Stream: students
Topic: FHIR Profiles - Student Track DevDays
Patrick Werner (Jul 21 2016 at 08:53):
I just had a closer look at the Norwegian profiles for the Student Track and couldn't find the profiles in XML representation, e.g. for the Obeservation profile: http://devdays-studenttrack.furore.com/Chapter.3.Profiles/structure.LabObservationNorway.html . I already searched simplifier, but the Norwegian profiles which i found there aren't matching the student track profiles (https://simplifier.net/ui/Search/Index?text=labobservationnorway).
Has anyone a clue where i can find the correct StrucureDefinition files? Maybe @Marten Smits could help?
Marten Smits (Jul 21 2016 at 08:54):
Hi Patrick, I can create a project on Simplifier with the XML representations and create a link to those in the IG. Would that be sufficient?
Patrick Werner (Jul 21 2016 at 08:55):
would be perfect! Thank you.
Marten Smits (Jul 21 2016 at 09:10):
https://simplifier.net/DevDaysStudents2016 Here is the project. I will edit the IG later
Marten Smits (Jul 21 2016 at 10:03):
You can download the XML (or Json) in the options menu
Marten Smits (Jul 26 2016 at 13:33):
@Patrick Werner 'I added a link in the IG now in the Profiles section and in the FAQ section
Patrick Werner (Jul 26 2016 at 13:41):
Thank you Marten
Patrick Werner (Aug 11 2016 at 14:38):
Hi @Marten Smits is just checked the LabObservationNorway. It seems that the discriminator for the slicing of the indentifier is missing.
Simone Heckmann (Aug 11 2016 at 14:51):
huh? the cardinality for Observation.identifier is 0..0 !
...or am I missing something here?
Simone Heckmann (Aug 11 2016 at 17:58):
oh. You mean the LabPatientNorway. Yes that's a surprising slicing with both slices fixed to the same identifier.system! I wonder what the use case is?
Patrick Werner (Aug 11 2016 at 20:24):
yes i meant the Patient. I just see a Slicing definition and one slice.
But if i load the profile into Forge i get the Error: "Invalid slicing definition. If there are no discriminators , there must be a definition.
Marten Smits (Aug 15 2016 at 11:46):
Hi Simone, Patrick, sorry for the late reply. I'll look into it later this week. Or are you in a hurry?
Patrick Werner (Aug 15 2016 at 11:54):
Hi Marten, no hurry needed from my side.
Simone Heckmann (Aug 17 2016 at 21:06):
Nope. All cool 
Marten Smits (Aug 17 2016 at 21:06):
@Patrick Werner Fixed it, should be visible in a couple of minutes!
Simone Heckmann (Aug 17 2016 at 21:06):
(deleted)
Simone Heckmann (Aug 18 2016 at 07:11):
(deleted)
Mahmoud Alakraa (Sep 28 2016 at 21:31):
Hi
in students track there is 5 profiles , should we use the structure that mention in student track or the general one because there is some variation between for example the general patient resource structure and patient resource structre for student track
Marten Smits (Sep 29 2016 at 07:47):
You should use the structures that are in the Student Track implementation guide.
Mahmoud Alakraa (Sep 29 2016 at 10:33):
thanks
Patrick Werner (Oct 11 2016 at 19:55):
Hi @ all Student Track Attendees and Mentors. Last year we had a nice meeting at the AMC before the Student Dev Days. Is something similar planned this year? It would be nice to keep this "tradition" going.
Mahmoud Alakraa (Oct 31 2016 at 00:10):
Hi
there is some elements in the profiles structure for student track are canceled with line , so should we ignore this or what ??
Patrick Werner (Oct 31 2016 at 10:58):
what do you mean with "canceled with line"
Monika Moser (Nov 01 2016 at 16:03):
Hi, again. In Simplifier there are some examples, but we are not able to open them. We generated some example data on our own. But if possible we'd like to try also the examples from simplifier, to check if we are on the right track.
Especially for validation it would be nice if we have some more example data to try.
Speaking of validation. Is it suitable to validate against a generated schema? Because validation wasn't working with the norway profiles. In reference to this post (https://chat.fhir.org/#narrow/stream/implementers/topic/.24validate.20with.20sliced.20profiles ).
Marten Smits (Nov 02 2016 at 10:44):
Hi Monika, in current DSTU2 validators, slicing is not yet supported. We are working on that as we speak, but that will not be finished before the DevDays.
Marten Smits (Nov 02 2016 at 10:45):
Also, I'm not sure why you can't see the examples. The XML and JSON views don't work? https://simplifier.net/DevDaysStudents2016/DiagnosticReport-example4/jsonview
Patrick Werner (Nov 02 2016 at 10:55):
Hi Monika, we are currently testing a minimal hapi-based validator for the Studenttrack implemented as a webservice. This validator has some basic functionality in regard of validating sliced resources against the Norway Student Track profiles. This service should be ready in time for the student track.
Monika Moser (Nov 02 2016 at 16:46):
@Marten Smits I see, yes absolutely my fault. I tried to stay in the example tab opening a example, but don't got it that it is already open in another tab. Sorry about that.
Marten Smits (Nov 02 2016 at 16:59):
@Monika Moser No problem, great that you use the stream to ask questions
Mahmoud Alakraa (Nov 03 2016 at 18:16):
Mahmoud Alakraa (Nov 03 2016 at 18:17):
for example :paca.JPG
Simone Heckmann (Nov 03 2016 at 18:24):
Wow. That's some strict constraining! :)
Marten Smits (Nov 03 2016 at 20:46):
@Mahmoud Alakraa The "stripe through" means the cardinality is 0..0, which means you are not allowed to use these elements. We "canceled" this many elements to make it easier for servers and clients to know what you can expect
Lloyd McKenzie (Nov 04 2016 at 03:05):
Ok, but it may suggest to the students that this is good practice. In general, profiles should avoid limiting the upper cardinality for elements as it essentially says "if you have data that I don't need, you have to spend money to remove data that I could ignore for free" - which isn't the ideal statement to make if you are genuinely interested in interoperability. (If you only want one element and an element could potentially repeat, best practice is to instead enforce a limitation such as "one instance must have this special use code or extension" rather than saying "there can only be one instance" - that way a system can emit a single instance that can consume both your "single element" requirement as well as the needs of other consumers who may need multiple or different occurrences of the same element.
Last updated: Apr 12 2022 at 19:14 UTC
