Refeds


Subject Re: mari plan & next steps
From Leif Johansson <leifj@xxxxxxxx>
Date Wed, 29 Oct 2014 18:25:56 +0100

On 2014-10-29 17:39, Peter Schober wrote:
> * Leif Johansson <leifj@xxxxxxxx> [2014-10-29 17:34]:
>> If R&S specifies releasing eduPersonScopedAffiliation and FEIDE doesn't
>> have that attribute, how will including it in a bundle help FEIDE?
> 
> That's probably my point: It won't help FEIDE per se, it would force
> FEIDE to align its practices with the spec IFF they wanted to be able
> to apply to (SPs) or support (IDPs) REFEDS R&S.
> 
> I.e., I wouldn't want to weaken the harmonization brought about by
> having standardized attribute bundles by making them refer to
> meta-attributes instead.

These are two different problems:

1. get X to release data with the required semantics
2. get X to release data with the required syntax

to my bind 1 is tractable and R&S will help but 2 seems to be much
harder. semantics is much more important than syntax if the syntax-
diff is small enough (which I believe it is in this case).

yes, everything is a matter of degrees

	Cheers Leif