FHIR Chat · Freeze · committers

Stream: committers

Topic: Freeze

Lloyd McKenzie (Mar 24 2016 at 04:14):

As a reminder, the freeze date for Connectathon and for IG ballots is end of day Monday Mar. 28 at midnight Eastern - so get in whatever you need by then

Grahame Grieve (Mar 29 2016 at 03:04):

1 hour to go - is anyone still planning to commit something prioer to the freeze?

Grahame Grieve (Mar 29 2016 at 03:04):

in particular, structural changes....

Lloyd McKenzie (Mar 29 2016 at 03:42):

I'm hoping to get in changes to ReferralRequest and Task to get them to align with most recent Workflow decisions.

Grahame Grieve (Mar 29 2016 at 03:42):

how long are you likely to be?

Rick Geimer (Mar 29 2016 at 03:43):

I have some minor C-CDA on FHIR tweaks to get in (HTML page links, etc.).

Grahame Grieve (Mar 29 2016 at 03:43):

my focus is structure right now. In order to QA the build, I have to get my server up to date on the latest code. It's not affected by non-structural changes

Lloyd McKenzie (Mar 29 2016 at 03:44):

Depends whether the first commit works. Working on examples now.

Lloyd McKenzie (Mar 29 2016 at 03:45):

Do you want me to commit without fixing the examples?

Grahame Grieve (Mar 29 2016 at 03:45):

no I'm no in that much of a hurry

Rick Geimer (Mar 29 2016 at 03:57):

Is there time for me to finish a local build before committing my changes to guides/ccda? Would need about 35 minutes.

Rick Geimer (Mar 29 2016 at 03:57):

Partial build worked, but feel better running a full.

Rick Geimer (Mar 29 2016 at 04:01):

Went ahead and committed anyway to make sure I got in under the wire.

Lloyd McKenzie (Mar 29 2016 at 04:03):

I didn't even kick off a build. Fingers crossed . . .

Lloyd McKenzie (Mar 29 2016 at 05:03):

I'm done

Grahame Grieve (Mar 29 2016 at 05:12):

great. Freeze is not quite on yet - I'm still trying to resolve validation errors from my last change

Brian Postlethwaite (Mar 29 2016 at 05:14):

Thought it was just my changes...

Grahame Grieve (Mar 29 2016 at 05:14):

nah. changing structure definition - would have been fine except I lost time earlier

Brian Postlethwaite (Mar 29 2016 at 05:19):

Since I cant build, I guess cant commit that change I had to roll back.

Brian Postlethwaite (Mar 29 2016 at 05:20):

If things get working, let me know and I'll grab the code and try again. (Keen to have it in, but not critical - only very draft content)

Lloyd McKenzie (Mar 29 2016 at 06:13):

I think the current build is now working

Brian Postlethwaite (Mar 29 2016 at 06:27):

Strange, I did a get and nothing came down. Will try build again anyway.

Brian Postlethwaite (Mar 29 2016 at 06:34):

Still getting this one:
[java] Exception in thread "main" java.lang.Error: Case mismatch of file C:\Work\org.hl7.fhir\Clean Build\publish\valueset-observation-vitalsignresult.html: found valueset-observation-VitalSignResult.html

Brian Postlethwaite (Mar 29 2016 at 06:34):

a cached valueset perhaps?

Grahame Grieve (Mar 29 2016 at 06:35):

don't know. maybe.

Grahame Grieve (Mar 29 2016 at 06:35):

delete everything in publish and try again

Grahame Grieve (Mar 29 2016 at 06:59):

ok. freeze is now on; any further changes to structures have to be cleared with me.

Grahame Grieve (Mar 29 2016 at 06:59):

changes to html are allowed for a little while yet

Richard Ettema (Mar 29 2016 at 12:45):

@Grahame Grieve, just wanted to confirm - is svn revision 8107 the "official" snapshot, or should we wait based on your recent commits?

Grahame Grieve (Mar 29 2016 at 12:46):

I'm still doing error correction and publication fixes

Grahame Grieve (Mar 29 2016 at 12:46):

there'll certainly be more yet

Grahame Grieve (Mar 29 2016 at 12:46):

I haven't yet got my server to run :-(

Richard Ettema (Mar 29 2016 at 12:48):

Ok. Please let me know if there's anything I can do to help. Thanks.

Grahame Grieve (Mar 29 2016 at 12:53):

thx, I will

Grahame Grieve (Mar 29 2016 at 22:08):

ok, build 8118 is the stable build for testing / review. There's now a 21 hour review period for fixing rendering problems, updating erroneous documenation. (once 8118 is completed)

Grahame Grieve (Mar 29 2016 at 22:16):

those of you that offered to help - reviewing the spec is best. my current action list:
- update RDF page to be correct
- update history page for the freeze
- change the version
- fix the v3 code system page
- task 8968
- task 8724
- task 9284 (maybe)

Rob Hausam (Mar 29 2016 at 22:17):

thanks - I'll help

Grahame Grieve (Mar 29 2016 at 22:27):

+ fix v2 oids

Brian Postlethwaite (Mar 29 2016 at 23:09):

I take it http://hl7-fhir.github.io/terminologies-v3.html is still rendering incomplete value set links?

Grahame Grieve (Mar 29 2016 at 23:09):

working on those pages now

Richard Ettema (Mar 29 2016 at 23:18):

I'll take the history page if you haven't done that yet.

Grahame Grieve (Mar 29 2016 at 23:25):

I haven't touched it, but what will you do with it?

Richard Ettema (Mar 29 2016 at 23:50):

This appears to be a high level list of substantive changes. So, list all new and modified resources to start. Beyond that I'm open to suggestions; e.g. list those FHIR domain areas that have undergone extension enhancements and modifications: CDS, Terminology (new value sets/code systems), etc.

Richard Ettema (Mar 29 2016 at 23:50):

Sound reasonable?

Grahame Grieve (Mar 29 2016 at 23:51):

we should particularly - and first - list breaking changes to the API and wire formats (I don't think there's been any)

Grahame Grieve (Mar 29 2016 at 23:51):

then we should list breaking changes to the conformance resources

Grahame Grieve (Mar 29 2016 at 23:51):

then we should list other resources that have had breaking changes (but not the changes)(

Grahame Grieve (Mar 29 2016 at 23:52):

then we list new resources

Richard Ettema (Mar 29 2016 at 23:55):

Ok. I'll work on those. Should just the conformance resources with breaking changes be listed or do want additional high level info on the changes?

Grahame Grieve (Mar 29 2016 at 23:57):

I think there's only been 2 breaking changes, and those are both worth being explicit about
- change to the way structure definition types work (constrainedType --> baseType, base --> baseDefinition)
- move CodeSystem out of ValueSet

Grahame Grieve (Mar 29 2016 at 23:57):

I may have forgotten some

Richard Ettema (Mar 30 2016 at 00:01):

Ok. I'll start with those 2 changes and then review the remain conformance resources for any others. Thanks.

Grahame Grieve (Mar 30 2016 at 00:03):

np

Richard Ettema (Mar 30 2016 at 01:06):

I just emailed you my updated history.html page for your review.

Richard Ettema (Mar 30 2016 at 01:19):

I just noticed a misspelled search parameter for EpisodeOfCare "claimindentifier"; s/b "claimidentifier".

Grahame Grieve (Mar 30 2016 at 01:19):

there's several. I'm not fixing them now. you can make tasks to do so

Richard Ettema (Mar 30 2016 at 01:21):

Sorry, I meant ExplanationOf Benefit. Ok. I'll create a tracker item for it. Thanks.

Richard Ettema (Mar 30 2016 at 01:39):

I guess that means you approve. :) your welcome.

John Moehrke (Mar 30 2016 at 14:07):

@Grahame Grieve The security.html, and security-labels.html are also pointing at nowhere... They need to be fixed with whatever magjic terminology fix.

Grahame Grieve (Mar 30 2016 at 18:30):

thanks

Grahame Grieve (Mar 30 2016 at 18:31):

which link in security.html?

John Moehrke (Mar 30 2016 at 20:11):

@Grahame Grieve it is the ones to vocabulary that are broken (404). You mentioned that absolute paths should not be used, but you didnt say what should be used. These pages are very old and have always used absolute paths. (your initial authoring).

Grahame Grieve (Mar 30 2016 at 20:29):

I fixed all the broken links I could find. there were lots in the security labels, but I didn't find any in the the security page itself

John Moehrke (Mar 30 2016 at 20:33):

agreed. I don't find anything broken on security.html either. sorry for that distraction. My search started with security labels.

John Moehrke (Mar 30 2016 at 20:35):

search.html has a link to the SUBSETTED vocabulary that fails. ... this was the initial failure that Kathleen noticed this weekend.

John Moehrke (Mar 30 2016 at 20:35):

in the _summary section

Grahame Grieve (Mar 30 2016 at 20:35):

got that one, thanks

Grahame Grieve (Jul 18 2016 at 09:24):

so how close are we to done?

Eric Haas (Jul 18 2016 at 09:49):

I'm done - tripped up by this issues with value sets:
and unfortunately only able to commit only a small portion of what I wanted.

Grahame Grieve (Jul 18 2016 at 09:55):

@Lloyd McKenzie you fixing the broken build?

Lloyd McKenzie (Jul 18 2016 at 10:20):

@Grahame Grieve Yes. Most recent commit should have fixed that issue and hopefully not introduced new ones.

Grahame Grieve (Jul 18 2016 at 10:20):

and are you ready to sleep?

Lloyd McKenzie (Jul 18 2016 at 10:21):

I wasn't necessarily expecting to

Lloyd McKenzie (Jul 18 2016 at 10:21):

But I will if you want to lock things down

Grahame Grieve (Jul 18 2016 at 10:21):

you can have a little longer. A little

Grahame Grieve (Jul 18 2016 at 11:36):

ok, finally, the we now have a substantiative freeze.

