| HL7 Terminology (THO) Publication | | - RM
- One of the many things the TSMG is going to be involved in is approving the publication of terminology.hl7.org (a.k.a. HL7 Terminology a.k.a. THO) which is a point in time look at the HL7 Terminology that it not included within the core standards.
- RM prepares to show the CI build of THO.
- TK
- I want to extend a personal thank you to GG.
- I downloaded release 82 of the IG Publisher and for the first time ever the THO build now reports zero errors and zero warnings.
- JB
- RM provides an overview of the THO website and how to find the QA report to find errors and warnings.
- TK provides an overview of the publication history of THO.
- TK then proposes that the next be 3.0.0 and offers to explain why.
- TK
- We changed the user interface significantly.
- We have broken out HL7 Code Systems from External Terminologies as two separate high level tabs.
- The External Terminologies tab has sub tabs named All Content, Code System Metadata Records, Code System Technical Identifiers, Code System Fragments, Complete Codes systems and Retired Content.
- The entries in the both the Code System Metadata Records and Code System Technical Identifiers are also separated according to those which are managed by the HTA and those which are not i.e. those which came in from the V2 and V3 imports.
- There is consultation underway to see which of these need to be retained and which needs to be retired.
- RM demonstrates that those items not deemed of value will move under the Retired Content tab.
- TK
- The significant change is the introduction of the Identifier Systems tab at the top.
- We have now added support for identifier systems with the first ones that got brought in through UTG rather than going directly to the FHIR pages.
- FHIR has a number of HTML pages that do not have the identifiers system metadata information captured in NamingSystem resources which this system (THO) does.
- The plan and decision from Vocabulary WG is for the the FHIR ones to move from the FHIR core specification to THO.
- To prepare for that, we needed to get the "landing zone" for that move in place so that you can begin the move.
- I am not suggesting that we do the move as part of the 3.0.0 release, but this is where they will land.
- Lastly, there is a tab at the top named Pending which is intended to be automated as an easy way for authors to notify the community participating in the CI build process of UTG and THO and to show that they're working on a code system that they intend to be in THO.
- But it's not ready yet, it's not complete and it's not been vetted by the project, but people should know it's coming
- If they want t know more about it, to give appropriate information, this is where it will go.
- I think this is inappropriate to be in a release, because this is notionally part of the CI build process for indexing.
- Others may disagree, that's fine.
- I don't know whether this should be removed for a new release, or left in the CI build or left in the release and I'm asking for that decision from TSMG.
- RM
- (Referring to the current CI build with no content in the Pending tab), it's easy for this one – it's empty.
- I'm going to suggest that we remove this, for now, because there's nothing in it and there's no value in including it in a published release this time anyway.
- And then we investigate whether we want to include it when we actually understand it better, and we have something in it.
- RD
- JB
- RM
- With no other comments, I'll take that as a support for that position.
- I think out vote is that we would vote on this being published without pending for now.
- I do think, for now, what we would do is that we would vote to publish 3.0.0 but then that needs to go with me to TSC.
- And then I need to find out if TSC always wants to have the final "we're going to vote", like they do for other things.
- I assume they're going to say yes to that, but perhaps they don't then when we vote you can publish.
- But for now, I don't know that we can do that, so we would just be voting to say it is ready to publish.
- TK
- It is ready to publish.
- Now there's just a couple of other questions of details around that.
- Basically, I have to coordinate with GG a time to do the publication.
- Because we do not yet have it as a one-button / turn-the-crank and it all just works.. yet
- There's still some little funnies that happen in the scripts and I'm hoping this one will go through without a hitch.
- But always whether it's a server side problem or a script problem or something else, something has gone awry every time.
- I have no reason to expect that won't happen again, and so I will need GG's assistance and so we will coordinate the time.
- I am also working very very part time this weekend and next week so I won't be able to start publishing it before the end of the next week anyway.
- RM
- LN
- If we publish a new version of THO, what impact will that have on currently published IGs built against FHIR R4 e.g. if a value set changes when we do this and that value set is referenced from an IG that is published right now, will it inherently absorb the changes? Will this cause a server to get updated and then, it will dynamically and immediately behind the scenes impact IGs that use that value set that is being updated in this IG?
- GG
- The answer is complicated because it depends on how the implementation guide is bound to the value set.
- Some cases, if they did a non-version specific binding and an update release of THO will cause an updated value set or code system to be used.
- But in other cases it won't.
- And that's by the validator. It won't affect the previously published IG in terms of what's published.
- But if you look at the expansions given it says the expansion is correct at a given point in time.
- And if you don't bind into a version specific value set, then the expansion is implicitly got an update when new versions of UTG are published.
