Metadata Distribution Subcommittee of TAC

["Cantor, Scott" <>]

On 10/3/13 3:32 PM, "Tom Scavo" 
<>
 wrote:
>>
>> I also think we'll greatly limit interfederation by making it opt-in.
>>So I
>> see this as at most an interim step.
>
>Yes, I'm beginning to understand the significance of this decision.
>The moment we expose two metadata distribution endpoints, we introduce
>the risk of fragmenting the InCommon participant community.

I feel that very strongly. I think multiple aggregates are a great way to
do:

- rollout of new features
- early adopter vs. mature production splits
- testing of impact of things like SHA-2

But I don't think they serve the general community well as a way to expand
the scope of the federation.

So in essence, interfed could be rolled out to a second aggregate to test
it out and identify problems (the first two bullets) but could then be
rolled into the production aggregate.

-- Scott