Grahame Grieve (Jul 18 2016 at 11:37):

from now on, no changes to resource/datatype elements, cardinaity, types, or codes in required code systems, without first discussing it with me here on this stream (where other people can express an opinion)

Grahame Grieve (Jul 18 2016 at 11:38):

there are few known miss-the-boat things; I or the editor committing will announce them here

Grahame Grieve (Jul 18 2016 at 11:38):

changes are still allowed to non-substantiative stuff for a week and to the module pages through QA period

Eric Haas (Jul 18 2016 at 16:20):

I'd like to rearrange the order of includes in http://hl7-fhir.github.io/valueset-referencerange-meaning.html so that hopefully the expansion put the 900 race codes after the others. Its very hard to read as it is now. I'm assuming that is ok.

Eric Haas (Jul 18 2016 at 16:20):

tracker 8941

Melva Peters (Jul 18 2016 at 16:35):

to confirm if we have non-substantive changes - for example, an update to a definition that provides clarification, can we comit those or do we need approval first?

Eric Haas (Jul 18 2016 at 16:36):

Yes worrds and examples too

Lloyd McKenzie (Jul 18 2016 at 16:41):

Non-substantive changes are permitted through to next Sunday. And technical corrections based on QA are allowed for the next 3 weeks

Eric Haas (Jul 18 2016 at 17:53):

here is one that is iffy. re updating diagnoisticrequest to follow request pattern I want to change the binding name of the requred status valueset so that is a generic request status for use by all request resource.

Eric Haas (Jul 18 2016 at 18:36):

I have a substantial one, finally got the value set issue sorted out on DeviceUseRequest and that is now aligned with the workflow request pattern as well. It a huge change so I can sit on it too.

Grahame Grieve (Jul 18 2016 at 18:58):

definitions are not frozen. what's the binding strength on the device use request binding?

Eric Haas (Jul 18 2016 at 19:12):

yeah that is the thing ...t he diagnostic request status binding is required. and he DeviceUseRequest is a major update and uses the same valueset. like I said it can wait ... its not lke its the last one that needs to be aligned.

Grahame Grieve (Jul 18 2016 at 19:14):

This is a change to DeviceUseRequest.status? Is there any profiles on this in the ballot scope?

Eric Haas (Jul 18 2016 at 19:35):

QI-Core is that Bryn? and I don't know about the CCDA profile -could not find it

Bryn Rhodes (Jul 18 2016 at 19:39):

The QI-Core profile there just adds a "reasonRejected". Any change you make to the underlying resource would just be absorbed, so no objections to any change you want to make there.

Eric Haas (Jul 18 2016 at 20:20):

@Bryn Rhodes I noticed those extensions. With the updates however see DiagnosticRequest there is no status = "rejected". The rejected status lives on the Task.

Eric Haas (Jul 18 2016 at 20:49):

still no objections?

Bryn Rhodes (Jul 18 2016 at 21:00):

I see. Yeah, no objections still, we'll move that extension as appropriate as part of updating the profiles.

Grahame Grieve (Jul 18 2016 at 21:01):

k Eric, go for it

Eric Haas (Jul 18 2016 at 21:03):

thanks

Gaston Fiore (Jul 18 2016 at 21:50):

@Grahame Grieve , the CG WG has a couple of new gForge items. Are our artifacts also under the freeze, or do we have until the 21st?

Grahame Grieve (Jul 18 2016 at 22:02):

what changes are you proposing?

Josh Mandel (Jul 18 2016 at 22:27):

On this subject: I think the discussion we had back in May was that since no outside workgroups are building on the Genomics resources/profiles, this shouldn't be a problem.

Grahame Grieve (Jul 18 2016 at 22:49):

well, I'm not entirely sure that's still the case, so wanted to be explicit about what is involved here to check

Josh Mandel (Jul 18 2016 at 22:52):

Agreed! Just providing some background.

Josh Mandel (Jul 18 2016 at 22:52):

Would be good to undestan from @Gaston Fiore what's involved.

Grahame Grieve (Jul 18 2016 at 23:09):

it's not clear to me who's leading the ballot preparation for the various IGs:
- SDC: Lloyd
- CQF : Bryn
- Quality (is this QI-Core?): ?
- CCDA: ?
- DAF + DAF Research: Eric/Brett

Grahame Grieve (Jul 18 2016 at 23:09):

can we fill out this list?

Grahame Grieve (Jul 19 2016 at 10:59):

@Bryn Rhodes - I've just had a nasty thought - what impact might fluent path changes have on the fhir spec? will we have to make any changes for a new version of fluent path?

Brian Postlethwaite (Jul 19 2016 at 11:08):

If any of the content in the invariant expressions or search path extractions is effected I would say so.

Grahame Grieve (Jul 19 2016 at 11:09):

that's what I'm asking - if any pertinent syntax or function names have changd

Gaston Fiore (Jul 19 2016 at 14:52):

Regarding CGWG new changes, I have so far:
http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=10378
http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=10379
There might be new ones after today's call as well.
@Grahame Grieve , I'll have to check how the source changed after you made a CG IG. When are you planning on undoing the IG since we don't qualify for one, meaning we belong to core?

Bryn Rhodes (Jul 19 2016 at 16:31):

Quality is QI-Core, yes, and for the actual publishing that's probably me (Claude is helping too).

Bryn Rhodes (Jul 19 2016 at 16:42):

On the FluentPath question, there are two that could potentially impact backwards compatibility of invariant expressions, yes.
.all() renamed to .allTrue()
string concatenation now uses an &

Grahame Grieve (Jul 19 2016 at 19:36):

ok I will make those changes during QA. Do you have a timeline for preparing the draft?

Grahame Grieve (Jul 19 2016 at 19:37):

Gaston - both those changes are to the profile, not to a base resource. Profiles are not frozen this week, so you can go ahead and make those changes

Gaston Fiore (Jul 19 2016 at 19:50):

Thanks for the clarification @Grahame Grieve . The third change, GF#10384, however, since is a change to the Sequence resource, is not allowed now?

Grahame Grieve (Jul 19 2016 at 19:53):

that is subject to freeze, yes. I believe we've established before that no one is profiling Sequence this time around?

Gaston Fiore (Jul 19 2016 at 19:54):

Correct, the only profile that existed got deleted.
http://hl7-fhir.github.io/sequence-profiles.html

Grahame Grieve (Jul 19 2016 at 19:56):

ok then that change is approved

Grahame Grieve (Jul 19 2016 at 19:57):

@Bryn Rhodes - ok, what is the plan for publishing QI-Core then? We need to migrate it to the new IG publisher, but do we leave the IG as close to it's current form as possible? or do we work it over somewhat? and what to do about the QUICK stuff?

Bryn Rhodes (Jul 19 2016 at 20:08):

Timeline on FluentPath updates is next week, this week I'm focusing on finishing up the changes to the CQF content.

Grahame Grieve (Jul 19 2016 at 20:10):

ok thaks

Bryn Rhodes (Jul 19 2016 at 20:12):

For QI-Core, we discussed that on today's CQF call and we are planning on updating the QI-Core profiles based on the STU3 content. I think in the interest of time, it makes sense to leave it as close to current form as possible, but I'm open to any changes. We'll also be submitting a white paper that has a CIMI archetype, a FHIR profile, and a StructureMap to transform from the CIMI to the FHIR.

Bryn Rhodes (Jul 19 2016 at 20:13):

We'd like that transform to actually be executable, but the first step is to do it as a thought exercise.

Bryn Rhodes (Jul 19 2016 at 20:14):

As far as QUICK, something about the changes to STU3 causes that tooling to hang now, and I haven't been able to determine the cause. So that's also on my list to fix so we can regenerate that, with the goal of distributing it as part of the QI-Core IG content.

Eric Haas (Jul 20 2016 at 01:00):

(deleted)

Grahame Grieve (Dec 05 2016 at 03:06):

Editors: 2 hours to freeze....

Grahame Grieve (Dec 05 2016 at 03:07):

thx

Grahame Grieve (Dec 05 2016 at 05:27):

ok, freeze is on - all commits, please discuss with me individually

Brian Postlethwaite (Dec 05 2016 at 07:04):

Can you please let me know when we can do the Client lib generations once we're stable and ready to go (version numbers etc)
Thanks

Grahame Grieve (Dec 05 2016 at 07:32):

ok I'll let you know

Grahame Grieve (Dec 05 2016 at 11:46):

@Brian Postlethwaite - all go. The build as committed has passed my QA tests

Brian Postlethwaite (Dec 05 2016 at 12:00):

Thanks

Grahame Grieve (Dec 05 2016 at 12:51):

wait - I missed something - I have to upgrade the version. So no substantiative changes, just a new version. will commit that tomorrow.

Grahame Grieve (Dec 05 2016 at 12:52):

Freeze remains in place for now

Grahame Grieve (Dec 05 2016 at 21:51):

ok. very big commit(s) coming up, taking the spec and all implementation guides to v1.8.0

Grahame Grieve (Dec 05 2016 at 21:53):

@Brian Postlethwaite - I saw somewhere comments about broken examples, but can't find that now?

Grahame Grieve (Dec 06 2016 at 01:24):

ok, I have tagged the current release as 2017Jan. That is the base for 1.8.0, and the Freeze is now lifted

Grahame Grieve (Dec 06 2016 at 01:25):

shortly - maybe in a day or 2 - I will get around to updating the current build to 1.9.0, but it's an exhausting process for me, so I'm taking a break from it today

Grahame Grieve (Dec 06 2016 at 11:28):

I changed the tag to a branch, because svn bitched at me about it. So if you want to work in the frozen version, you need \branches\2017Jan

Brian Postlethwaite (Dec 06 2016 at 12:20):

And for those on dotnet, there is a GitHub branch available, details in the dotnet stream.

