FHIR Chat · Logical Model page layout · IG creation

Stream: IG creation

Topic: Logical Model page layout


view this post on Zulip Jose Costa Teixeira (Apr 17 2020 at 16:22):

How would a Logical Model page look like in an IG?

view this post on Zulip Jose Costa Teixeira (Apr 17 2020 at 16:23):

  1. should it show directly the model (differential) instead of the Text summary ?

view this post on Zulip Jose Costa Teixeira (Apr 17 2020 at 16:24):

  1. I supposed Detailed Descriptions and Mappings would remain..?

view this post on Zulip Jose Costa Teixeira (Apr 17 2020 at 16:25):

  1. I guess the header instead of being "StructureDefinition: ..." could be "Logical Model: " (or "Logical Data Model:") ?

view this post on Zulip Lloyd McKenzie (Apr 17 2020 at 17:05):

  1. Text summary could still be relevant. We're probably going to change where it appears in the tab order for everything. (And make it more useful)
  2. y
  3. Logical Model is the term we use in the spec, so keeping consistent would be good

view this post on Zulip Giorgio Cangioli (Apr 18 2020 at 16:13):

Jose Costa Teixeira said:

  1. I supposed Detailed Descriptions and Mappings would remain..?

I hope so .. :-)

view this post on Zulip Jose Costa Teixeira (Apr 18 2020 at 17:48):

Any other ideas for things that should be different?
(Besides the "examples" section for which there is a separate discussion)

view this post on Zulip Frank Oemig (Apr 19 2020 at 10:55):

@Jose Costa Teixeira do you have a sample IG? This is something we gave to work on for v2+ but I would like to benefit from existing work and thoughts...

view this post on Zulip Jose Costa Teixeira (Apr 19 2020 at 11:48):

@Frank Oemig you mean sample logical models? There are a few.
If you mean "sample new rendering of Logical Model" then no, but that may be a good idea to put in the guidance IG.

view this post on Zulip Jose Costa Teixeira (Apr 19 2020 at 11:50):

We don't have the new rendering yet because work is just starting (hence collecting ideas).

view this post on Zulip Jose Costa Teixeira (Apr 19 2020 at 11:50):

I can try to make a PR soon to the base template so that Lloyd can merge it.

view this post on Zulip Frank Oemig (Apr 19 2020 at 16:55):

IGpublisher can do a lot. So it would be good to have a max sample IG that demonstrates the different features.

view this post on Zulip David Hay (Apr 19 2020 at 20:49):

Not sure if it's useful - but here's one I'm working on - http://igs.clinfhir.com/healthalliance/StructureDefinition-HaPatient.html

Would like to lose from the snapshot:

  • meta info (id, meta, implicit rules, language)
  • extensions, contained
  • slice definition

Not sure if the differential tab adds value

(I am using clinFHIR to create the models as a precursor too creating the profiles - so may well be introducing oddities from there.)

Not sure what I'd add - I've just started using LMs in the IG...

view this post on Zulip Jose Costa Teixeira (Apr 19 2020 at 20:58):

@David Hay do you mean this
http://igs.clinfhir.com/healthalliance/StructureDefinition-HaPatientLM.html
instead?

view this post on Zulip David Hay (Apr 19 2020 at 21:50):

bother! yes...

view this post on Zulip David Hay (Apr 19 2020 at 21:51):

I should check what I'm doing before I click send! My comments above can be ignored...

view this post on Zulip Jose Costa Teixeira (Apr 19 2020 at 22:28):

Well, if we make the LM at least say "Logical model" instead of StructureDefinition you will spot it more easily ;) I struggle in my profiles as well

view this post on Zulip Jose Costa Teixeira (Apr 21 2020 at 13:50):

  1. Do we want Terminology bindings ?
  2. and Constraints?
  3. The Schematron link would go away, right?

view this post on Zulip Jose Costa Teixeira (Apr 21 2020 at 13:51):

image.png

view this post on Zulip Jose Costa Teixeira (Apr 21 2020 at 13:52):

on 4. I think yes, we must preserve terminology bindings. - So far our practice is : the logical model may contain bindings that are defined at the functional / legal level - for example civil states.

view this post on Zulip Lloyd McKenzie (Apr 21 2020 at 14:02):

Terminology bindings and constraints are both appropriate in logical models

view this post on Zulip Lloyd McKenzie (Apr 21 2020 at 14:03):

Schematrons aren't so relevant because you're not supposed to create and share instances. However, that's not 100% true. @Grahame Grieve thoughts?

view this post on Zulip Grahame Grieve (Apr 21 2020 at 19:50):

that's not true at all. They are not resources but that doesn't mean that they aren't meaningful models

view this post on Zulip Lloyd McKenzie (Apr 21 2020 at 19:59):

Schematron has nothing to do with whether they're meaningful, it has to do with whether you're producing instances for them.

view this post on Zulip Grahame Grieve (Apr 21 2020 at 20:02):

well, any of them may have instances

view this post on Zulip Lloyd McKenzie (Apr 21 2020 at 20:49):

So even though those models are not supposed to be used for exchange purposes, supporting schematron would still be appropriate?

view this post on Zulip Grahame Grieve (Apr 21 2020 at 20:50):

could be. Depends on whether XML is in scope e.g. CDA

view this post on Zulip Jose Costa Teixeira (Apr 22 2020 at 08:01):

So, about the layout - if the Logical Model layout header becomes "Logical Model" image.png

view this post on Zulip Jose Costa Teixeira (Apr 22 2020 at 08:04):

would we also change the Profile layout to be "Profile"?

view this post on Zulip Jose Costa Teixeira (Apr 22 2020 at 08:04):

image.png

view this post on Zulip Lloyd McKenzie (Apr 22 2020 at 14:00):

Sure


Last updated: Apr 12 2022 at 19:14 UTC