Cross-team building block standup meeting minutes 15-07-2021

Hello All,

Once again, please make sure your team’s documents are accessible from Teams from the appropriate working group’s folder in Microsoft Teams. If you need help with this, let me know!

If you have a use case that connects multiple building blocks, please link to it in the Logical Process Blueprint so other teams can find it: Logical process blueprint - Google Documenten

Agenda for this meeting:
• Review each group’s progress (standup!)
• Review progress on PPTs
• Reminder about using the building block definition template
• Reminder about writing down significant changes and decisions in a log under a header in your building block definition template
• Reminder to provide source for all images diagrams, ideally draw.io, web sequence diagrams so they can be edited and evolved.
• Review last week’s key decisions and see if they’ve been addressed
• Work on the next sprint’s logical process outlines and flow diagrams

Key decisions made in this meeting:

  • All use cases that connect multiple building blocks should be linked from the Logical process blueprint: Logical process blueprint - Google Documenten
  • Transactional system of record registries MUST use a document-based approach and MUST copy all records, e.g. a doctor’s record must be copied into a snapshot as part of a prescription. If you store a foreign key, e.g. doctor’s reg number and their phone number or address changes the record will be incorrect. Therefore, documents stored in system of record registries MUST copy all the data onto the document to provide a record as a snapshot in time, and SHOULD retain references to all foreign keys/references to objects copied for auditing, e.g. to the doctor’s address and phone number record URIs. This capability will be part of the registration building block.
  • Groups should show a first cut of their powerpoints at this monthly review meeting with Ramkumar’s suggested format and template.
  • Groups will provide source for all images diagrams, ideally draw.io, web sequence diagrams so they can be edited and evolved. If possible, check them into your group’s forked copy of GitHub - GovStackWorkingGroup/BuildingBlockAPI to provide versioning consistent with your document versions.
  • Groups will submit initial revisions of their building block specifications for review by the Architecture group as soon as possible.

Previous key decisions:

  • Security highlighted the need for a Networking group to handle networking-related issues, specifically firewalls, reverse proxies that have been identified as key components. They will work on an OpenAPI/Swagger specification for talking to an API Gateway building block, see below:
  • Information Mediator showed an excellent diagram highlighting a more detailed overview of how requests coming into Govstack will be handled: https://drive.google.com/file/d/1bp0DkRnITod0h96jIf6-AL2ks6NfGaoP/view
  • Payments is debating whether or not they will handle cash or not. They will handle mobile to mobile and bank payments. If they decide not to handle cash/teller/reconciliation/accounting functions, we will need a separate group for that functionality.
  • All groups will review the Data Verification and Validation - Unconditional Social Cash Transfer draft use cases at Logical process blueprint - Google Documenten
  • Groups will look at https://discourse.govstack.global/ and see how it fits their needs. See the Architecture Team Resources topic for an exampe of how groups can list their resources for public consumption. Architecture will continue to evaluate discourse.
  • Groups will fork a copy of GitHub - GovStackWorkingGroup/BuildingBlockAPI so they have a centralized, versioned repository for editable diagrams and other resources
  • Next week, Architecture will give an overview and demo of how we use github for editing diagrams

Outstanding tasks

  • Architecture will work on a flow diagram outlining how webhooks work with Information Mediator and will meet with Information mediator team to review
  • Payments will move language specific to their group from the Logical Process Blueprint Logical process blueprint - Google Documenten and link to it
  • All groups will review link to the cross-cutting requirements of the Security building block definition Security Building Block Definition.docx - Google Documenten. Groups will review additional functional requirements and point to those as well. Groups are free to consult with the Architecture and Security groups once this is done.
  • All teams still need to work on updated powerpoints
  • Groups can request a formal review from the Architecture group at any time.

We opened with standup, this time reviewing each team’s building block specification updates, questions and changes. Highlights included:

  • Architecture continues to work on clarifying the interactions between existing applications and the GovStack System and how existing DPGs can become GovStack-compliant. We talked about today’s call with OpenIMIS and their work on a FHIR to OpenAPI adapter that could be very useful for GovStack.
  • Information mediator gave an overview of their progress and the group gave feedback.
  • We talked in depth about system of record/transactional registries, see Trevor’s logical model slide deck eGovernment - Logical Building Blocks v0.3.pptx - Google Presentaties
  • Registries will ensure they support systems of record/documents and reported they may grow the team.
  • We reviewed the Working group cookbook for onboarding new members and groups: Working group cookbook - Google Documenten
  • Architecture showed how https://app.diagrams.net/ can be used to edit a file right on GitHub, via the File → Import → GitHub menu option and browsing to repository to easily edit a diagram.
  • Trevor suggested we make a video recording presenting the onboarding document, including the how and why of forking the GitHub - GovStackWorkingGroup/BuildingBlockAPI example repository and editing/creating diagrams there.