Eric Haas (Dec 06 2016 at 19:12):

stillfrozen right?

Rick Geimer (Dec 06 2016 at 19:29):

Where are the IGs published post-freeze? I went to http://hl7.org/fhir/2017Jan/index.html, clicked on the IG link at the bottom left, but it took me to the old IGs. Specifically looking for the latest C-CDA on FHIR updates we made over the weekend.

Grahame Grieve (Dec 06 2016 at 19:35):

I haven't updated the IG registry yet - last thing for me to do. Some of the other IGs were not ready, and I've been tussling with them for 24 hours

Rick Geimer (Dec 06 2016 at 19:35):

Ok, thanks. Let me know when ready so I can review with SDWG.

Grahame Grieve (Dec 06 2016 at 19:37):

from implementers stream:
- fhir spec: http://hl7.org/fhir/2017Jan
- US Core: http://hl7.org/fhir/us/core
- CCDA on FHIR: http://hl7.org/fhir/us/ccda
- Meds Maturity Project: http://hl7.org/fhir/us/meds

Rick Geimer (Dec 06 2016 at 19:39):

Thanks!

Rick Geimer (Dec 06 2016 at 19:47):

Hmm. The DAF links are still not showing up. Grahame, is this something you are working on? Example:
http://hl7.org/fhir/us/ccda/StructureDefinition-CCDA-on-FHIR-Continuity-of-Care-Document.html
The Allergies section slice has a constraint on Composition.section.entry pointing to http://hl7.org/fhir/us/core/StructureDefinition/us-core-allergyintolerance, but it is not showing up in the rendered view.

Grahame Grieve (Dec 06 2016 at 19:52):

where exactly?

Rick Geimer (Dec 06 2016 at 19:55):

    <element id="111442">
        <path value="Composition.section.entry" />
        <sliceName value="section_slice_pos3" />
        <short value="entry" />
        <definition
            value="SHOULD contain zero or more [0..*] U.S. Core AllergyIntolerance Profile (identifier: http://hl7.org/fhir/us/core/StructureDefinition/us-core-allergyintolerance) (CONF:2219-2088)." />
        <min value="0" />
        <max value="*" />
        <type>
            <code value="Reference" />
            <profile value="http://hl7.org/fhir/us/core/StructureDefinition/us-core-allergyintolerance" />
        </type>
    </element>

Rick Geimer (Dec 06 2016 at 19:55):

But the differential rendering just shows this:

.. entry 0..* Reference() entry

Grahame Grieve (Dec 06 2016 at 19:56):

this cannot have been rendered properly in last month or so

Grahame Grieve (Dec 06 2016 at 19:57):

because you need to use targetProfile not profile.

Grahame Grieve (Dec 06 2016 at 19:57):

I think it's too late to deal with this

Rick Geimer (Dec 06 2016 at 19:58):

We were never able to test this because we just got the new US Core URIs.

Rick Geimer (Dec 06 2016 at 19:58):

If we change profile to targetProfile everywhere will it work?

Grahame Grieve (Dec 06 2016 at 19:58):

maybe

Grahame Grieve (Dec 06 2016 at 19:59):

understand that I have a few hours of post-processing to take the IG and post it to the web site.

Grahame Grieve (Dec 06 2016 at 19:59):

I *may* have time to do it again

Rick Geimer (Dec 06 2016 at 20:02):

Understood. We should discuss how to coordinate multiple interrelated IGs better in the future.

Grahame Grieve (Dec 06 2016 at 20:03):

we're going to. I don't want to repeat the last 24 hours.

Grahame Grieve (Dec 06 2016 at 20:22):

I've updated http://www.fhir.org/guides/registry now

Rick Geimer (Dec 06 2016 at 22:15):

Ok, I verified locally that changing profile to targetProfile for reference types now correctly renders the DAF links. Assuming you "may" have the time to do it again, should I check the changes into the current build or the 2017Jan branch when I wrap this up tonight?

Grahame Grieve (Dec 06 2016 at 22:41):

check them into both

Grahame Grieve (Dec 06 2016 at 22:41):

then let me know

Rick Geimer (Dec 07 2016 at 06:51):

Updated ccda2 files are now in both trunk and the 2017Jan branch.

Grahame Grieve (Dec 07 2016 at 19:49):

ok. posting the updated CCDA now

Rick Geimer (Dec 07 2016 at 19:50):

Much appreciated. Thanks!

Eric Haas (Dec 09 2016 at 00:25):

Are we unfrozen yet?

Lloyd McKenzie (Dec 09 2016 at 01:15):

We've been unfrozen for a few days

Grahame Grieve (Feb 20 2017 at 12:52):

ok, committers, we have entered the start of the R3 release period. From now on, no substantiative changes to the content without discussing them with me first. Narrative changes, re-working of modules etc is still allowed this week

Rob Hausam (Feb 20 2017 at 13:20):

I barely missed the deadline for GF#12880. It look longer to work through all of the details than I had hoped, but it is building correctly. It did require an update to igtools CodeSystemRenderer (among others).

Eric Haas (Feb 20 2017 at 16:38):

We still have an open ticket on applying a couple of pre-release LOINCs see https://chat.fhir.org/#narrow/stream/committers/subject/New.20LOINCs.20for.20STU3. The substantive part is adding them to the valueset

Grahame Grieve (Feb 20 2017 at 17:03):

We added them already, I thought. Please add them today

Eric Haas (Feb 20 2017 at 17:24):

OK

Eric Haas (Feb 20 2017 at 18:11):

Done

Eric Haas (Feb 22 2017 at 21:59):

Is an edit to the comment element in an operation definition a substantial change?

Grahame Grieve (Feb 22 2017 at 22:09):

no

Grahame Grieve (Feb 22 2017 at 22:09):

no more substantive changes *at all* for any reason. I will not approve anything; I have started (again) doing the downstream work

Eric Haas (Feb 22 2017 at 22:12):

To clarify:

• no - don't change the operationdefition comment element or
• no - not a substantive change

Grahame Grieve (Feb 23 2017 at 05:53):

Not a substantiative change, you can make that one

Bryn Rhodes (Feb 23 2017 at 23:18):

So is adding an example capability statement for a service considered substantive?

Lloyd McKenzie (Feb 23 2017 at 23:25):

Examples are not substantive. Substantive = anything that would require someone to write/change code.

John Moehrke (Feb 24 2017 at 14:26):

Is updating the _summary selection substantive? It seems it would be, but in the most minor way possible.

Lloyd McKenzie (Feb 24 2017 at 15:33):

It's substantive - involves changing code on all of the reference implementations

John Moehrke (Feb 24 2017 at 15:48):

that is what I worried about. It comes up in the QA check, where it was not something someone put in a formal comment. Thus it was not driven to the freeze deadline... Oh well, there is always STU4.

Grahame Grieve (Feb 24 2017 at 19:46):

reminder: 2 days until QA freeze.

Grahame Grieve (Feb 24 2017 at 19:49):

FMG will need to resolve exactly what is allowed to be deferred into the QA freeze, but I will be recommending that the following things are allowed to be changed once QA freeze starts:
- grammar, typos, spelling
- R2/R3 conversion scripts
- FMM levels
- module pages + home page
- Generated RDF definitions

Any other change will require approval by me in advance. Any substantiative change will require approval by FMG too.

Lloyd McKenzie (Feb 24 2017 at 20:20):

Graphical layout fixes are ok too.

Lloyd McKenzie (Feb 24 2017 at 20:21):

Spelling/gramar/typos can only be fixed without permission if they're non-substantive (e.g. not names of resources or elements)

Lloyd McKenzie (Feb 24 2017 at 20:22):

QA freeze applies to all FHIR core content, including profiles published as part of FHIR core. Separately published IGs are responsible for their own QA process (which means they can make changes into the next two weeks if they wish, recognizing that they're cutting into their own QA time).

Grahame Grieve (Feb 24 2017 at 20:22):

y

Cooper Thompson (Feb 24 2017 at 20:24):

Adding new w5 mappings is considered non-substantive as long as it doesn't force a re-ordering of the resource properties - correct?

Lloyd McKenzie (Feb 24 2017 at 20:25):

Adding/correcting mappings is fine. It'd be the re-ordering that would be substantive.

Bryn Rhodes (Feb 24 2017 at 20:57):

And just to double-check, workflow mappings are part of "mappings" there, right?

Grahame Grieve (Feb 24 2017 at 20:58):

yes

Grahame Grieve (Feb 24 2017 at 20:58):

though I expect that they'll be substantiative in the future

Lloyd McKenzie (Feb 24 2017 at 21:35):

I.e. In R4 we might generate code based on workflow mappings, but no limitations on making them for this release.

Brian Postlethwaite (Feb 26 2017 at 00:17):

The Consent resource code generates duplicate definitions for the Actor and Data backbone element classes.
Can I add in the classname hint to the spreadsheet, its not substantive from a wire definition format point of view, but does change what is code generated. (don't know if you even see it on the spec itself anymore)

Brian Postlethwaite (Feb 26 2017 at 00:18):

Will make the Actor and ExceptActor, and Data and ExceptData

Grahame Grieve (Feb 26 2017 at 03:37):

yes you can do that. but you should also fix your generator

Brian Postlethwaite (Feb 26 2017 at 03:41):

Thanks.
If I change that then classnames may change from what others already use where its not pre-defined. :(

Grahame Grieve (Feb 26 2017 at 03:43):

that's ok right now

Andy Stechishin (Feb 26 2017 at 22:56):

Are changes to the W5 mappings allowed during the freeze?

Grahame Grieve (Feb 26 2017 at 22:58):

maybe. what kind of changes?