- RM
- An IG's access to what's in THO that is not bound to a specific version is it always looking at the CI build or is it only looking at the Curren published snapshot of THO?
- GG
- RM
- Help me understand what depends on what.
- TK
- This is a hard problem and I went down the rabbit hole to figure it out.
- The best way to think of it is that the IG is generated rendering of the expansions that you see in the IG get generated based upon a composite value set vocabulary that has build during IG publisher processing that takes input from:
- core.
- THO
- The IG and
- a number of special parameters in the IG that may reference other depends on linkages
- That's where some of the depends on comes from.
- Every time that the same URI occurs for the same object, then the one that remains in the composite is the one with the latest version ID once that composite set is built and the IG is generated.
- So, if it's not a version specific object that's referenced in the IG, it is the one that is the most current from all of those sources.
- If it's not found in that composite set, then it looks in tx.fhir.org and sees if it's part of the publisher server infrastructure and then it happily generates it based on that content from there.
- If it's not found in any of those places it gives an error.
- RM
- The order is published THO, then CI THO - does it look at both?
- TK
- Only the published version, unless there's a parameter requiring the CI build.
- GG
- By default, an IG that doesn't reference HL7 Terminology as a dependency gets the published release - the last published release of HL7 Terminology – as its dependents which defines a version of any non-version value of references.
- If it explicitly references the CI build it depends on the CI build but only if it explicitly does so.
- If you reference values that in a previous version of HL7 Terminology at present you'll get an error and it's on my to do list to fix.
- RM
- That speaks to THO being a snapshot — only one version — it doesn't have the older versions and that's part of what's going on here.
- In order words, when the IG needs an older version of a value set or code system that's not in the current or latest published version of THO, the publisher won't be able to and GG's working on fixing this.
- GG
- Yes - mostly correct.
- There is something special about THO in that you don't need to depend on it.
- It's implicit that you depend on it.
- It's the only IG that's like that.
- As part of me fixing the version issue, it may be that IG's will need to depend on the enter version of THO that contains that value set. And that will raise questions as to whether you've implicitly frozen every version of every rerelease to a value set. This is what the Germans have done.
- RM
- I can tell you that this is very similar to what the USA does for its ECQMs. Essentially that it assumed that when it does a reference, it's a static reference to a series of expansions.
- GG
- I will propose a resolution but won't resolve this without coming back to Vocabulary WG or the TSMG.
- LN
- Is there an example of a static binding in FHIR?
- GG provides the following example in chat
- Version-non-specific binding: <binding><strength value="required"/><valueSet value="http://terminology.hl7.org/ValueSet/xxx"/></binding>
- Version-specific binding: <binding><strength value="required"/><valueSet value="http://terminology.hl7.org/ValueSet/xxx|1.2.3"/>
- RD
- The approach taken in Australia by an large is a bit different to THO in that we have things versioned individually – not as a whole group or an entire release but that's also because we tend to keep thing in FHIR terminology servers which just get added to continually and there's no idea of a sort of single release like we have with THO.
- As things get added to as they become available so it's a bit more real time.
- Whether or not that model will work for THO remains to be seen.
- But it's just another on of the considerations.
- RM
- It's just another way to tackle the problem.
- We should never do something that would preclude doing what you're doing in Australia.
- GG
- On the External Terminologies tab, the name "technical identifiers" will be misleading for people given exactly what this contains.
- Can we change it to "Code System Identification" ?
- Everyone agrees.
- LN points out some issues with wording on the Code System Metadata Records and Code System Technical Identifiers pages.
- TK
- If attendees don't think the release is ready we can hold off and make the necessary changes to improve clarity.
- RM
- Given that publication cannot occur till the next of next week, maybe we can see what we can do.
- For those who have an interest in doing some word smithing, put in a UTG ticket.
- JB
- This could be a pro-forma tickets to avoid consensus review.
- RD
- The changes would need approval from HTA which can also occur next week (Wednesday).
- JH
- Are we considering putting THO in brackets after the HL7 Terminology at the top because a lot of people still don't know what "THO" is.
- RM
- That is also a possibility.
- TK
- (To all attendees): Summarize the change, create a ticket, I will put it in the right resources if you create a ticket and have a good description of what it is you want change. Not huge changes small clear, unambiguous description, create a UTG pro-forma ticket and assign it to me.
- GG
- If I was going to make a motion, it would be that we go ahead and publish this as soon as we can, addressing any word smithing issues that we can but not holding up the release for them and but still that we make them for future releases.
- RM
- That's what I was just suggesting but just suggesting that these word smithing tickets be added as pro forma for review next week.
- TK
- The sponsor can be TSMG or HTA and this are the ones that need to approve that the changes are good to implement.
