FHIR Chat · Revamping resource summary · ig publishing requirements

Stream: ig publishing requirements

Topic: Revamping resource summary

Lloyd McKenzie (Aug 03 2021 at 23:06):

Today, we reviewed a proposed enhancement to the IG publisher templates around the 'metadata' table that currently shows up at the beginning of all resources. To refresh your memory, the current table looks like this:
image.png

The proposed table looks like this:
image.png

or, if all possible metadata on a canonical resource is filled in, like this:
image.png

Lloyd McKenzie (Aug 03 2021 at 23:13):

A whole bunch of things have happened:

Whether a profile is abstract or an artifact is experimental is now made immediately obvious on the title. (Abstract profiles and logical model titles are also italicized)
If you hover over canonical URL or the version, you'll now get a little icon that lets you copy either the canonical URL or a versioned canonical reference to the resource with one click. Looks like this:
image.png
If FMM and/or 'standards status' (i.e. informative, trial use, normative, deprecated, etc.) extensions are present, they'll show up
If FMM or standards status are declared on a root artifact, they'll automatically cascade to related artifacts. How they've cascaded is shown in a flyover
All of the metadata from canonical resources is now exposed, including publisher, jurisdiction, computable name, identifiers, keywords, usage, and purpose so these elements can now be reviewed if present - previously they were only seen if you looked at the XML, JSON or TTL
the information is presented in a more compact form and only elements that are actually present show up.

Lloyd McKenzie (Aug 03 2021 at 23:15):

In addition, regular pages now also have a header (if extensions are present in the IG with relevant information to make that happen):
image.png

Lloyd McKenzie (Aug 03 2021 at 23:16):

The color of the heading is consistent with the core spec - yellow for STU, green for normative, grey for informative, etc.

Lloyd McKenzie (Aug 03 2021 at 23:18):

The only action required by authors is to declare the extensions to assert the FMM and/or standards status on the artifacts or pages they wish, overriding where propagation would be incorrect. E.g. If an overall IG is FMM 2/STU, but a couple of artifacts are FMM 0/draft because they were added post-ballot, you'd put FMM 2/STU on the IG and FMM 0/draft on the new resources and everything else would update. Similarly, you'd assert FMM/STU on the pages that actually contain STU content and everything else would automatically be marked as "Informative".

Lloyd McKenzie (Aug 03 2021 at 23:21):

This change involves both updates to the publisher (to allow propagating status/FMM across resources as well as exposing metadata in a way the templates can access it. It also involves changes to the templates. Because few people can run publisher changes in Eclipse, I can't practically let you run your own IGs through it yet. I don't want to move the publisher changes up until post-ballot opening, just in case they have a negative impact. (Very unlikely, but better safe than sorry.)
If feedback about the change is positive, I'll move up the publisher changes immediately post-ballot, give people a few days to play with a local branch copy of the template and make sure it seems happy on a variety of IGs, then move it to the 'current' version of the template.

Lloyd McKenzie (Aug 03 2021 at 23:21):

Please provide your thoughts/feedback.

Richard Townley-O'Neill (Aug 04 2021 at 06:33):

Much nicer.

Where is the relationship between StructureDefinition.status and standards-status documented?

Craig Newman (Aug 04 2021 at 13:30):

Is there a reason that the Publisher (some WG) isn't labeled the same way other cells are? As well, what I'm assuming is the realm (where the flags are isn't labeled either. I know this is just an example, but have flags that don't seem to match the text is confusing.

Craig Newman (Aug 04 2021 at 13:31):

Is "Usage" new? From the example (and I know it's just an example), I can't tell what this is supposed to be.

Craig Newman (Aug 04 2021 at 13:31):

Otherwise it looks much nice!

Lloyd McKenzie (Aug 04 2021 at 14:00):

I don't know that we've documented a formal relationship between the 'status' values (draft/active/retired/unknown) and the 'standards status' values (Draft/Trial Use/Informative/Normative/Deprecated), though I'd say that Trial Use, Informative and Normative would all be subtypes of active, draft = Draft and retired=Deprecated.

Lloyd McKenzie (Aug 04 2021 at 14:03):

Could certainly add a label for publisher, just didn't want to risk it wrapping. There are actually 5 jurisdictions listed for this sampleIg (to test out the behavior). The two with text didn't have flags I could look up. The flags themselves have flyovers. Usage corresponds to the useContext element. It's primarily useful for localized artifacts in registries. E.g. forms specific to a particular age range, gender, location, etc. I doubt it'll be used much at the international level

Lloyd McKenzie (Aug 05 2021 at 17:24):

Based on feedback received, I've made a few changes:

put the flags in the header for the IG overall to represent the jurisdiction(s) of the overall IG
suppress the flags on individual resources unless they differ from what's on the overall IG (e.g. some artifacts are specific to some jurisdictions)
suppress the publisher on individual resources unless it differs from what's on the overall IG
So it now looks like this:
image.png
image.png
image.png