Andy Stechishin (Feb 26 2017 at 22:58):

adding them to a resource

Grahame Grieve (Feb 26 2017 at 23:00):

can't be done now? changing w5 mappings have the habit of introducing hints and warnings

Andy Stechishin (Feb 26 2017 at 23:03):

So the answer is no? Trying to be certain I am operating within the boundaries. Currently the resource in question has 0 Warnings (I would work very hard to ensure that remains), so I would hope the W5 additions would have no effect on the warnings and informationals. The resource currently does not have the W5 warning.

Lloyd McKenzie (Feb 26 2017 at 23:31):

The primary issue with w5 mappings is that they may tell you the order of elements in your resource is wrong - which would be a substantive change to fix. No harm in adding them and seeing what happens. If it creates warnings that would require a substantive change to fix, then back them out and stay at FMM 2.

Andy Stechishin (Feb 26 2017 at 23:35):

Understand, as I said, not looking to push boundaries but trying to make sure I know what they are.

Lloyd McKenzie (Feb 26 2017 at 23:42):

There will probably be a few places where we at least explore pushing boundaries (e.g. artifacts that based on use ought to be FMM 5, but have a search parameter that's not following naming conventions that got missed). So go ahead and do the w5 work, see what the impact is and - if there's impact - decide how urgent it is to try to get that change into this release. FMG will talk about issues that have been brought forward on Wed.

Grahame Grieve (Feb 27 2017 at 03:31):

ok, everyone, last call for changes before QA freeze

Grahame Grieve (Feb 27 2017 at 03:34):

reminder: the following things are allowed during QA Freeze:
- grammar, typos, spelling
- R2/R3 conversion scripts
- FMM levels
- module pages + home page
- Generated RDF definitions

At least, I have recommended this list to FMG. This is the list of things I'll be working on during the freeze:
- fix snomed and loinc references (and snomed displays - see 11365)
- fix UCUM expansions (12438)
- check bread crumbs (8422, 8702)
- ordering in profiles.*.* (9109)
- fix glossy (9120)
- v2 value set versioning - see task 9961
- extension binding display = see 10552
- Add a jump past narrative to the XML & JSON rendering of resources (9969)
- example validation - 10425 + 12577
- glossary - 12306

Grahame Grieve (Feb 27 2017 at 03:35):

Other changes by approval by FMG. There's a bunch of search parameter issues that came up during final QA; Lloyd will be taking them to FMG for approval. That's the only substantiative change in the air that I know of

Lloyd McKenzie (Feb 27 2017 at 03:39):

QA can also include adding/enhancing/correcting example instances

Lloyd McKenzie (Feb 27 2017 at 03:41):

Also, implementation guides published outside the core spec are on their own - they're not part of the formal QA process, which in theory means they can continue to make substantive changes. That said, given that US-core is a dependency of many other IGs, it should probably be treated as frozen from a substantive perspective as of tonight.

Eric Haas (Feb 27 2017 at 06:40):

Thanks for the suggestion on US Core @Lloyd McKenzie but US-Core is not frozen yet. Need another day to resolve race and ethnicity code system and language code system issues.

Grahame Grieve (Feb 27 2017 at 06:42):

need a day when I'm available ;-)

Grahame Grieve (Feb 27 2017 at 06:48):

ok, so the main build is frozen for QA now.

Grahame Grieve (Feb 27 2017 at 06:48):

I'll do a run on the IG publisher and us-core tomorrow

Eric Haas (Feb 27 2017 at 06:49):

media is missing from resource page

Rob Hausam (Feb 27 2017 at 06:54):

could I still make a non-substantive change to ndc.html?
that's not on the qa freeze list, so I assume the answer is no?

Eric Haas (Feb 27 2017 at 06:55):

Well i still dont have a clue how to resolve the language tracker so that one may not change. for race and ethnicity checking with cdc on a potential url.

Grahame Grieve (Feb 27 2017 at 20:14):

@Rob Hausam you need FMG permission now. @Eric Haas which is the language tracker?

Eric Haas (Feb 27 2017 at 20:23):

http://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=11919&start=0 and others there are some emails too: lagnuage-emails.pdf to which you agreed to create a valueset which we could then use as an extensible binding. ( at least that is mine and @Brett Marquard understanding)

Grahame Grieve (Feb 27 2017 at 20:23):

this is a us core issue

Eric Haas (Feb 27 2017 at 20:24):

yes sorry all US-Core

Grahame Grieve (Feb 27 2017 at 20:24):

k. so no freeze there, and I'll work on language today

Eric Haas (Feb 27 2017 at 20:24):

thnks

Rob Hausam (Feb 27 2017 at 22:37):

I assumed I would need to get permission. I guess the question is if it's worthwhile to try to get that to add the additional text for the NDC page (GF#9555) at this point?

Grahame Grieve (Feb 27 2017 at 22:38):

I think it's worth it

Grahame Grieve (Mar 06 2018 at 02:20):

one of the editors missed making commits due to the storm in the NE. Does anyone have any issues with some late changes to device:

Grahame Grieve (Mar 06 2018 at 02:20):

Device-Resources-Feb-2018-V1.1.xlsx

Brian Postlethwaite (Mar 06 2018 at 11:44):

Doing another check of trackers outside the ballot, one has slipped through on Patient that is really a technical correction on a search parameter.
Time to slip that in still?
https://gforge.hl7.org/gf/project/fhir/tracker/?action=TrackerItemEdit&tracker_item_id=15554&start=0

Lloyd McKenzie (Mar 06 2018 at 15:43):

@Grahame Grieve Looking at the spreadsheet it's hard to tell what the change is

Grahame Grieve (Mar 06 2018 at 15:43):

I agree.

Lloyd McKenzie (Mar 06 2018 at 15:44):

@Brian Postlethwaite If it's not going to break any implementation guides, should be no issue. Will it affect any of the US Core or other patient implementation guides?

Lloyd McKenzie (Mar 06 2018 at 15:45):

@Grahame Grieve I think the proponents of the change need to provide a proper description of what's being changed.

Grahame Grieve (Mar 06 2018 at 15:47):

I have asked

Eric Haas (Mar 06 2018 at 16:39):

I have several issues with the proposed changes and how/whether they have been reviewed by the affected Workgroups and lack of details and since I am charged with applying them. I would like to hold off.

Eric Haas (Mar 06 2018 at 16:45):

@Rob Hausam ? what do you think?

Rob Hausam (Mar 06 2018 at 17:36):

I think the changes may be OK in and of themselves, but I agree that it's hard to know for sure looking at the spreadsheet. Even if the changes are OK, it's not clear to me what is being accomplished and why it's worth making the effort to do it. I think we need at least to get the further information that has been requested before thinking about proceeding, then the decision should be clearer (as if there isn't enough justification, then we wouldn't do it - or at least wouldn't do it now).

Grahame Grieve (Mar 06 2018 at 23:36):

we have agreed to just put a text note in to get balloter feedback

Eric Haas (Mar 09 2018 at 19:19):

Is the freeze for substantive changes still on. Assuming yes but seeing a couple of substantive commits

Grahame Grieve (Mar 09 2018 at 20:05):

yes it's still on - what did you see?

Eric Haas (Mar 10 2018 at 00:12):

not naming names but @Craig Newman :-) made this change: https://chat.fhir.org/#narrow/stream/committers/subject/fhir-svn.20.2F.20master/near/132845

Eric Haas (Mar 10 2018 at 00:15):

though to be honest ... I can't tell if it is just examples.

Bryn Rhodes (Mar 10 2018 at 15:38):

On that, adding an invariant is not considered substantive, right?

Lloyd McKenzie (Mar 10 2018 at 16:04):

Yes

Lloyd McKenzie (Mar 10 2018 at 16:04):

You need Grahame's permission

Bryn Rhodes (Mar 10 2018 at 16:09):

Understood.

Eric Haas (Mar 11 2018 at 18:58):

I just noticed that Observation.focus lost its TU flag. I am certain I applied that when I committed. can that be added before ballot?

Grahame Grieve (Mar 11 2018 at 20:29):

Yes

Bryn Rhodes (Mar 12 2018 at 00:07):

In updating the CDS documentation to reference CDS Hooks, I found a few items that I missed in DataRequirement. I completely understand if it's too late to apply these, but can I apply GF#15746, GF#15747, and GF#15748? The last two are canonical changes I missed, the first one adds a subject to DataRequirement.

Bryn Rhodes (Mar 12 2018 at 00:10):

Also, I swear I thought I applied GF#13515 a long time ago, but apparently I didn't. If possible, I'd love to squeeze that one in (just the DataRequirement change to add limit and sort elements).

Craig Newman (Mar 12 2018 at 12:44):

I made two changes last week. One was just adding STU notes PH and CDS WGs agreed to. The second one (the one Eric notes) is a change to a couple of elements to allow a choice between a positive integer and a string. It didn't feel all that significant to me but if someone wants to pull it, that's fine. I do still have a number of changes I'd like to make, mostly related to workflow patterns (but at least a couple of elements would be added to better match the patterns). I can either make those or hold off on them depending on what you all prefer.

My original intent had been to get these in place on March 2 (after they were WG approved), but real-life intervened and I wasn't able to make any changes until just recently. Sorry for any heartache or confusion. Please let me know if I should go ahead with the remaining workflow changes.

Lloyd McKenzie (Mar 12 2018 at 13:43):

Hi Craig - the issue is that IGs were working to build on top of the "frozen" content. Certainly don't roll-back now as all changes should now be frozen for QA. @Grahame Grieve is in transit, but can provide more guidance when he gets back on line.

Craig Newman (Mar 12 2018 at 17:40):

Understood. Sorry for the confusion. I'll make no more changes unless directed to by @Grahame Grieve

Eric Haas (Mar 12 2018 at 18:49):

(deleted)

Grahame Grieve (Mar 12 2018 at 19:30):

@Bryn Rhodes I believe that there's no reason for anyone to be profiling or using metadata types in their IGs. is that true?

