.  These are not necessarily
>           technical issues but may involve other things such as (a)
>           usability of various authentication form factors (hard
>           tokens, soft tokens, etc) by consumers; (b)  public policy
>           issues pertaining to levels of assurance necessary for
>           authentication to access consumer health records, financial
>           accounts, etc; (c) the economics of providing stronger
>           authentication technologies to consumers; (d) who will be
>           the identity providers that will verify a consumer's
>           identity and issue identity credentials for online use? 
>           Banks?  Motor Vehicle bureaus?  Other financial
>           institutions?  Entities involved with the development of
>           online consumer health records?; (e) How can Kantara, and
>           this WG in particular, make a difference in this space?
>         * Kantara is planning its budget for next year and the various
>           WGs have been asked to think about what kinds of resources
>           they will need to perform their work.  What kinds of
>           resources do you think may be needed for this WG to
>           accomplish its goals? 
>     Thanks for your participation.
>     Bob
>     ---------------------------
>     Bob Pinheiro
>     Chair, Consumer Identity WG
>     908-654-1939
>     consumerid at bobpinheiro.com <mailto:consumerid at bobpinheiro.com>
>     www.bobpinheiro.com <http://www.bobpinheiro.com>
>     _______________________________________________
>     Wg-consumer-identity mailing list
>     Wg-consumer-identity at kantarainitiative.org
>     <mailto:Wg-consumer-identity at kantarainitiative.org>
>     http://kantarainitiative.org/mailman/listinfo/wg-consumer-identity_kantarainitiative.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kantarainitiative.org/pipermail/lc_kantarainitiative.org/attachments/20090831/5ef041d1/attachment-0001.html>

More information about the LC mailing list