DCAT Question: SSA inSeries

[SueValente]

• Answer question in comments for this ticket
• Add Q & A to FAQ list with your initials and last edited date: https://github.com/GSA/dcat-us/wiki

On the DCAT-US Schema v3.0 | resources.data.gov main page, there are 3 references to converting the old DCAT 1.1 “isPartOf” to “inSeries” – but I don’t see guidance about “inSeries” in either the Implementation Guide or github.

It appears as though the true new 3.0 implementation is via the new DatasetSeries class.

Can you confirm that for me?

Also – Rachel Yuan is leaving SSA effective 9/18 – so you can remove her as a contact, if you keep a list of contacts.

Thanks!

-Jessie

---

Jessie Dalrymple
Visual Information Specialist
CIO > TCDAI > Reporting Branch
Social Security Administration

[SueValente]

@jbrown-xentity

[jbrown-xentity]

I think this is a somewhat legitimate criticism that we could explain further.

In order to get them something that they can run with, here is my short response.

Our guidance could use some more detail in this area, thank you for bringing it to our attention! We will attempt to address this soon.

The general overview is that if you had a "parent" dataset that other datasets referenced with the `isPartOf` link, then that parent should either turn into a [dataService](https://resources.data.gov/standards/catalog/dcat-us-3/catalog/#service) or a [datasetSeries](https://resources.data.gov/standards/catalog/dcat-us-3/catalog/#datasetSeries). Each of these objects has the ability to list sub-datasets according to the DCAT-US3 schema. That list would be where the datasets that come from or belong in the dataService or datasetSeries should be listed.

If this is accomplished, the end result will be the same in catalog. These references will be displayed both upward (dataset page to dataService or datasetSeries) and downward (dataService or datasetSeries to all sub datasets).

[JessieD-SSA]

Fabulous. Thanks, @jbrown-xentity - this was my impression. I'll try a test case in our next harvest to confirm the results.

While you're working on updating this documentation - I noticed that in the DatasetSeries schema samples, the Datasets samples that are under seriesMember do not have their own Distribution classes. But the Dataset classes inside DatasetSeries should be able to have their own Distributions, right?

The DCAT 3.0 Implementation Guide from GSA says "For datasets that are part of a dataset series, agencies only need to define the datasets within the seriesMember property (see below); they do not need to be defined within the dataset property for the catalog."

So I'd think we could include Distributions inside those nested Datasets. Actually, that we should be able to include any "regular" Dataset properties/classes, right?

[saitanallensantiago26-beep]

https://www.useparagon.com/resources/build-vs-buy-content?submissionGuid=f9981b94-42b8-4fcb-bb05-ffb64805ec04

[saitanallensantiago26-beep]

• Answer question in comments for this ticket
• Add Q & A to FAQ list with your initials and last edited date: https://github.com/GSA/dcat-us/wiki

On the DCAT-US Schema v3.0 | resources.data.gov main page, there are 3 references to converting the old DCAT 1.1 “isPartOf” to “inSeries” – but I don’t see guidance about “inSeries” in either the Implementation Guide or github.

It appears as though the true new 3.0 implementation is via the new DatasetSeries class.

Can you confirm that for me?

Also – Rachel Yuan is leaving SSA effective 9/18 – so you can remove her as a contact, if you keep a list of contacts.

Thanks!

-Jessie

---

Jessie Dalrymple
Visual Information Specialist
CIO > TCDAI > Reporting Branch
Social Security Administration