Thanks to everyone that attended!

Action items for each working group:
• Continue to work on your own copy of the Building Block Definition Template Building Block Definition Template.docx - Google Documenten. At least the Key Digital Functionalities, Cross-cutting requirements and Functional Requirements sections should be filled in based on the specific requirements outlined in the logical process blueprints: Logical process blueprint - Google Documenten and functional components in the logical process checklists for the use cases already covered (e.g. registration, payments and case management for Postpartum and Infant Care): Logical process checklist - registration - post partum and infant care.xlsx - Google Spreadsheets Logical process checklist - payments - postpartum and infant care.xlsx - Google Spreadsheets and Logical process checklist - case management - postpartum and infant care.xlsx - Google Spreadsheets.
• Ensure your Building Block Definition Template and other documents are accessible via Teams so other groups can see them.
• Review the logical process blueprint for the current sprint (Case Management - Postpartum and Infant Care) and add any comments or questions that come up: Logical process blueprint - Google Documenten
• Confirm the components for your building block designated in the logical process checklist are correct: Logical process checklist - case management - postpartum and infant care.xlsx - Google Spreadsheets
• Review the building block flow diagrams for Logical process blueprint - Google Documenten UC-P-USCT-001: Payment - Unconditional Social Cash Transfer (bank payments) - Google Documenten and UC-P-USCT-002: Payment - Unconditional Social Cash Transfer (non-electronic/cash payments) - Google Documenten
• Identify folks who could be good for the new working groups we’re spinning up: consent management, messaging, marketplace, scheduling and workflow!

Next week we can review our progress and perhaps begin another sprint!

Agenda for next week:
• Review each group’s progress (standup!)
• Review group’s documents in the context of how a specific arrow in a flow diagram maps to specific functionality in the block
• Reiterate that use cases are not to be taken as gospel, they can be rewritten and new use cases can be created that more accurately reflect reality
• It’s fine to break a building block into multiple blocks that better map to functional domains
• Reminder to provide source for all images diagrams, ideally draw.io, web sequence diagrams so they can be edited and evolved
• Work on the next sprint’s logical process outlines and flow diagrams
• Github and discourse overview

From the chat:
Information mediator bb:

Security BB:

Please link to the cross-cutting security requirements from your doc, and refer to any specific security requirements by linking to those.
GovStack System Architecture:

https://openimis.org/
15:02
Taylor (InfoMed)
layer 1 - secure communication layer
layer 2 - pubsub facility
workflow may often use something like openfn with an OCL adaptor:

15:16
Please add links to all use cases for review in:

Trevor’s logical semantics deck:

Here is the workgroup cookbok for onboarding new members:

16:18
psramkumar@apcogsys.net
bye
16:40

Meeting minutes (Cross-block meeting minutes has the recording and transcription):
Taylor 0:03
So we can we can at least talk about our next steps we

Unknown Speaker 0:07
talked about where were we going,

Taylor 0:09
because we’re not super far from. We’re not far from I think asking for, sort of more formal interview.

Max Carlson 0:17
Okay well that’s exciting. Yeah, Max. How

Taylor 0:21
are you,

Max Carlson 0:22
I cannot complain. Doing well, um I managed to sneak in a little bit of camping. So that was good. Yeah, yeah I got out I got out of the city back now. Um, so yeah but everything’s good. Yeah, pretty exciting. We’re looking at onboarding some additional groups and trying to figure out how all of that process is gonna work out. So that’s super exciting. Yeah. Yeah. So let’s see, uh I guess we’ll wait a few minutes, see who else can join.

Taylor 1:07
I’m gonna go on mute for one second, I just think it was like finishing up an email.

Max Carlson 1:11
No worries,

might just be a slow week I don’t know I mean I think I know a lot of people around here on vacation, so we’ll see. Might be a really quick meeting.

All right, well, Um, let’s see, I guess we might as well just go through the agenda. You know right, why not, uh, so let me just pull up the notes. So, uh, I guess we could just go through and talk about each group’s progress. Um, so given that there’s just, you guys and security group we can do that pretty quickly. Do you want to just show your documents and show what you’re, you know, or maybe your PowerPoints and talk about your progress or what you’re doing next.

Guess I was muted, so it could be a pretty short meeting. It could just be us so why don’t we just go through the, you know, the agenda. So the first thing I’d like to cover is just, you know, his cover the progress information mediators made

Ramkumar 4:41
sure. Okay.

