The meta-data for the quickstarts can be obtained from here: https://hub.openshift.com/api/v1/quickstarts.json

The OpenShift team don't want us to duplicate the Quickstart detail page (E.g.https://hub.openshift.com/quickstarts/1-wordpress-4) as it will affect SEO having duplicate pages. Instead we will display the meta-data on on the learn page, and link to the detail page on hub.openshift.com. This is the same pattern as followed by the customer portal articles and solutions.

I suggest a DCP river is used to index this content in the same way as the Customer portal documents are indexed. This needs to be indexed in both DCP v1 and DCP v2 instances. This removes the risk of requiring migration to DCP v2 for the Q1FY16 release.

The meta-data looks pretty complete, apart from author details. I don't think this should be a problem.

Content type

An outstanding question is, which content type do we use?

jbossdeveloper_quickstart

Add OpenShift quickstarts to the current quickstart bucket, with a 'target_product' of 'openshift'.

• This makes them easier to search over along with all the other quickstarts we have
• The OpenShift quickstarts are quite different to the JBoss ones, so maybe it doesn't make sense to group them together
• We'd need to filter these out on JBoss Developer somehow.
• There might be a problem with the meta-data being slightly different. For example, some required fields might be missing.

openshift_quickstart

Make a new bucket for just the OpenShift quickstarts.

• It might be harder to handle this on the 'learn' page. Do we have a separate format, or include them in the results for the current "Quickstarts" format? I suggest the later, in which case the query would need to be updated to support this.

'Early Access' vs 'Supported'

Do we need to differentiate the quickstarts ('Early Access' vs 'Supported') as we do the middleware ones? The OpenShift quickstarts do have a 'provider' field that has values [partner, openshift, community, trusted]. Maybe we translate these values.