Stream: committers
Topic: Profile Names
Grahame Grieve (Mar 20 2019 at 21:56):
Many IGs are producing this warning: "Name should be usable as an identifier for the module by machine processing applications such as code generation" for lots of resources
Grahame Grieve (Mar 20 2019 at 21:57):
editors please review these - implementers to care about this warning
Grahame Grieve (Mar 20 2019 at 21:57):
the IG publisher will consider this an error, not a warning, in the future
Lloyd McKenzie (Mar 21 2019 at 00:22):
An error for HL7 or an error for everyone?
Grahame Grieve (Mar 21 2019 at 00:23):
HL7 at least
Lloyd McKenzie (Mar 21 2019 at 00:23):
And presumably this will take into account the tracker item where it yells about it even where there isn't a name (and there's no reason for there to be a name)?
Grahame Grieve (Mar 21 2019 at 00:25):
Why would there be a reason not to have a name?
Lloyd McKenzie (Mar 21 2019 at 00:28):
Sticking names on Questionnaires or questionnaire-contained value sets doesn't make a whole lot of sense most of the time.
Lloyd McKenzie (Mar 21 2019 at 00:28):
The chances that anyone's going to generate a class model for a particular questionnaire is pretty low.
Grahame Grieve (Mar 21 2019 at 00:29):
really? what on earth makes you think that?
Lloyd McKenzie (Mar 21 2019 at 00:29):
And in instance examples we typically only include the elements we identify as 'mustSupport' - and name isn't one of those.
Lloyd McKenzie (Mar 21 2019 at 00:29):
I didn't say systems can't do that, but just that most don't.
Lloyd McKenzie (Mar 21 2019 at 00:29):
And I don't want to stick stuff in instance examples that systems don't have to do.
Grahame Grieve (Mar 21 2019 at 01:34):
well maybe we make Name mandatory to get right for non-examples in Hl7 guides
Lloyd McKenzie (Mar 21 2019 at 02:28):
I'm fine for that in principle, though we have a bit of a challenge distinguishing examples from non right now. For example, we often mark CapabilityStatements and similar resources as 'examples' because that's the rendering template we use for them. We'll need to change that if we want to distinguish 'true' examples and enforce proper constraints on resources that are part of the definitional content of the resource.
Bryn Rhodes (Mar 21 2019 at 02:32):
+1 for being able to distinguish between examples and definitional content, I run in to that a lot.
Grahame Grieve (Mar 21 2019 at 02:50):
well, you can distinguish. Can't do that much if people are not
Lloyd McKenzie (Mar 21 2019 at 02:51):
Distinguishing is currently too tied into publishing format. We need to split those things.
Grahame Grieve (Mar 21 2019 at 02:51):
how so? the publisher doesn't care
Lloyd McKenzie (Mar 21 2019 at 02:52):
It certainly used to...
Eric Haas (Mar 21 2019 at 05:47):
I’m think that cap statement is the most common place re run into this.
Lloyd McKenzie (Mar 21 2019 at 14:04):
y
Last updated: Apr 12 2022 at 19:14 UTC
