Metadata Distribution Subcommittee of TAC

[Ian Young <>]

On 18 Dec 2013, at 14:09, Tom Scavo 
<>
 wrote:

> 1) The plan calls for the preview aggregate to be *identical* to the
> production aggregate but the technique we're using causes the ID
> attributes to be different. We don't think that's a big deal but I
> wanted to ask and make sure.

I don't see it as a problem, I can't see anyone actually using that other 
than as part of the signature verification.

> 2) What should be the content type of the metadata response?
> Historically, we've responded with text/xml but we're guessing
> application/samlmetadata+xml is preferred. The two content types have
> different user experiences so we want to get this right out of the
> box. What content type should we use?

The spec is clear that the type should be application/samlmetadata+xml:

> If the publishing protocol permits MIME-based identification of content 
> types, then
> use of the application/samlmetadata+xml MIME media type is required.


I don't know why that isn't written as REQUIRED, but it's probably just a 
typo.

In the UKf, we moved from the former to the latter a fair while back and I 
don't recall any problems resulting from that.

        -- Ian

Attachment: smime.p7s
Description: S/MIME cryptographic signature