Stream: IG creation
Topic: Publishing an IG for ballot
Max Masnick (Aug 06 2021 at 17:52):
I want to submit an IG for publication for the ballot next week. I filled out the publication readiness checklist form from the bottom of this page and got the email "receipt" indicating this was received.
Is there anything else I need to do to get the IG content over to HL7 for publication? @Lloyd McKenzie
Lloyd McKenzie (Aug 06 2021 at 17:56):
That's an @Melva Peters question rather than a me question :)
Melva Peters (Aug 06 2021 at 17:58):
Lynn Laakso will be in touch if anything is needed. @Max Masnick which IG is it?
Max Masnick (Aug 06 2021 at 17:59):
Sorry Lloyd -- you're my go-to @ when I don't know who to @
Max Masnick (Aug 06 2021 at 17:59):
@Melva Peters http://build.fhir.org/ig/HL7/fhir-shc-vaccination-ig/branches/main/ -- you just got several emails from/about us :smile:
Melva Peters (Aug 06 2021 at 18:00):
I’ll check the qa.HTML later when I get home and let you know if there are issues remaining that need to be resolved
Lynn Laakso (Aug 06 2021 at 18:07):
package and version should be 0.1.0
Lynn Laakso (Aug 06 2021 at 18:13):
and your pull request for the change to the Jira-spec_artifacts repo should reflect the same, which will resolve your build warning
Paul Denning (Aug 06 2021 at 18:58):
In FHIR-shc-vaccination.xml do need to change
<version code="0.6.2" url="http://hl7.org/fhir/uv/fhir-shc-vaccination/2021Sep" />
to
<version code="0.1.0" url="http://hl7.org/fhir/uv/fhir-shc-vaccination/2021Sep" />
Can we remove
<version code="0.6.0" url="http://build.fhir.org/ig/HL7/fhir-shc-vaccination-ig/branches/0.6.0" /> ?
Max Masnick (Aug 06 2021 at 19:10):
For this IG the versioning made its way up above 0.1.0 outside the HL7 process. I’m concerned publishing 0.1.0 as the ballot version will be confusing to implementers who have been seeing 0.6.x for months
Lynn Laakso (Aug 06 2021 at 19:11):
I can see that it could be an issue; however that's the numbering scheme in HL7 IGs
Max Masnick (Aug 06 2021 at 19:31):
@Lynn Laakso I found this thread with Lloyd and Grahame -- this is what we were basing our version numbering off of as it sounded like 0.x.0 was ok for the ballot version (not limited to just 0.1.0)
Lynn Laakso (Aug 06 2021 at 19:38):
0.x.0 does not mean that the first ballot would be 0.6.0... If an IG goes to multiple STU ballots before publishing e.g. http://hl7.org/fhir/us/bser/history.html then the middle number would increment. However if the FMG approves that release number I can do that but FMG does not meet again until next Wednesday
Max Masnick (Aug 06 2021 at 19:46):
Ok, that makes sense. I'll update everything to use 0.1.0.
Max Masnick (Aug 06 2021 at 19:47):
Is this versioning convention documented anywhere? I tried looking for it back when I asked Grahame and Lloyd but couldn't find anything
Lynn Laakso (Aug 06 2021 at 19:48):
funny I just had that same conversation with @Melva Peters the other day and noted that we need to consolidate these numbering references and bring them forward into an easily-findable document
Max Masnick (Aug 06 2021 at 19:50):
@Lynn Laakso is this a new rule? It looks like there are other STU1 ballots that are not 0.1.0, like http://hl7.org/fhir/us/mcode/history.html
Max Masnick (Aug 06 2021 at 19:50):
Max Masnick (Aug 06 2021 at 19:50):
Lynn Laakso (Aug 06 2021 at 19:51):
This guidance dates back to the old HL7 Wiki, at https://wiki.hl7.org/index.php?title=FHIR_Package_IG_Release_Policy_Issue and approved by TSC on 2019-04-15. It needs updating from the perspective that it specifies a major version should conform to the STU publication release but not specifically that the ballot is an incrementation to the "minor" portion of the semantic versioning.
Lynn Laakso (Aug 06 2021 at 19:52):
again, it can be done but must be approved by FMG / FHIR Product Director
Max Masnick (Aug 06 2021 at 19:56):
Is there any way to get this approved without a full vote of the FMG? Since it does appear to be relatively common to not use 0.1.0, and going back to 0.1.0 is potentially confusing to our implementers, it seems worth fixing
Max Masnick (Aug 06 2021 at 19:58):
@Lloyd McKenzie or @Grahame Grieve - do you know if we can publish our STU1 ballot version as 0.6.2 rather than 0.1.0 without a full FMG vote? This is purely to avoid implementer confusion -- we've got lots of use of the 0.6.x version of this spec in the wild.
Lloyd McKenzie (Aug 07 2021 at 03:06):
It's fine to start a ballot at a higher version that 0.1.0. A work group might have had frozen versions they 'published' informally for connectathon, peer review, etc. We often do that with the core spec. No reason to prevent it for implementation guides.
Max Masnick (Aug 08 2021 at 11:21):
Ok thanks! @Lynn Laakso the version we submitted should be set for ballot then
Paul Denning (Aug 09 2021 at 02:30):
@Lloyd McKenzie do you need something more done to approve https://github.com/HL7/JIRA-Spec-Artifacts/pull/253 ?
Paul Denning (Aug 09 2021 at 02:57):
Thanks for approving the PR, but still have the warning in http://build.fhir.org/ig/HL7/fhir-shc-vaccination-ig/qa.html
Do we need to remove the line for 0.6.0? https://github.com/HL7/JIRA-Spec-Artifacts/blob/master/xml/FHIR-shc-vaccination.xml#L5
Lloyd McKenzie (Aug 09 2021 at 02:58):
It's there now, can't be removed. If it's not in your package-list.json file, you should flag it as deprecated
Last updated: Apr 12 2022 at 19:14 UTC