Max Carlson 4:44
On the building block specification. Yeah.

Taylor 4:50
Right now, we’re basically we. I’m in the process of moving some of our guy and Graham, this is like into web sequence diagrams. I’ve got a bunch of diagrams set up and Lucid Chart and have either links to them or are just images of them in the, in the mediator spec itself. And at Ramkumar is working on mapping the building block to an example use case, but we actually do have four hubs, but for the secure communication layer. We do have, sort of, like, fairly detailed specs laid out already. And then I guess Aleksandre on the, on your plate still is, is sort of this around transaction logging or exception failure handling is that right,

Aleksander 6:01
yes. Yeah, it’s it’s not covered. Okay,

Max Carlson 6:05
okay, cool. Is there anything that you’d like to just review now or show now. I mean, cuz I know I owe you, I apologize, I think that I and the security architecture group owed you, some flow diagrams just talking about how web hooks might

Aleksander 6:22
work. Yeah, it would be nice to have it.

Max Carlson 6:25
I mean, maybe we could do that together, you know, that could be a good productive use of the time, we could just hash something out now. Okay, so maybe

Taylor 6:36
we’ll actually, before we dig into that we could I share my screen for one second and just sort of show you what the spec looks like and you can have a chance to say like, oh whoa that’s why I asked for it.

Max Carlson 6:46
Yeah of course. Yeah, perfect. I mean that’s that’s the idea. I’d love to see a screen share. If you’re up for sharing. Yeah, yeah. Cool. All right so, and then who’s joins hg. Who is he,

Ingmar 7:06
emergency. Oh Hey Mark,

Max Carlson 7:10
welcome. Okay, great.

Taylor 7:16
Yeah, sure. So, can you guys see my screen.

Ramkumar 7:19
Yes.

Taylor 7:24
Right. So we have. First, we have some basic terminology sections where we’re trying to define a couple of a couple of sort of critical, critical words and phrases that we’re using. We then dig into actually this at the wrong level here. I’ll pull that up to one again. We then dig into the key digital functionalities. And we have this little comparison with existing standards so, you know, very early on in the process we wanted to sort of map for work that we were doing against the work that’s being done by open HIE, which again is limited in scope, they’re just thinking about interoperability and information mediation and health care, but it was a really useful process that we wanted to document and sort of get into the final product here, how another big big sort of government focused interoperability focused effectively working group or community of practice is thinking about the standards for an information mediation layer. We have the cross cutting requirements, which we still have a couple of open questions around. And there are a couple of places where like around Kumar has asked maybe to get some input from the security group but we think that we want this sort of, we kind of want to submit this document to you guys on architecture with these comments in it, so you can see the comments and see the places where we’re still sort of looking for external guidance.

Max Carlson 9:05
Okay, so, so far. Yep, good, quick question in the cross cutting functionalities. Did you link to the cross cutting functionalities in the security building block yet.

Aleksander 9:17
It was the architecture.

Max Carlson 9:19
Yeah, so maybe, maybe you want to add a note to do that for yourself. Yeah, okay. Do you want the link, just so we have no problem. Oh, you have the security building block okay. Yes. All right. Um, I’m just going to paste the link to your document just so other people can follow, follow along.

Aleksander 9:42
Yeah. Okay. Okay.

Max Carlson 9:48
Okay. Yeah, so I just want to make sure everybody is linking to the security building blocks vestigation The security building block definition. I’ll put that as well. Right. And so, you know, the cross cutting

any specific security requirements. By linking to those. Okay, So, yeah. Great. All right.

Taylor 10:32
Cool. So then, functional requirements, we’ve broken it down into into basically these different layers. So we have a secure communication layer. And the easiest way to think about that is extra code. It’s, it’s the thing that allows your if you go into the basic requirements, folks must be able to register, and we talk about organizations applications and services that folks must be able to communicate via this layer. So we sort of describe that entire process of how communication happens. This layer also needs to provide a directory service. It needs to require lung or it needs to provide logging capabilities and monitoring capabilities. And then this is going to be fleshed out a little bit more, more, but there are some, Generic Requirements for scaling and throughput for services secure communication, those are the basic requirements, and then we break into sort of more detailed flows like how you go about registering a member or registering an application or how you actually send a message from one member of this superior communication layer to another, and we do the same for directory services. Now, on. So that, that’s basically the that is the full requirement for that one layer but information mediator we’re saying is broken down to at least these two different layers, which sort of kick, the third one out again. But the second one is Pub Sub communication. And here we are falling short of the same structure, where we have basic requirements for folks to broadcast messages to provide logs requirements for retries requirements for updating, and deleting and registering subscriptions, so how to subscribe to certain kinds of events in the Pub Sub layer. And we’ve actually provided sort of sample, you know, sample JSON that enforces standards about communication for the Pub Sub layer.

