Stream: shorthand
Topic: shorthand roadmap (not IG-related)
Jose Costa Teixeira (May 26 2020 at 20:15):
I assume the roadmap is separated between the IG stuff and the shorthand stuff. For the shorthand stuff, are these is in the plan?
- Logical models
- convery structuredefs back to shorthand
- a way to do "must support=false"
Jose Costa Teixeira (May 26 2020 at 20:18):
These are the things that I am personally missing most. The IG stuff is nice but it's existing functionality already
Chris Moesel (May 26 2020 at 21:33):
Hi @Jose Costa Teixeira -- we're first determining how to address the comments from the ballot before getting too deep into the next set of features.
We agree that logical models are important and we do think FSH needs to support them.
We also agree that round-tripping between SD and FSH is important. @Rick Geimer did a prototype of this with FSH Food; we need to determine if that work will continue under Lantana or if someone else needs to pick it up.
As for must support = false, we've inquired in the past (link) and learned that you cannot set MS from true to false in a profile. Since default MS is false, we don't see use cases where setting MS to false is needed (until and unless FHIR makes it valid). The only thing I can think of (which I just thought of now) is if you want to set it to false to prevent any sub-profile from ever setting it to true. That seems very off the beaten path, so in that case, I think it's fair to require authors to use the escape syntax for that (e.g., ^mustSupport = false). Do you have another use case in mind (that is valid in FHIR)?
Jose Costa Teixeira (May 26 2020 at 21:36):
just for me to tone my expectations: how much of the ballot commenting was about shorthand vs the ig stuff?
Chris Moesel (May 26 2020 at 21:36):
Edit on that last one -- I had mustSupport in my head, but obviously that thread was about isSummary. We'll have to check in on mustSupport then. For now, however, assuming it is valid in FHIR, you should be able to use the escape syntax: ^mustSupport = false.
Chris Moesel (May 26 2020 at 21:40):
how much of the ballot commenting was about shorthand vs the ig stuff?
The IG stuff is actually out of scope for the ballot -- as it is not part of the spec. So, I don't think you need to worry about IG stuff taking too much of our time in that regard. In fact, aside from figuring out what to do with the IG stuff we already worked on (like the new config beta), I think we're going to primarily focus on FSH-proper stuff from here on out -- and try to encourage the IG Publisher to take on some of the more user-friendly IG features SUSHI currently provides.
Jose Costa Teixeira (May 26 2020 at 21:40):
Thank you
Jose Costa Teixeira (May 26 2020 at 21:42):
user-friendly features = the config file?
Jose Costa Teixeira (May 26 2020 at 21:46):
for the ^mustSupport = false, it doesn't work in my case (just tried it) possibly because I have in the same file "identifier MS" and then the escape syntax; the upper line prevails. I will describe my needs tomorrow but it was a non-standard use of shorthand that I was testing to have more control of changes
Chris Moesel (May 26 2020 at 21:53):
The new config file currently serves three purposes: 1. it provides the minimum data necessary to create FHIR definitions (e.g., the canonical URL); 2. it provides a way for users to have fine-grained control over the ImplementationGuide resource; and 3. it provides a single-file to manage what is typically represented in multiple files (including ig.ini, package-list.json, menu.xml, and the ImplementationGuide resource).
Chris Moesel (May 26 2020 at 21:54):
The user-friendly IG features I was referring to are more along the line of #3, such as a friendlier way to create menus and package history -- as well as a friendlier way for managing pages (by auto-generating the pages parameter based on conventions).
Chris Moesel (May 26 2020 at 21:56):
I'd say whether or not SUSHI should provide a friendlier way to author the ImplementationGuide resource itself (#2 above) is up in the air. But anything that we do in #2 above that is "magic" should be considered a candidate for the IG Publisher, I think.
Last updated: Apr 12 2022 at 19:14 UTC
