InCommon Technical Discussions

[Keith Hazelton <>]

On 2017-06-08, 12:51, 
"
 on behalf of Cantor, Scott" 
<
 on behalf of 
>
 wrote:

    > I do think it would
    > be helpful to figure out how we get from where we are, to having 
reliable,
    > long-term identifiers that can be used for SSO account linking from 
every IdP.
    > If there is interest in this topic, I would be really excited to 
participate in it.  (Laura Paglione, ORCID)
    
    +1
    
    I think the generally correct answer(s) are becoming fairly clear. All of 
them appear to be very hard to make happen due to lack of incentives, and so 
that is the next challenge. To date, nothing's really been proposed because 
of the seeming impossibility of making it happen.
    
    Getting to the point of proposing change is a good first step, but it's 
not the hard one.
    
    In the end, I believe federation in the sense we want it will work for a 
relatively small number of SPs and IdPs, and if that isn't sufficient scale, 
then success will elude us.
    
  -- Scott
  
  The R&E federations will have to be involved in promoting the adoption of 
any solution that emerges. I see that as the hard part. I can imagine 
bringing a proposal to REFEDS to enlist their member federations’ help in 
getting enough IdPs and SPs to adopt the solution to make a dent in the 
original problem.
  
  My point in the earlier email is that this is a problem that will just go 
away eventually as fewer and fewer science SPs attempt to federate with 
campus IdPs.  But that will leave those IdPs increasingly irrelevant to the 
research-mission-related SPs (I explicitly include ORCID).
  
  Are there people interested in helping put such a proposal together and 
pitching it to the inter-federation community?
  
      -- Keith H