Bryn Rhodes (Mar 12 2018 at 20:04):

Strictly speaking, that's true, but it depends a little bit on how much we can support in the spec. For example, I want to profile ValueSet to add a relatedArtifact extension, but that's only because the ValueSet resource doesn't have a relatedArtifact element.

Bryn Rhodes (Mar 12 2018 at 20:05):

I also want to profile EventDefinition to add a parameter extension (referencing ParameterDefinition), but again, that's only because the base resource doesn't have the element defined.

Bryn Rhodes (Mar 12 2018 at 20:05):

So I think yes.

Simone Heckmann (Mar 12 2018 at 21:42):

I just realized that when updating Encounter.hospitalization.origin from Reference(Location) to Reference (Location|Organization), I failed to do the same for Encounter.hospitalization.destination (which was part of the resolution of GF#15021), so I fixed this...
Hope that's ok...

Grahame Grieve (Mar 12 2018 at 21:42):

hmm. who is profiling encounter? this cycle?

Simone Heckmann (Mar 12 2018 at 21:49):

HSPC does

Simone Heckmann (Mar 12 2018 at 21:54):

Can't find any others (at least not from this list: http://www.fhir.org/guides/registry)

Grahame Grieve (Mar 12 2018 at 21:55):

place to look is this forthcoming ballot cycle - and guides may not be listed there.

Grahame Grieve (Mar 12 2018 at 21:58):

• "HL7 FHIR® Implementation Guide: Breast Cancer Data Logical Model and FHIR® Profiles
• HL7 FHIR® Implementation Guide: Clinical Genomics, Release 1
• HL7 FHIR® Implementation Guide: International Patient Summary, Release 1
• HL7 FHIR® Implementation Guide: Healthcare Associated Infection Reports, Release
• HL7 FHIR® Implementation Guide: LOINC – IVD Test Code (LIVD) Mapping,
  Release 1
• HL7 FHIR® Implementation Guide: Personal Health Device (PHD), Release 1
• HL7 FHIR® Implementation Guide: Validated Healthcare Directory, Release 1
• HL7 FHIR® Profile: Pharmacy; Medication, Release 1

Grahame Grieve (Mar 12 2018 at 21:59):

pretty likely that IPS profiles encounter. @Rob Hausam ?

Simone Heckmann (Mar 12 2018 at 22:04):

Can't find their IG, but FWIW, their frontpage doesn't mention encounters: http://international-patient-summary.net/

Simone Heckmann (Mar 12 2018 at 22:07):

According to the source files in SVN, Encounter happens in QICore and HSPC, neither listed for ballot

Rob Hausam (Mar 12 2018 at 23:42):

IPS hasn't done Encounter yet, although I think you're right that it's likely that we will. But right now we are on STU3, so the R4 changes won't affect us at the moment.

Grahame Grieve (Mar 13 2018 at 03:57):

ok Simone approved...

Grahame Grieve (Mar 13 2018 at 10:10):

@Bryn Rhodes approved

Grahame Grieve (Mar 13 2018 at 22:32):

@David Pyke there's a lot of consent tasks at 'change required'....

Grahame Grieve (Mar 13 2018 at 22:34):

@Rob Hausam where are with the about 20 vocab tasks waiting for to be applied?

Rob Hausam (Mar 13 2018 at 22:59):

There are two remaining open change required trackers from the Jan. 2018 ballot that remain to be applied
I can add the example for GF#14865 this evening
GF#14882 requires adding an invariant, and I assumed we probably couldn't do that at this point, but I'm happy to get it in if we can
I also intend to do the 4 terminology module ones (back from Sept. 2016!)
and then as many of the other 21 as is feasible in whatever time is left that will be allowed
I believe that at some you said you would look at GF#14118 (but it hasn't been assigned - we could assign it to you)

Grahame Grieve (Mar 13 2018 at 23:31):

you can apply the invariant today for GF#14882

Grahame Grieve (Mar 13 2018 at 23:32):

GF#14118 is on my list for today.

Grahame Grieve (Mar 13 2018 at 23:32):

I have some others on my list - let me know if you take them on

David Pyke (Mar 13 2018 at 23:57):

I should probably go back and close them all as completed. I'll do that tomorrow.

Grahame Grieve (Mar 13 2018 at 23:59):

ok that would be pretty useful if they've been made - use the status "applied"

Grahame Grieve (Mar 14 2018 at 00:18):

@Rob Hausam I guess we've missed the boat with GF#13869 ?

Rob Hausam (Mar 14 2018 at 00:55):

Well, maybe. I can still try to scan CP and see if we've missed something critical (don't think so?), but won't have a lot of time to spend on it.

Rob Hausam (Mar 14 2018 at 01:18):

I have most of the editing done on GF#15721 for SNOMED CT on FHIR, and am finishing that up.
And I've worked out with Stefan Lang and Julian Sass what's needed for GF#13210 for ICD-10 dual (multiple) coding and I want to get that in.

Grahame Grieve (Mar 14 2018 at 01:29):

ok you can commit both of them

Grahame Grieve (Mar 14 2018 at 02:07):

@Rob Hausam GF#13520 - did this go to harmonization? It's missed the boat either way right?

Rob Hausam (Mar 14 2018 at 02:17):

This will go to harmonization - which is next week. The initial proposal was submitted and is acceptable. However, there is this note from the tech review that says "However, is the workgroup aware that in the US the Joint Commission a year or two ago has disallowed the use of these codes for frequencies in hospitals?". I'm not sure what the implications of that exactly will be - at least not until next week. Maybe we should put the codes in now, and then we could remove them during QA if they end up not getting approved? Just a thought.

Grahame Grieve (Mar 14 2018 at 03:54):

no if they're going to harmonization, they'll be in the ballot.

Grahame Grieve (Mar 14 2018 at 03:54):

thanks for following through

Grahame Grieve (Mar 14 2018 at 04:21):

GF#13815 says to use a definition from the glossary project... I can't find any output that matches what I would expect to find.... other than several old glossaries that don't mention value sets, or provide a not-useful definition

Rob Hausam (Mar 14 2018 at 04:39):

For the glossary reference it says "if possible". I'm not too surprised we didn't find anything useful. I think that just making the first sentence say "A value set specifies a set of codes drawn from one or more code systems for use in a particular context." will probably satisfy this one.

Rob Hausam (Mar 14 2018 at 04:42):

Actually, it should probably be: "A ValueSet resource specifies a set of codes drawn from one or more code systems for use in a particular context."

Grahame Grieve (Mar 14 2018 at 04:54):

ok. thanks. With regard to GF#14118 - there's some confusion there. The HL7 table presentation is cross version. That's why it's not shown for a particular version. Instead, the version history is shown for each code. So I'm not going to resolve it by making it version specific. What I'm going to do instead is show the expansion with version if it's anything other than 'all the codes'

Grahame Grieve (Mar 14 2018 at 06:03):

I'd like to enter a full freeze for FHIR core spec at 4pm eastern tomorrow (FMG call). Any editors have a problem with that? (@Bryn Rhodes @Brian Postlethwaite @Eric Haas @Rob Hausam @Lloyd McKenzie @David Pyke @John Moehrke )

note that this is the start of QA, so there's one more opportunity for QA type changes at the end of QA

Grahame Grieve (Mar 14 2018 at 06:29):

my outstanding task list right now:

• GF#6255
• GF#13042

David Pyke (Mar 14 2018 at 11:53):

My changes are to my local only. I'll push them once the ballot is done. So Freeze away!

John Moehrke (Mar 14 2018 at 13:32):

I am done...except for that resource identifier thing you just dropped on us? (Which I am not clear about).

Bryn Rhodes (Mar 14 2018 at 14:32):

Are we okay to add examples during the QA period?

Lloyd McKenzie (Mar 14 2018 at 15:21):

Yes

Bryn Rhodes (Mar 14 2018 at 15:27):

Then yes, I think I'm good. Looking at QA output now to see why we're not hitting our QA targets.

Rob Hausam (Mar 14 2018 at 17:54):

@Grahame Grieve For GF#14118 I guess that can work. But I'm not sure how useful it is to to mix all of the versions in a single expansion. If someone was actually going to use it I think they would need "all codes" for a specific version. But we can see how it works out (more ballot comments if needed!). I will try to have all of the important stuff done by 4:00 PM Eastern, but it's pretty tight and I'm not sure that it will all make it. Had to take time out for the dentist this morning. :(

Grahame Grieve (Mar 14 2018 at 19:37):

FHIR expansion is based on the latest version. Don't support expansions of older versions in FHIR itself. UTG has all the definitions of all past versions, so if people really want older versions, they can go look in UTG

Grahame Grieve (Mar 14 2018 at 19:40):

for an example of v2 expansion: http://build.fhir.org/v2/0131/index.html

Grahame Grieve (Mar 14 2018 at 19:41):

@Rob Hausam - what have you yet to commit?

Rob Hausam (Mar 14 2018 at 20:13):

ICD page
and the example for GF#14865
beyond that it would be whatever else you want me to do (if anything) - or I can stop whenever needed

Rob Hausam (Mar 14 2018 at 20:15):

oops - the 4 terminology module trackers, too - if we can still squeeze it in

Grahame Grieve (Mar 14 2018 at 20:16):

ok. will talk to FMG and advise

Grahame Grieve (Mar 14 2018 at 21:15):

@all. ok freeze is on. Please talk to me from now on before making any commits.

Grahame Grieve (Mar 14 2018 at 21:18):

Rules for content changes after QA: changes are ok (as long as approved by committee, of course) if they do not impact on IGs. So:

• grammar / typos/ spelling changes
• reworking modules, fixing module pages
• adding / fixing / extending examples
• fixing / clarifying narrative

These things are all ok.