Max Carlson 12:52
That’s great. So you might want to consider actually writing a swagger, or open API that describes that Yeah.

Taylor 13:01
Yeah, that’s a nice idea so maybe like you know and that would be under the details for this section.

Max Carlson 13:09
But this is great,

Taylor 13:11
consider writing this as a swagger spec as

Ingmar 13:20
well, like,

Taylor 13:23
considering that most of the details flow section in layer one is effectively an API is this some time and make the output, more useful.

Max Carlson 13:43
Yeah, that would be great, because then you know, all the requirements for all of these API’s are fully articulated. Yeah, great.

Taylor 13:54
Okay, but this is this is the general pattern for, for, for Pub Sub and again we need to, like, we just have little like mini screenshots of some of the actual diagramming we’re doing in Lucid Chart, but I’m going to bring those either bring those over and make sure that we can use Lucid Chart going forward or convert that Lucid Chart stuff into diagrams that can be maintained using some of the software that you’ve been sending around Max,

Max Carlson 14:20
yeah I don’t, I don’t really mind too much. As long as you know ideally, you know there is this concern about versioning of diagrams. Right yeah

Taylor 14:30
and that’s what’s cool about the stuff that you’ve been using is that you can like, check it into GitHub, you know,

Max Carlson 14:35
exactly, so you could check it into GitHub and then you could have a branch for a specific version of a document that has all the specific versions of those resources and then you can link to that, to those you know what I mean, that’s kind of, yeah, that would be ideal. But, you know, yeah, the version of the documents is kind of inherent in the spec anyway, like it’s got all the versioning built in Google Docs. So, all is not lost, I guess for now what we’re asking is that everybody at least provide a link to an editable version. Yeah, and ASCII, wherever it is like you know if you want to keep using, if that’s possible with this tool and you’re happy with this tool that’s totally fine. I mean we’re not, you know, we don’t want to. We don’t step on your flow, like if you’re most comfortable with that tool provided there’s some way, a way to link to an editable version of the document. That’s great.

Taylor 15:29
editable version of the doc and.

Max Carlson 15:34
And then there’s like an inline image insert some way

Taylor 15:38
of understanding changes version for a time. Exactly, so that’s why we’re,

Max Carlson 15:46
that’s why we’re recommending that people use a GitHub repository. But, but we’re not mandating it because I mean, You know, some people are not as comfortable with that, you know what I mean. But seems like you guys in the long run. I’m sure it’ll save time and headaches. Okay.

Taylor 16:04
And actually we are we are, how do I do this, this, sorry. Oh welcome Ron from our

Unknown Speaker 16:18
arrow Kumar.

Ramkumar 16:20
I,

Taylor 16:22
yeah, we so we are, we do have in GitHub, this spec right so when we have a go stack repo, where we sort of listed out in this specification so it wouldn’t be a stretch to just start doing the web sequence diagrams in here and GitHub. So I think that’s a pretty nice way forward because then we can just focus people’s energy on the GitHub repo and that contains like you know a useful swagger specs. It contains the diagrams properly version, and then we can link to that from the Google Doc, so I think that’s, that’s a pretty good way to do it.

Max Carlson 16:55
I agree. I mean, so that’s one of the things I really like about APA diagrams that never draw that I O is that you can connect it directly to your GitHub, and then make your edits and then save it and it commits back to get GitHub for you. So, okay, great. I don’t want to go too much down a rabbit hole so what anything else you want to show.

Taylor 17:20
Let me get back to the bottom of the spec quickly information building blocks back.

Max Carlson 17:29
Sorry we can bump up the font size a little bit.

Aleksander 17:33
Yeah. A little better. Yeah, definitely.

