Interfederation

[Scott Koranda <>]

Hi,

> On Wed, Sep 25, 2013 at 10:21 AM, Scott Koranda 
> <>
>  wrote:
> >
> > InCommon joining eduGAIN is only useful to LIGO at the point that
> > international IdPs release a persistent identifier to all of our SPs
> > published in InCommon without LIGO having to take any extra steps.
> 
> I believe you're conflating attribute release with technical trust. It
> doesn't really matter from where you obtain your partner metadata, you
> still have a separate attribute release issue.

I am confident I am not conflating attribute release with
technical trust. I know what I want my workflow to look like
as I join the trust fabric and I know that until attribute
release meets a threshold the trust fabric is of no use to me.

> 
> The best solution we have to that problem at the moment is R&S. Are
> you aware that REFEDs has proposed an R&S spec with international
> scope? 

Yes.

> As soon as that spec is finalized, we (InCommon) will support
> it, and then of course it's up to IdPs in the UKf to jump on the band
> wagon.

Understood, though I expect Ian can explain that there is more
to it then having IdPs "jump on the band wagon".

> 
> > I am not confident that happens immediately upon InCommon becoming an
> > eduGAIN participant.
> 
> Nor upon consuming UKf metadata either, as I believe you discovered
> during your interop with Cardiff.

Yes, but because of the more nimble and flexible working
relationship we were able to make progress at a faster rate then I expect
to be able to make with eduGAIN en masse.

> 
> > I see the InCommon and UK interfederation as another vehicle for
> > pursuing interfederation and exploring the space--a vehicle that will
> > be much more flexible and nimble.
> 

> Again, unless I'm missing something, I think eduGAIN is the
> way to go here.

I cannot say if you are missing something or not. 

I can say that in my experience and judgement the
collaboration between myself, Steven, and Ian was and will be
more nimble and flexible and make progress faster than with
eduGAIN en masse.

> 
> > If I am wrong and InCommon joining eduGAIN is quick and leads quickly
> > to what we need that is great.
> 
> It won't streamline attribute release (but then nothing---except
> perhaps R&S---will do that). eduGAIN is, however, the easiest way for
> us to get Cardiff metadata into your hands (and your metadata into
> theirs).

If it is easy then please do it. I am not objecting to having
InCommon leverage eduGAIN.

> 
> > I simply do not want to put all my eggs in that basket
> 
> May I ask why?

The work with Steven, Ian, and myself led directly to enhanced
functionality for users.

Thus far all talk from InCommon regarding eduGAIN has had no
impact for users.

My bias is to continue to support collaborative efforts that
have produced measurable impacts for users.

> 
> > and would like to continue to build on the work Steven,
> > Ian, and LIGO have done.
> 
> No offense, 

No offense taken.

> but that's not the totality of interfederation, not even
> close. The hard work of interfederation are the out-of-band agreements
> that assure relying parties that what they are consuming meets their
> requirements (whatever they happen to be).
> 

I do not argue that the work we did was the totality of
interfederation. 

I have expressed my desire for InCommon to continue to pursue
the bilateral interfederation agreement with the UK. I leave
it to you to have the last word if you like, I will not
comment further. 

I ask for the rest of the community to judge my input as it
shall.

Thanks,

Scott