Making narrative based conformance rules, or changing structures: not ok if they impact on IGs without approval from IG authors. Reconciliation of what is ok or not should happen by discussion here

David Pyke (Mar 15 2018 at 14:31):

I should have said that all approved changes are in the build. My apologies.

Michelle (Moseman) Miller (Mar 16 2018 at 12:57):

When QA changes are allowed (later this month), will the following types of changes be allowed since they should all be non-substantive?

• updating/removing workflow mappings
• adding workflow exemptions to suppressed-workflow-warnings
• updating w5 mappings (provided it does NOT change the sequence of elements)

Lloyd McKenzie (Mar 16 2018 at 14:09):

Workflow exemptions is allowed right now. The other two are fine after freeze (and with Grahame's permission can probably be done during QA, at least for non-normative candidates).

Grahame Grieve (Mar 22 2018 at 12:20):

reminder for all IG editors:
- 2018-04-01 - Final content freeze - no changes except with permission of product director

This deadline is for both core spec and all IGs

Grahame Grieve (Mar 22 2018 at 12:44):

Also, all IGs undergoing ballot must have a value for the "npm-name" in the config json file. Editors must ask for the name to use for the IG. This means you must all have a license in the json file too. All HL7 IGs are

  "license" :  "cc0-1.0",

Eric Haas (Mar 22 2018 at 14:55):

note that

"npm-name" : "[node Package Name - see below]",
"npm-template" : "[Optional npm package template file]",

can't just sit there with a null value or you get this

Stack Dump (for debugging):                                                      (32.0223sec)
java.io.FileNotFoundException: /Users/ehaas/Documents/FHIR/IG-Template2 (Is a directory)
    at java.io.FileInputStream.open0(Native Method)
    at java.io.FileInputStream.open(FileInputStream.java:195)
    at java.io.FileInputStream.<init>(FileInputStream.java:138)
    at java.io.FileInputStream.<init>(FileInputStream.java:93)
    at org.hl7.fhir.igtools.publisher.Publisher.generateNPMPackage(Publisher.java:2822)
    at org.hl7.fhir.igtools.publisher.Publisher.generateSummaryOutputs(Publisher.java:2743)
    at org.hl7.fhir.igtools.publisher.Publisher.generate(Publisher.java:2285)
    at org.hl7.fhir.igtools.publisher.Publisher.createIg(Publisher.java:395)
    at org.hl7.fhir.igtools.publisher.Publisher.execute(Publisher.java:354)
    at org.hl7.fhir.igtools.publisher.Publisher.main(Publisher.java:4152)
Exception in thread "main" java.lang.NullPointerException
    at org.hl7.fhir.igtools.publisher.Publisher.main(Publisher.java:4162)
Erics-Air-2:IG-Template2 ehaas$

Since I have a ig.json template this creates a issue for me. Can ig-pub ignore it if not filled?

Rob Hausam (Mar 22 2018 at 17:24):

how (or who) do we ask for the npm-name?

Chris Moesel (Mar 22 2018 at 17:40):

According to IG Pub docs,

This name is assigned by the FHIR product Director

So... I think that's @Grahame Grieve ?

Rob Hausam (Mar 22 2018 at 17:43):

yes - not sure if its OK (or best) to ask here, or by email, or otherwise?

Grahame Grieve (Mar 22 2018 at 19:19):

@Eric Haas what does this mean? "can't just sit there with a null value or you get this"

Grahame Grieve (Mar 22 2018 at 19:42):

@Rob Hausam - "yes - not sure if its OK (or best) to ask here, or by email, or otherwise?" - email please.

Eric Haas (Mar 22 2018 at 20:40):

basically I would lke the npms parameters to work with value = null like this null works for broken links...

  "special-urls": [],
  "npm-name": null,
  "logging": [],
  "npm-template": null,
  "structuremaps": [],

causes an error whereas

"broken-links". null

does not

Grahame Grieve (Mar 22 2018 at 20:45):

why not just leave it out then?

Eric Haas (Mar 22 2018 at 23:10):

I can scrub it if null.

Grahame Grieve (Mar 22 2018 at 23:34):

great

Sarah Gaunt (Mar 29 2018 at 09:41):

@Grahame Grieve - do I need to get a value for npm-name from you for the HAI IG to put in the config json file? And add that licence?

Sarah Gaunt (Mar 29 2018 at 09:47):

Have added "license" : "cc0-1.0", - just need the npm-name... There might be a few minor tweaks to the IG tomorrow but other than that, it's pretty much good to go for ballot.

Grahame Grieve (Mar 29 2018 at 09:53):

is it auto-building?

Grahame Grieve (Mar 29 2018 at 09:53):

npm name: us-hai

Sarah Gaunt (Mar 29 2018 at 09:55):

is it auto-building?

Yes

Grahame Grieve (Mar 29 2018 at 09:56):

great

Sarah Gaunt (Mar 29 2018 at 09:57):

Taking forever to auto-build though - is there something up with the terminology server?

Sarah Gaunt (Mar 29 2018 at 09:58):

Keeps failing.

org.hl7.fhir.r4.utils.client.EFhirClientException: Error sending Http Request: Read timed out
    at org.hl7.fhir.r4.utils.client.ClientUtils.sendRequest(ClientUtils.java:310)

Grahame Grieve (Mar 29 2018 at 09:59):

hmm

Sarah Gaunt (Mar 29 2018 at 10:02):

And then it works again. All good now.

Grahame Grieve (Mar 29 2018 at 10:03):

I restarted the server

Sarah Gaunt (Mar 29 2018 at 10:04):

Oh, cool. Though it's been doing that on and off all evening - failing with that error and then resurrecting itself... I'm done anyway... :-)

Grahame Grieve (Mar 29 2018 at 10:05):

wonderful

Grahame Grieve (Apr 02 2018 at 12:17):

well, I'm running a few hours behind.... so I'm going to sleep shortly. Once I wake up we'll actually go into freeze for all content and I'll start processing. The first biggest task I have to do is upgrade the version to 3.3. So freeze will be in about 6 hours or so

Grahame Grieve (Apr 02 2018 at 20:03):

I've just run into a few infrastructural issues updating the FHIR version... looks like a few hours work... so freeze is still a few hours away

Grahame Grieve (Apr 02 2018 at 20:58):

ok. all of it is done. Freeze is on. For the next 30min, commits with my approval only. And then nothing...

Rob Hausam (Apr 02 2018 at 21:02):

(deleted)

Grahame Grieve (Apr 02 2018 at 21:40):

if you're interested in following my progress, this is my tracking sheet:

Grahame Grieve (Apr 02 2018 at 21:40):

https://docs.google.com/spreadsheets/d/1J8Dl1OQJCqcvMoZZh5OYXwlmG5kt4NuefwA3QwbbRtg/edit?usp=sharing

Grahame Grieve (Apr 02 2018 at 22:02):

@Rick Geimer @Sean McIlvenna what version should be the CCDA IG have when I publish it?

Grahame Grieve (Apr 02 2018 at 22:09):

@Brett Marquard what version should the onc meds IG have when I publish it?

Grahame Grieve (Apr 02 2018 at 22:09):

ah - I see 1.0.0 - so that, yes?

Grahame Grieve (Apr 02 2018 at 22:12):

@Rick Geimer @Sean McIlvenna @Sarah Gaunt the specified version of HAI-guide is "1". I don't think that this is correct for a ballot publication.

Grahame Grieve (Apr 02 2018 at 22:15):

Core is on full freeze now. no changes until I fork

Eric Haas (Apr 02 2018 at 22:26):

@GG just a reminder that Argo-Scheduling is going to http://www.fhir.org/guides rather than hl7.org/fhir/ - not part of ballot.

Grahame Grieve (Apr 02 2018 at 22:27):

yes. I'm doing all the builds etc at the same time because other than that everything is the same

Eric Haas (Apr 02 2018 at 22:28):

Thanks - I can edit the fhir.org registry once you are finished.

Sarah Gaunt (Apr 02 2018 at 22:32):

@Grahame Grieve I put 1 because the release number that Lynn gave us is 1: https://www.hl7.org/special/Committees/tsc/ballotmanagement/pi_index_nibs.cfm?ProjectNumber=1192

Grahame Grieve (Apr 02 2018 at 22:44):

so that will be the final number of the publication. it's not the technical version. I will call this 0.1.0

Sarah Gaunt (Apr 02 2018 at 22:47):

Ok, cool - thanks

Sarah Gaunt (Apr 02 2018 at 22:50):

Why are there errors on my IG? It was error-free last week when I finished up??? Has something changed?

Grahame Grieve (Apr 02 2018 at 22:50):

I changed something. I haven't the actual errors yet

Grahame Grieve (Apr 02 2018 at 22:50):

@Rob Hausam - I am changing the IPS version from 1 to 0.1.0

Grahame Grieve (Apr 02 2018 at 23:05):

@Rick Geimer @Sean McIlvenna the CCDA guide has another problem: it refers to ServiceRequest, which doesn't exist in R3, but it is otherwise an R3 implementation guide

Grahame Grieve (Apr 02 2018 at 23:15):

@Sarah Gaunt yes both errors caused by changes I made. I will fix

Grahame Grieve (Apr 02 2018 at 23:24):

@Eric Haas can you review the broken links in the argonaut guide?

Rob Hausam (Apr 03 2018 at 00:12):

no problem - I thought about that, and forgot to do it

Eric Haas (Apr 03 2018 at 00:39):

by review I assume you mean fix. just discovered that git ignores case changes in directories and file names.

Sarah Gaunt (Apr 03 2018 at 00:51):

@Grahame Grieve Thanks! I'm in Port Douglas on holiday - if there are any issues you need me to fix can you either skype or email - might not see Zulip...