Taylor 17:39
Okay, cool. Yeah, so I mean, our, as I mentioned before, we need to build out detailed flows on Pub Sub communication. Rom Kumar was thinking about the like mapping this to the use case, and potentially could also lead that bit around could lead the bit around cross cutting requirements, make, making sure that those are in line with the security group. And Alexander is going to is going to add a bit more on the specs around the scaling and throughput, and logging and monitoring, but but when that’s done and then start right and I’m going to do this. I’m going to bring everything over to GitHub, get it into version control, the specs the swagger specs and the web sequence diagrams with those sort of three, not super super long remaining work streams. It’s certainly not perfect but I think it’s ready for ready for for other people to start contributing and getting critical. And the one of the big open questions in our group has been. We just got a whole bunch of CVS sent over to us of folks from a different organization that he connected with. So we suddenly have like a bunch of people who would be willing to contribute more to this, but we also seem to be sort of right at the edge of a working, you know, version one or version zero, or whatever you want to call it. So I don’t know sort of an open question on like, you know, should we. What’s the best way to engage going forward right because like the three of us Ramkumar Alexander Taylor will be proved. We’re not saying it’s, it’s great but we think we’ll sort of have a consistent, and somewhat intelligible way of presenting the spec, and then we need all of this outside input to get it over the finish line and that’s like the architecture group primarily. And then it also could be like new contributors, if we have a stack of five CVs of people who said yeah I’ll help out on the, on the information mediation working group, so I kind of defer to you guys on like what the best way to solicit that kind of feedback is and and how we get from basically a document that’s been created by three people in conjunction with with the weekly check ins with you all how we get that into into the shape where it’s like, everybody’s on the same page and happy about it.

Max Carlson 20:23
Yeah 100% Um, I think, you know, feel free to ask architecture to review it at any point, whenever you feel like it’s ready. I mean it doesn’t have to be you know 100% Obviously. Um, but, uh, you know, in terms of bringing additional people on, I mean, you know, I think, uh, I would I would defer to you, I mean, my feeling would be it would be good to get them on sooner than later, but then again you don’t want too many cooks in the kitchen. Right, so I mean it might be useful to, if given that you’re so close you might want to like kind of get a first draft out, you know, and then start bringing people on, it’s really up to you.

Ramkumar 21:09
You know, like, everything. Based on the first review of the draft that if there are so many gaps and I think we need the extra. And the heads to fill those gaps. Then, let us look out, but you have come very close to submitting version one. Whenever we go post version one, different is proposed version one.

Max Carlson 21:37
Yeah, okay. So it sounds like we got consensus on that. Yeah, good. Reasonable. Okay, great. Any other questions, comments, concerns you guys want to raise from it, or any anyone else wants to raise about information mediator.

Okay, great. Excellent.

Taylor 22:04
If I, if I just jot down the decision here as we’re going to run from our Alexander Taylor will hopefully in the next four or five days, you know, get to this shareable v1 Send it to architecture for formal review and then consider, you know, inviting outside reviewers effectively, and saying to the people in that group of CVS like, Yes, we have this sort of draft information mediator building blocks back if you would like to come in here and review it and add a bunch of comments and make a bunch of suggestions then Heck yeah, that’s fantastic, but we’re sort of not letting that hold us back from sharing is the one with with architecture, is that a decent summary of the decision.

Max Carlson 22:48
I think so. Cool. All right, that’s all. Okay, awesome. So, let’s see, just general stuff. So, okay, cool. So who wants to go next, I guess let’s continue. Just doing the stand up I think this is a this is a good way to go. So Ingmar, do you have anything you want to share.

Ingmar 23:16
Sure. We’re on vacation mode at Disney. Oh, there’s a real progress. Currently, there’s some, some work being done with the with the document. Reviewing meetings. Then we also had a start on on bringing onboarding. More and more people have been waiting, actually, one person on board, or quite many months already so hopefully it will get to. Other than that, We have been experimenting with Alexander with practical part of it just to see where are the obstacles. And I just a few seconds ago I got a success message from the next instance set up for a correct payment the paragraph tag. The United Nations on that system. So, there’s some, some experiments going down.

Max Carlson 24:36
Very exciting. Other than

Ingmar 24:39
that, no, no big change. Okay. Um,

Max Carlson 24:46
do you want to show your documents, and just give an overview, I mean, feel free to or if you don’t want to right now, That’s okay.

Ingmar 24:54
I think currently we’re kind of in the middle of transforming the document so better All right now, the show.

