Integration of Discovery Service

[Scott Koranda]

On Wed, May 6, 2015 at 9:09 AM, Rod Widdowson <rdw at steadingsoftware.com> wrote:
>> Rod might have a different view, but I tend to think that's more a benefit
>> in theory than practice now, though I suppose a redesigned CDS on top of
>> the new code base would inherit certain benefits in that respect.
>
> He doesn't.  Were we to do a CDS, that would definitely be part of the
> requirements.
>
> Having said that if there is function which the CDS could currently support
> via the plugins which cannot currently be achieved in the EDS I am very are
> amenable to RFIs on the EDS.  The only thing that the EDS will never support
> is (1) non javascript operation and (2) The legacy Shibboleth protocol for
> SAML1.
>

Thanks.

The plugin functionality looked attractive to address a use case like
this: if the IP address is from one of two subnets, and the target SP
is one from a configured list, then the user is sitting in a control
room at one of the detector sites and is an operator and an employee
of the laboratory so we know which IdP the user must use.

We have use cases with that kind of flavor.

I am happy to file RFIs on the EDS.

Thanks,

Scott K