Eric Haas (Apr 03 2018 at 01:06):

@Grahame Grieve I fixed one link the rest I believe are due to a ig-publisher error in how it handles markdown for extensions. Here is an example :

• Markdown in profile: [Foo](patient-scheduling.html#use-case-3-prefetching-open-slots results in correct relative address in html:
  http://build.fhir.org/ig/argonautproject/scheduling/patient-scheduling.html#use-case-3-prefetching-open-slots

• same Markdown in extension
  [Foo2](patient-scheduling.html#use-case-3-prefetching-open-slots)
  results in an absolute address to the main build:
  http://hl7.org/fhir/STU3/patient-scheduling.html#use-case-3-prefetching-open-slots

Eric Haas (Apr 03 2018 at 01:08):

I brought this issue up earlier here: https://chat.fhir.org/#narrow/stream/IG.20creation/topic/IG.20build.20rendering.20issue

Eric Haas (Apr 03 2018 at 01:20):

for US-Meds should be 1.1.0 not 1.0.0 , it was not updated from the prior release.

Grahame Grieve (Apr 03 2018 at 01:22):

k. just updated it to 1.1.0

Eric Haas (Apr 03 2018 at 01:23):

Need to update the title, footer and publish box too - can I do that now

Grahame Grieve (Apr 03 2018 at 01:24):

yes please

Grahame Grieve (Apr 03 2018 at 01:48):

@Eric Haas I don't see how this can possibly be ok:
<comment value="If the expression is a FHIRPath expression, use the fhirPath Expression extension."/>

Grahame Grieve (Apr 03 2018 at 01:49):

grr. it processed the markdown. you can't put {{}} in the content of a resource

Eric Haas (Apr 03 2018 at 01:55):

really? .... :-(

Eric Haas (Apr 03 2018 at 02:09):

OK i'll work on those links and help brett out with some of the broken links in US-Meds

Grahame Grieve (Apr 03 2018 at 02:13):

tanks

Grahame Grieve (Apr 03 2018 at 02:26):

you want the correct path to the main spec, right, @Eric Haas ?

Eric Haas (Apr 03 2018 at 02:32):

that is the thing. In some cases the main FHIR spec which it defaults to. In other cases to the IG pages. for the conformance resources I just changed it to the canonical uri. The relative paths seem to depend on the element for descriptions they are relative to the IG. For the comments the main fhir specification. Anyway I have zero bad links now.

Grahame Grieve (Apr 03 2018 at 02:34):

where do you have zero bad links?

Eric Haas (Apr 03 2018 at 02:35):

local - pushing in a minute

Eric Haas (Apr 03 2018 at 02:35):

for scheduling

Grahame Grieve (Apr 03 2018 at 02:36):

k

Eric Haas (Apr 03 2018 at 02:54):

scheduling links all fixed

Grahame Grieve (Apr 03 2018 at 03:11):

thanks.

Eric Haas (Apr 03 2018 at 03:36):

I can't figure out the problem for all the bad links for US-Meds they all resolve for me and mirror what we did in US Core. Most are in CapabilityStatement Narrative.

for example:

/Users/ehaas/Documents/FHIR/US-Meds/output/CapabilityStatement-server.html#/html/head/link/body/div/div/div/div/div/div/ul/li/a at Line 204, column 64
    error

The link 'http://hl7.org/fhir/STU3/http.html#read' cannot be resolved

but (http://hl7.org/fhir/STU3/http.html#read) resolves just swell.

Grahame Grieve (Apr 03 2018 at 03:58):

have you tried updating the cached igpack.zip?

Grahame Grieve (Apr 03 2018 at 06:05):

no. the URLs in config.json were wrong. I've fixd most of the broken links

Grahame Grieve (Apr 03 2018 at 11:38):

ok. Freeze is lifted for the main build. Not for any of the IGs yet.

Grahame Grieve (Apr 03 2018 at 11:38):

editors are welcome to review the ballot publication here: http://hl7.org/fhir/2018May/index.html

Grahame Grieve (Apr 03 2018 at 11:53):

also, FHIRPath 1.1. Normative ballot is published here: http://hl7.org/fhirpath/2018May/index.html

Eric Haas (Apr 03 2018 at 17:54):

@Grahame Grieve TY ... I can fix those last bad links if you want.

Grahame Grieve (Apr 03 2018 at 18:08):

which bad links?

Eric Haas (Apr 03 2018 at 18:56):

the last 5 on us meds

Grahame Grieve (Apr 03 2018 at 19:01):

ok sure

Grahame Grieve (Apr 03 2018 at 19:01):

thx

Grahame Grieve (Apr 03 2018 at 19:39):

@Sarah Gaunt http://hl7.org/fhir/us/hai/index.html

Eric Haas (Apr 03 2018 at 21:03):

US Med is broken link free

Grahame Grieve (Apr 03 2018 at 21:05):

thx, publishing now..

Grahame Grieve (Apr 03 2018 at 21:56):

@Eric Haas http://fhir.org/guides/argonaut/scheduling

Grahame Grieve (Apr 03 2018 at 22:18):

@Chris Moesel http://hl7.org/fhir/us/breastcancer/

Grahame Grieve (Apr 03 2018 at 23:48):

@Brett Marquard @Eric Haas http://hl7.org/fhir/us/meds/2018May

Grahame Grieve (Apr 03 2018 at 23:48):

@Lloyd McKenzie http://hl7.org/fhir/uv/genomics-reporting/index.html

Grahame Grieve (Apr 03 2018 at 23:49):

editors: please check your IGs, and indicate your approval of them.

Eric Haas (Apr 04 2018 at 01:59):

Scheduling looks good except that:

• the canonical uri for the ConceptMaps are not resolving from the description element in the valuesets
• THIS errorMapping from http://fhir.org/guides/argonaut-scheduling/ValueSet/visit-type to (not specified) in http://www.fhir.org/guides/argonaut/scheduling/ConceptMap-visit-snomed-map.html not sure why is 'not specified' should be taken from:

 <targetUri
             value="http://fhir.org/guides/argonaut-scheduling/ValueSet/sct-services"/>

• in http://www.fhir.org/guides/argonaut/scheduling/ConceptMap-visit-snomed-map.html http://snomed.info/sct doesn't resolve should be https://www.snomed.org/ no error in validator though?.
• http://fhir.org/guides/argonaut-scheduling/ValueSet/sct-services --> bad code "77132007" -> "277132007"

Eric Haas (Apr 04 2018 at 02:02):

I can fix 3 and 4 and maybe 2 will be magically ok after that. I can't fix 1 and I imagine that will go away when the redirects are in place. is that OK to fix now or can do later in a future rev?

Grahame Grieve (Apr 04 2018 at 02:12):

canonical urls - a to do for me. We can patch this in later

Eric Haas (Apr 04 2018 at 02:12):

US Meds looks good to me. But @Brett Marquard might want to add the 'Not open for May 2018 ballot' box to the cap statements. Nothing major though.

Grahame Grieve (Apr 04 2018 at 02:13):

validator won't raise an error for snomed ct issue. They were supposed to sort this out

Eric Haas (Apr 04 2018 at 02:13):

They being Snomed?

Grahame Grieve (Apr 04 2018 at 02:14):

y. I just pinged them about it

Grahame Grieve (Apr 04 2018 at 02:16):

I don't know why the not specified thing happens

Eric Haas (Apr 04 2018 at 02:18):

Could be because the code is wrong in the valueset?

Grahame Grieve (Apr 04 2018 at 02:21):

doesn't look like it. But the code says it can't happen ;-)

Grahame Grieve (Apr 04 2018 at 02:21):

debugging now

Eric Haas (Apr 04 2018 at 02:24):

I almost forgot the history page for us meds, I'll update that locally and push to the repo.

Grahame Grieve (Apr 04 2018 at 02:25):

I edited that manually.

Grahame Grieve (Apr 04 2018 at 02:25):

is there a problem with it?

Eric Haas (Apr 04 2018 at 02:35):

here is the updated html

<div class="col-9">

<h2>All Published Versions of the US-Meds Implementation Guide</h2>
<p>
The US-Medication FHIR Implementation Guide promotes consistent implementation of
the pharmacy FHIR resources in US Realm Electronic Health Record Systems (EHRs)
to provide patient and provider access to patient medications.
The following versions of US-Core have been published:
</p>

<table class="grid">
  <tbody><tr>
    <td><b>Date</b></td>
    <td><b>US-Meds Version</b></td>
    <td><b>FHIR Version</b></td>
    <td><b>Description</b></td>
  </tr>
  <tr>
    <td colspan="4" style="background: #EFEFEF"><b>Current Versions</b></td>
  </tr>
  <tr>
    <td><a href="index.html">June 14, 2017</a></td>
    <td>1.1.0</td>
    <td>3.0.1</td>
    <td>Current Official Published Version<i>(Currently: STU 1)</i></td>
  </tr>
  <tr>
    <td>(last commit)</td>
    <td>(last commit)</td>
    <td>3.0.1</td>
    <td>Current Development build (may be incoherent and change rapidly)</td>
  </tr>
  <tr>
    <td colspan="4" style="background: #EFEFEF"><b>STU 2 sequence</b></td>
  </tr>
  <tr>
    <td><a href="STU1/index.html">April 4, 2018</a></td>
    <td>1.1.0</td>
    <td>3.0.1</td>
    <td>May 2018 Ballot version</td>
  </tr>

  <tr>
    <td colspan="4" style="background: #EFEFEF"><b>STU 1 sequence</b></td>
  </tr>
  <tr>
    <td><a href="STU1/index.html">June 14, 2017</a></td>
    <td>1.0.0</td>
    <td>3.0.1</td>
    <td><b>STU 1</b>  (Permanent Home - v1.0.0)</td>
  </tr>
  <tr>
    <td><a href="2017Jan/index.html">Dec 6, 2016</a></td>
    <td>n/a</td>
    <td>1.8.0</td>
    <td>1st Draft</td>
  </tr>
</tbody></table>

</div>

which outputs this....