Max Carlson 25:04
Okay, that’s fine. Great, so, um, yeah, I guess in the architecture group. A lot of the conversations I’ve been having are around how we integrate with other existing applications right and I know we’ve talked about this. But, you know, there’s, there are people so I had a meeting with open Ms. Earlier today, um, which is really interesting and just gave them, showed them the same slides that I’ve been showing you, I mean I guess I can just, they’re not even really slides but I can share my screen here, and we can talk through them and see it, you know, get, get feedback if there is any. So, uh, yeah so mostly we’ve been focusing in on this specific, you know system architecture of how God stack connects to existing applications. Right. So let me just post this in the chat here. Oh, sorry I clicked the wrong button. All right, Uh, yeah, just so it’s it’s a. So, um, so, and one thing that’s kind of, so this went over reasonably well. Oh, as far as, you know, the strategy, general strategy of having adapters. Right. So, in this case, open MS is essentially speaking the fire standard and is based on open HIE, so they’ve, they’ve essentially adopted as much of open HIE as they could. So in that case, we talked through this scenario of having an adapter that essentially. And they, one thing that’s cool about open MS is they are working on a tool to automatically generate open API specifications from fire, specifications, which is really cool because if they can pull that off, then that means that we can essentially just connect to an open ms instance of, you know, patient record registry or whatever it is, and then be able to use that as foundational registry. And we can essentially just proxy the open API specification that they give us through to the information mediator and then this adapter that’s adapting building block access to open ms can be super lightweight, like it pretty much all it’s gonna have to do is handle authentication. So we talked about that so that’s pretty exciting. So that’s sort of a, you know, another variant of these adapters which again are for allowing existing building blocks such as a registry or a citizen ID, an off building block to connect to an existing application, right. So, at least with open Ms. This strategy holds up pretty well. And this adapter doesn’t really need to do that much in terms of data translation or protocol translation. It will need to do some authentication, however, um, but as far as like the open API, The fact that it must publish an open API spec to the information mediator including JSON schemas. They’re working on that and something that pretty comprehensively exposes their API’s, their fire based API’s over open API, and JSON and rest so that’s a good news.

Aleksander 28:52
How is this what is open Ms. I tried to find it but it’s mass spectrum.

Max Carlson 29:01
So open MS is open for for health financing. So it’s like, automated claims processing digitize premium freight paperwork right and it’s all basically based, so you know patient verification paper app registration monitoring, reporting of, you know claim submission claim review claim approval. Right so enrollment health services claims management and monitoring. So that’s essentially what they do. So I can post a link to this as well. Um, so yeah, that’s pretty exciting. Potentially, because then we, we have a pretty good story for bridging into existing, you know, open Ms. And again open MS is built around open HIE so potentially that could, you know this adapter could could expand across the full scope of open HIE API’s and flows, but open MS is essentially adapting open HIE standard wherever possible, and that means fire. So, you know, the exciting thing is, with a little bit of collaboration and even if we don’t have collaboration right, we can essentially pick up a copy of their open API spec that they’re going to publish, and then publish it here, right, and then we just need the auth credentials to be able to connect to this API. And then we’re good to go. So that was exciting. So this adapter story holds up. You know, and then the question of course is, if you want, if you’re, say you’re an open ms instance out here, then how does that connect in to say to validate the citizens foundational identity, right. And so the story for that is again we have these API gateways, you know, it could be public, it could be private. But if it’s public, right, which it could be the application just connects the API gateway, it wraps the request the information mediator, the essentially the open API for the information mediator is exposed here. So, same story. This open ms just needs to read the open API here, if it has access, or maybe it has access through a private API gateway that gives it more ability to do more types of searches or something like that against citizen ID and off using the ID building block, then it can connect either way it reads the open API specification from here, and then effectively is routed directly to the citizen ID and off building block gets its response back as JSON, and everything’s happy so they’re pretty happy with that story. So I guess the next question that we really need to flesh out is how to be a part of the GOV stack ecosystem, right, because that is the next thing they asked is like okay well. But what if we want to be an official so you know we went we started out this conversation saying we don’t want to demand anything of anybody in terms of bridging into gov stack, right, if you have an existing application will have these API gateways and adapters, right and then API gateway and adapter could conceivably be combined, right into one kind of component right like this right you could essentially have an API gateway and an adapter that would that would kind of handle in combination, providing the service, but there is sort of an open question about how somebody becomes part of the GM stack ecosystem, and open MS was asking this right so, you know, I mean there’s, there’s a, essentially, you know, you have to have an open API and you have to be packaged as a container with a security with a information mediator. Security server sidecar. When that’s the basic requirements right so we’re just trying to, I want to have a really crisp clean presentation about how somebody could join the substack ecosystem. And part of that is getting listed in the catalog on go step, global, right, there’s like how onboarding, how we onboard project owners in the catalog. This is kind of an open question that we need to kind of review so for instance, open Ms. How can they be on boarded. And then how could they edit, uh, you know the listings themselves because they’re like the experts on what’s going on and open Ms. And they’re, you know they have links to GitHub repos that aren’t there, how do they, how do we keep the listings, up to date and accurate. Right. Um, So, yeah, all of this so, you know, but there are a bunch of open questions that we need to work through in terms of how somebody actually becomes GenStat compliant, right, and what that what that really looks like what’s the testing and certification process. So, anyway, that’s what we’ve been working on, mostly. And then other than that, a bunch of Doc reviews so we’ve done,

security, and then we also implemented. We also did one for payments. Yes, so that’s about it. Anybody have any questions comments concerns about what we’re talking about here in the architecture group. That’s our stand up

Ramkumar 34:21
club in the API and not be a great idea. API gateway required for different adapters. Fire.