- RM
- So please add your tickets with TSMG as the sponsor and we will review at next week's meeting.
- JB
- We talked in the past about wanting to get some changes to CVS and CDC Race and Ethnicity in.
- We are very close to getting those approved with HTA so I don't think it will be a problem, but just wanted to make it clear that we might want to staff until we can get those in the CI Build.
- RD
- We should be able to get these approved via email by next week's TSMG meeting.
- PJ
- Is this a Vocab meeting or a TSMG meeting because i'm looking at an agenda on Confluence and i'm not seeing any of this on the agenda.
- This is all good stuff but surely this is about Vocab putting in issues for UTG – it's not really for TSMG. I thought we were discussing things like merging.
- RM
- This is a TSMG discussion because we have to vote on the release of THO.
- These are things that were germane to the release of THO and this item moved to the front because of our guests.
- PJ
- At a high level that does make sense, but some of the individual issues.
- RM
- We will come back to this in a week.
|
Administration | All | - HTA
- DG provides an overview of the last HTA meeting.
- CM provided a detailed and accurate description of the situation without proposing an opinion to the HTA.
- CM and I then provided our opinions about what we thought might happen, and then we talked to the other three members about the HTA and it was expressed that there are some concerns.
- The HTA function is very well and very clear at this point and there is concern from two primary perspectives.
- One, to ensure that the process that actually working fairly well and well defined at this point persistent and
- Two, had to do with the fact that the people on the HTA are terminologists and that this is not so for TSMG and thus joining the TSMG may be dilutive and there may be some loss of focus and excess duties.
- CM and I were clear about the fact that we didn't think there'd be a lot of extra duties.
- Nonetheless we did not come to any sort of conclusion
- There was also concern about the HTA being asked to make a decision when the TSMG hasn't sorted out its own processes and roles, etc.
- SM
- Of concern is TSMG taking on what the HTA is doing well enough while the TSMG is not yet very organised.
- RD
- (To DV) Can you comment on the three points in the request?
- RM
- What was just said does not line up with the request.
- It sounds like the HTA wants to remain separate.
- SM
- No, we want more information and would come into the TSMG if we knew that the HTA process could be taken on and continued.
- RD
- Points out that the request is actually consistent with what SM and DV are saying in that the HTA would continue doing what is is very effectively doing, but as a subcommittee of the TSMG.
- RM
- And so it has always been my assumption that if HTA was to become a part of the TSMG it would operate in the ways that are currently was other than before it was a function of the Board under the control of TSC, we will have to make GOM changes to have the HTA function as a subcommittee of the TSMG but continue to function in the same way.
- PJ and RM discuss how changes to the GOM are made and agree that the TSC need to recommend the changes to the EC who then recommend it to the board.
- RM
- I also understand that the HTA is like — TSMG are slogging, why do we want to be a part of you — and I think that's a valid concern.
- Other than the fact that it's a part of you know, the the the the overall scope that has been stated as the TSMG scope, so if HTA remain separate from the TSMG as currently specified by the GOM, which is that it is a TSC reported committee then we are going to have issues and that we're going to have to work to try and talk to each other.
- We don't want our groups to find out about through our TSC reps or whatever it just seems like a long arc.
- RD
- I don't have a problem with with the HTA request.
- My firm belief is that having three terminology-related committee / workgroups isn't good for our community.
- We should be able to consolidate like everywhere else, and just have a methodology group and a management group.
- So, on that basis i'm very supportive of us making moves to get the transition going now and, specifically, so that as part of this merge, if you will, those HTA members who wish to become members of the TSMG have the opportunity to do so if they so choose.
- And i'm also comfortable with interim arrangements, but, for the first year that was members that come over if all they want to do is what they've been doing as part of HTA they can very happily continue doing that, under the TSMG umbrella.
- I do think we need to signal to the community that we acknowledge that three is too many and there's no reason that external engagement cannot become a function of the TSMG.
- RM
- So if the HTA is requesting that they be incorporated as a subcommittee and I would really like to also hear from CM.
- SM
- I really like what RD said, where we need to make sure that you know that there's a management group and things at different parts of us to do different things to make sure that it gets done because we can all do the same thing so but i'm not a voting member of this but I I would be thrilled to be a part of TSMG, just thrilled.
- RM
- SM
- I think you need to talk to CM.
- RM
- Does HTA want TSMG to bring a request to TSC to make the HTA part of TSMG?
- DV and SM
- RD
- So here's what I will do as Vice-Chair of the HTA, I am going to call an out of session HTA meeting to ratify this request.
- RM
- If you can make that happen by next meeting, that would be great.
- RD
- We will try our best.
- I would also like to hear views/input on HTA being part of TSMG.
- JB
- I agree with HTA being incorporated.