Eric Haas (Apr 04 2018 at 02:36):

Screen-Shot-2018-04-03-at-7.35.34-PM.png

Eric Haas (Apr 04 2018 at 02:36):

(in atom previewer so the background is off the second bar is the STU2 sequence)

Grahame Grieve (Apr 04 2018 at 02:38):

the (not specified) comes from the resource directly.

Grahame Grieve (Apr 04 2018 at 02:39):

you can either remove the narrative so it is regenerated, or fix it by hand...

Eric Haas (Apr 04 2018 at 02:41):

Oh I'm sorry , I forgot I manually created the narratives. Can I fix that and the bad codes ?

Grahame Grieve (Apr 04 2018 at 02:42):

yes

Eric Haas (Apr 04 2018 at 02:43):

(there was some issue with the generated narrative and I 'd rather manually fix it rather than risk it popping up again)

Lloyd McKenzie (Apr 04 2018 at 02:44):

Genomics looks good. I found an error while reviewing, but an issue with the source, not the publication process and far from critical

Eric Haas (Apr 04 2018 at 02:44):

What about the history for US-Meds - do you want me to commit to build as history.html or just take from here?

Grahame Grieve (Apr 04 2018 at 02:50):

I think the history page as posted is fine

Grahame Grieve (Apr 04 2018 at 02:50):

thx Lloyd

Eric Haas (Apr 04 2018 at 03:01):

OK fixed errors in terminology for Scheduling. Thanks Grahame

Grahame Grieve (Apr 04 2018 at 03:28):

k. reuploaded - check please

Lloyd McKenzie (Apr 04 2018 at 03:34):

Is it hard for you to re-load if I were to commit the fix?

Grahame Grieve (Apr 04 2018 at 03:54):

not so much hard as just taking time

Brian Postlethwaite (Apr 04 2018 at 05:24):

Hardly important, but the Exchange section doesn't have an image in the nav bar
http://hl7.org/fhir/2018May/exchange-module.html

Grahame Grieve (Apr 04 2018 at 05:29):

ah yes. that's google searches fault - the search for exact image size is broken

Brian Postlethwaite (Apr 04 2018 at 05:40):

Is the image actually there?

Grahame Grieve (Apr 04 2018 at 05:40):

no. I couldn't find one and thought I'd get back to it when the stupid search was working, but forgot

Brian Postlethwaite (Apr 04 2018 at 05:40):

lol. Cool.

Grahame Grieve (Apr 04 2018 at 05:42):

and it's still not working

Grahame Grieve (Apr 04 2018 at 05:57):

so there's a bug in the validator that an editor figured out how to exploit, and so now this serves as a good test for reference implementations....

Grahame Grieve (Apr 04 2018 at 05:57):

plandefinition-example-cardiology-os.xml:

 <action>
    <action>
      <!--    Patient disposition and triage group    -->
    </action>
    <action>
      <title value="Consults and Referrals"/>

Grahame Grieve (Apr 04 2018 at 05:57):

in json, this becomes:

Grahame Grieve (Apr 04 2018 at 05:58):

  "action": [
    {
      "action": [
        null,
        {
          "title": "Consults and Referrals",

Grahame Grieve (Apr 04 2018 at 05:58):

test out the robustness of your parsers.... I'll figure out why the validator missed that, but.. should I fix the ballot, or let it stand?

Chris Moesel (Apr 04 2018 at 12:05):

Thanks, @Grahame Grieve! I noticed that you added a "History" menu item. Should I have done that or is that an expected step in your publishing task? I also noticed that on the history page, none of the nav links are preserved, so you have to hit "back" button to get back to the IG.

Chris Moesel (Apr 04 2018 at 12:07):

I'll have the content team take a look to confirm the rest of the IG looks right.

Grahame Grieve (Apr 04 2018 at 12:07):

I don't mind creating the link. But I committed it back to github .

Grahame Grieve (Apr 04 2018 at 12:08):

not sure what "none of the nav links are preserved' means. If you mean, the links in the navbar... indeed. because the history page is not part of any particular publication, it doesn't have internal links to any of the publications

Chris Moesel (Apr 04 2018 at 12:21):

Thanks. Yes, I was talking about the links in the navbar (Home, Profiles, etc.). Are we talking about the same history page? It _looks_ like the history page is specific to our IG, and I noticed that US Core seems to have links in its history page navbar (although it looks like maybe they custom built that page). If it's my job to build out that history page (which I didn't), then that's ok -- I just wanted to understand the situation.

I also noticed, however, that at the bottom of our history page it says:

© HL7.org 2016+. C-CDA on FHIR Implementation Guide Version Directory

Grahame Grieve (Apr 04 2018 at 12:22):

oh crap

Grahame Grieve (Apr 04 2018 at 12:24):

well, they slipped the navbar thing by me. I think it's probably a bad idea in the long run. As for who maintains it... I hand edit it when I do the uploads. But I do reconcile content if the project is maintaining one. There's no expectation that you will though

Grahame Grieve (Apr 04 2018 at 12:26):

I'll fix the history footer once the hl7 ftp server starts accepting my conections again

Chris Moesel (Apr 04 2018 at 12:40):

OK. Understood re: the navbar. I hope I didn't get the US Core guys in trouble. I like them! Thanks for taking care of the history footer!

Brett Marquard (Apr 04 2018 at 13:06):

CHRIS! No problem we can fix on next update

Brett Marquard (Apr 04 2018 at 13:07):

US Meds -- i just committed an update to add 'not open for ballot' banners to the capability statement pages. It would be great to have those updated, but understand if too late

Eric Haas (Apr 04 2018 at 13:19):

I'm not clear what is wrong with our history pages navbar, when I create the IG I create the history page like any other so we can review it locally. I assumed it got moved outside the specification for publishing. Also , more important, I'd l ike to know who is the editor who exploited the validator bug. :-)

Eric Haas (Apr 04 2018 at 13:30):

Scheduling looks good to go. thanks though the the history page is the same as for US-Core. Not sure why this is a problem but need to work out a process for the editor maintaining this and moving to its permanent home.

Chris Moesel (Apr 04 2018 at 14:02):

@Eric Haas -- I think the problem, as I understand it, is that the history page is supposed to be outside the scope of any particular IG version. It's an index of IG versions. If it has nav links, which version do those nav links go to? The most recent official release? The most recent ballot release? The current build? I think that's the concern.

Eric Haas (Apr 04 2018 at 15:13):

Yes I see the issue there, But I still want editorial control over the content of the page and would like to be able to access and update it as well. We will need that workflow piece worked out so is off of GG's plate.

Lloyd McKenzie (Apr 04 2018 at 15:34):

There's a character set issue on the ballot and current release pages that isn't on the CI build. For an example, look at the snapshot tab of these: http://build.fhir.org/ig/HL7/genomics-reporting/obs-gen-panel.html http://hl7.org/fhir/uv/genomics-reporting/2018May/obs-gen-panel.html http://www.hl7.org/fhir/uv/genomics-reporting/obs-gen-panel.html

Lloyd McKenzie (Apr 04 2018 at 15:34):

The flags are wonky in the last two

Grahame Grieve (Apr 04 2018 at 19:54):

@Chris Moesel history text updated

Grahame Grieve (Apr 04 2018 at 19:54):

@Rob Hausam IG is posted now

Grahame Grieve (Apr 04 2018 at 19:56):

@Lloyd McKenzie fixed now

Grahame Grieve (Apr 04 2018 at 19:58):

@Eric Haas argonaut scheduling history is fixed. US core history should be. But it's not clear how the history file should be managed; these sorts of issues are why I manage it carefully

Grahame Grieve (Apr 04 2018 at 20:37):

ok Freeze is lifted globally. But you all need to create a tag on the current version marking it as the basis for the 2018May ballot.

Grahame Grieve (Apr 04 2018 at 21:36):

i tagged them. now that I figured out how

Eric Haas (Apr 04 2018 at 22:04):

??? what does this mean in practical terms for editors committing to the build?

Grahame Grieve (Apr 04 2018 at 22:07):

you can go ahead now. Any changes will go in to build.fhir.org but have missed the boat for ballot - that's published and closed now

Lloyd McKenzie (Apr 04 2018 at 22:49):

If you're making substantive changes to normative content, make sure you submit a ballot on it. Making substantive changes to normative content without ballot is something we should be cautious about.

Grahame Grieve (Apr 04 2018 at 23:00):

I think it goes beyond cautious - why would that be valid now?

Lloyd McKenzie (Apr 04 2018 at 23:03):

I'm not clear if there's a strict prohibition in HL7 Essential Requirements. If you have a need, we'll have to investigate whether it's possible - and what hoops need to be jumped through - if any.

Brian Postlethwaite (Apr 05 2018 at 05:43):

That plandefinition-example-cardiology-os.xml resource doensn't round trip alright!
Get's a parsing exception in the .NET fhir client json serializer.

Rick Geimer (Apr 06 2018 at 18:56):

@Grahame Grieve C-CDA on FHIR should be version 1. Regarding ServiceRequest, I changed to that from ProcedureRequest when that was renamed in the current build, but am fine rolling it back to ProcedureRequest if that gets the thing published.

Grahame Grieve (Apr 06 2018 at 19:26):

ok

Grahame Grieve (Apr 06 2018 at 21:57):

@Rick Geimer ok published at http://hl7.org/fhir/us/ccda

Rick Geimer (Apr 09 2018 at 15:25):

@Grahame Grieve Thanks!

Last updated: Apr 12 2022 at 19:14 UTC