Aleksander 34:47
Right,

Ramkumar 34:51
do separate two separate blocks and chain that could have. Yeah. Multiple adapters or could have multiple gateways, whichever way the implementation demands.

Max Carlson 35:02
Yeah, fair enough. I agree, that’s fine. Anybody else have any questions, concerns, comments about what I just showed. Okay, great. So, and then we do have a call with open HIE later today and I’m basically going to talk to them along the same lines, so I know information mediator, I did have a question. So you’re kind of can you point out exactly again. Which part of your document, and maybe send me a link of is modeled. Oh, I see it, it’s a it’s an open HIE comparison. That’s great. So I’ll bring that as well and kind of share that

Taylor 35:41
token with that open HIV today. Uh, let me see. It’s not like a Jennifer shivers or Jamie Thomas, or Paul beyond Mitch.

Max Carlson 35:54
Yeah, let me just see, I just got that invite forwarded to me I’m sort of like, cool,

Ramkumar 36:01
let’s see, so.

Max Carlson 36:04
Alright so, you know, I don’t know, it’s just gonna be a surprise. So honey just sent me a zoom link and is like, yeah, join if you can, so, so, uh, you know, I don’t know I think he did mention Jennifer, but, um, so she yeah she was

Taylor 36:23
the one who, who said so early, early days on the information mediator spec we basically sat down with Jennifer and went through the initial requirements that you guys had put forward, and the open HIE, like version 3.0 requirements, and sort of that’s what you see in this doc now was our sort of notes on how the two mesh together and Jennifer’s Jennifer’s fantastic she’s really switched on, she’s also, just to put on your roadmap, she’s also part of the open source steering committee for open FM. Okay, so, in so far as we’re thinking about how open FM microservices fit in as like a component in the workflow building block. Okay, she’ll be she’ll hopefully have good good ideas there on how that all how it all relates. Okay, awesome.

Max Carlson 37:21
Yeah, that’s gonna be really fun work I think. Super excited about that. Um, so, okay great so maybe you can kind of explain to me. So this table now that I’m looking at it is not super clear. So, maybe you can just explain to me what this, you know, means is this. How are you comparing to. To open HIE, you’re just basically saying that we recommend all this stuff but this is the full, this is the only one that’s required of the open HIE spec require you.

Taylor 37:58
Yeah, let’s dig back into this. What’s parallel let me see if I can find the Google doc where this stuff is actually stored. So there’s a

Max Carlson 38:09
spreadsheet down here.

Aleksander 38:10
Yeah, we’ll go. Go there. Go there. Okay. Right. That’s probably functional specs

Ramkumar 38:24
mapping the functional specs to correspond items in this list. This list was contract so we’re covering all those facts.

Max Carlson 38:35
Okay. Well anyway, I mean, it’d be, it’d be helpful that you know this is not entirely clear to me. So I mean, maybe I’ll just add a comment, you know. So, just to, or maybe you can add a comment if you don’t mind my mouse is acting super weird. Yeah. But essentially just the comment would be, you know, can we please, you know, describe how to just how to read this table. Right. Yeah, cuz there’s a little bit it’s a little bit opaque to me. But anyway, yeah. Oh flesh that out a little bit. Basically, I guess, that table, maybe shouldn’t end up in the final document in the the table above it, the key digital functionalities table. Yeah,

Taylor 39:28
we basically point out there’s a column that says open HIE question mark. Yeah we need, We need more infrastructure. And so this table is the one that references that other table, perfect. It could go in both directions but based on how blank the column is in the other table you’re looking at. Yeah, it apparently wasn’t right. So this is sort of saying here’s KDF one, we think that that is actually addressed by ol iOS two and KDF two is also addressed by or left to right. Okay.

Max Carlson 40:02
Yeah, so I mean, I guess it would be good if it was clear in this document, which things are met by information so I see right okay so the stuff that’s blank here are requirements that are not met.

Taylor 40:15
Yeah, exactly. So, let me go through and add something in here, I’ll say, an introduction explaining, Julian this and the subsequent

Max Carlson 40:26
comparison tables, okay yeah but I mean. Anyway, this will be really helpful for me probably when I talk to that you know I can I can point, point this out as well, when we’re talking to open HIE later today. Okay, great, awesome, uh, anyone else have any questions comments concerns about, you know any of this stuff that we’ve been talking about this adapters API gateways, you know how to be part of the Go stack ecosystem which needs a lot of fleshing out obviously, like

Trevor 41:07
I was just wondering, does the information mediator, do any semantic modeling, or is it purely just a kind of a routine and technical tool.