- I think it will be tough, knowing that a lot of them are in Europe as as PJ stated in the chat and with the assumption that we plan to form those sub task groups and that they would be able to focus on the activities that they are on a bi weekly basis.
- RM
- Agreed - it's tough no matter what.
- I am vey interested in having Europeans, Australia, New Zealand, etc. but it's gonna inconvenient for somebody.
- RD
- I also get the sense that, if this request is carried forward, it may well be the case that that the European members of the HTA may not want to attend the main call that often as long as they can do the HTA call.
- LN
- I think it sounds like a good idea to allow them to be a sub group and establish their own sort of cadence and time to work and maybe they have they have someone whose job it is to be the connector and report up and bring issues for governance up to this group as needed.
- All the meetings, all the time they could you know organize themselves more efficiently with a spokesperson, or something.
- RM
- I do think that process could be something we could follow through on other major sub groups that are working on activity too.
- LN
- Yeah, the only thing you have to you have to worry about is, if we add them and their count changes are quorum requirements yeah.
- BR
- I honestly don't feel qualified to answer that question.
- I really don't have enough insight into what HTA does, to be able to say you know I mean sounds to me like a good idea, but I don't I don't know that my input is relevant.
- DV
- I really think it's a good idea to have the HTA be a part of this group, I think I honor what and address the concerns with regards to not note not diluting the established process and would look to.
- You know, do carry on in a way, such that that process and those assignments kind of stay somewhat you know protected and that we benefit from the inclusion of the mindset where that you know and the mind the trust.
- Rather than a abuse that and have you know access or additional things to do so i'm very much a proponent of bringing that that group.
- I'm also very happy to confer the the the Chair position to the chair of the HTA, and so I am and i'm enthusiastic about this and I shared that actually at the meeting.
- MF
- I am going to abstain, I don't think I am deep enough into what HTA does and doesn't do.
- PJ
- My main driver is I don't think it's makes any sense to have both HTA and TSMG reporting to the TSC - there should only be one.
- RM
- So what I heard, I think, is that all the folks who voiced an opinion are in support of the HTA joining the TSMG.
- As a subcommittee it would continue to do the work that it's currently doing and I have separate meetings would report into the TSMG, instead of the TSC for the first year.
- RD
- So only DV commented on the HTA Chair — CM — becoming a co-chair of the TSC – can we take a minute to just consider that point, which is part of the request?
- I think that's a great idea. Apart from CM obviously being really capable co chair and would be great for TSMG, I think that would really help the transition over the next year of making HTA a sort of built in part of TSMG rather than just subgroup as it will be to begin with.
- So I don't personally have an issue with that at all.
- RM
- Does anybody else have any comments with regards to that particular issue any concerns with that being a locked-in co-chair slot?
- No responses raised
- RM
- I'll take that as a no.
- You know i've expressed an interest in being a Co chair so has RD and CM and DV who would be coming from the HTA.
- If we accept CM as a co-chair, then I am going to remove my nomination as a co chair and I would suggest that we petition the TSC to allow us to have three and I think we just vote for it, and tell them that's what we've done and see if they slap on the wrist.
- I say that for a couple of reasons number: One I am persuaded with regards to discussions that have occurred I don't know if they've happening on the TSMG call that RD would be not only a great co chairs current co chair, but having an affiliate is a benefit, but we can't lose three co-chairs in Vocabulary WG in one fell swoop.
- TSC has also made clear that they're open to the idea of potentially not having to enforce the requirement that TSMG co-chairs have to resign as Vocabulary WG co-chairs — we'll see how far that goes.
- It was also under consideration that RD would not need to lose his Vocabulary WG co-chair position.
- If I am not a TSMG co-chair, then I would not need to resign, but CM will have to.
- That's one approach to this that I would endorse.
- Does anybody have any kind of after meeting the comments, because we are actually officially after meeting?
- RD
- i'll say quickly that all of that is something I would endorse as well, but I think we should postpone any assumptions about whether or not any of the existing Vocabulary WG co-chairs would need to give up their positions necessarily.
- think in our last year's TSMG meeting Wayne was very clear that this there's lots of flexibility, especially in the first year.
- So I think the focus really needs to be installing the three TSMG co-chairs as quickly as possible and also to decide who the TSMG rep to the TSC will b.
- And so my last point is, I would endorse that you RM remain on the TSC as the TSMG rep.
- RM
- Right well and I would endorse that too so well
- RD
- One one last thing, I would like to let everyone know on the infrastructure side the TSMG listserv lists have been created, so if you haven't joined, please join and you'll get all forthcoming meeting conference call invitations once that side of things are sorted out.
- RM demonstrates how to find the listserv page on the HL7 website.
|