Max Carlson 41:23
So, can you give an example of semantic modeling just so everybody’s on the same page.

Trevor 41:28
Yeah, so when you see the open IMS, they’ve got, you know the concept of patient and claim and which kind of logic, which are kind of business semantic things, right, so if you’re competing with a standard for someone like open IMS. I we saying that, then in order to process something in that industry sector, it has to comply to that standard, and if so, does information media to do any kind of verification of that or is, is that just something disinformation media doesn’t care about because it’s more of a technical thing that then does routine, those kind of things. Yeah.

Aleksander 42:22
Endpoints endpoints, will take care of that and mediation is just transport.

Ingmar 42:34
I say it’s just the typical transporting of

Trevor 42:38
information so it doesn’t really care about the payload more cares about routing and ends person

Aleksander 42:45
cares only to the level described in an open API, You have types.

Trevor 42:52
Okay, so as transport layer stuff rather than kind of business information.

Ingmar 43:01
So, I was actually thinking, I was actually thinking that maybe we should add to the open API description a link to the link data standard or late link and data concepts, so that the semantics and ontology would be described somewhere as well. But I haven’t had time yet to come up with an example.

Max Carlson 43:25
Yeah, I think that would be really really helpful. Right, I mean it’s kind of essential because you don’t want to just be, I mean, in some cases though, you want links and in some cases you actually want a full copy of that link as it was at that point in time, right, so it depends on if it’s a transactional thing that you’re trying to look at or if it’s, you know. Either way, you know, most times I think you definitely want referential links for sure.

Ramkumar 43:54
No, but as you said, for claims specifically, you require a copy of the snapshot.

Max Carlson 44:00
Exactly. So,

Ramkumar 44:02
yeah, is that, at that point in time. Exactly chain link Let me save the content might change tomorrow

Max Carlson 44:08
depends on if it’s a transactional registry or not, but. So, it, you know, does that answer your question though forever I mean it seems like it’s pretty much on the application to comply to the open API spec.

Trevor 44:23
Yeah, it’s just it’s basically it’s technical mediation and and it’s, it’s that is, it’s clear.

Taylor 44:31
Yeah, and hey, Trevor, I, I dropped three small chats in the, in the chat thread. Basically we kicked out that that from information mediator. And we’re back down to sort of two layers one is the one that, that you guys were just discussing that’s the secure communication layer. And the next is a Pub Sub facility, which actually is a bit opinionated and enables people to do sort of broadcasts and subscribe to various channels. And then, the thing that you’re talking about, you know, open concept lab OCL, have you checked those guys out. No haven’t limited now yeah so that’s really cool, a really cool project. And I know that they were working on an open fn adapter. So the idea is that a you know a workflow engine like open fn or whatever, or even an application that’s just building some sort of, you know, new, new service, they could use something like OCL to do the kind of semantic mapping that you’re talking about right like no we, we say, we say surname and you say last name, like let’s not call them, theories, you know what I mean. It does, it does, I think it does just what you were talking about and that’s a cool Jennifer will know a lot about OCL as well. At max, if you’re meeting with open HIE OCL is a big part of their plans going forward, it’s a really interesting project.

Max Carlson 46:08
That’s great. So yeah, if you wouldn’t mind just you posted the link already in the chat. Great.

Taylor 46:15
Yes, okay, yeah, overground SoundCloud dot o RG,

Max Carlson 46:19
yeah, yeah, yeah, it looks. I mean this looks really interesting so I mean going back to this, you know, adapters the adapters really will probably be implemented using a workflow building block like these will be instances or pre configured instances of a workflow, building block, most likely, I mean, so that’ll be kind of a neat fit right. If there’s an adapter to OCL then you can do that sort of terminology translation. If you want, and you can also do it in the other direction with an API gateway, right. If you want, although that doesn’t have the same level of programmatic control. And I think yeah, you could argue that maybe you don’t want to do that because you want you’re encouraging people to comply with the GOV SEC API’s, right, at least when they’re calling in. Right. So I would argue that, I think.

Trevor 47:17
Yeah, it’s just that sometimes you know that the lines have blurred between, you know, if you look at some applications, they, they might be an API major, but have portions of this in it. And then in some cases, you have, you know, a routing and authentication, like dispatcher, or, you know load balancer or something like that, which also does a similar thing but there’s nothing on the API, so it’s just, I just wondered where it was on that spectrum but I can see it’s, it’s kind of focusing more on on routing security routing and, you know, making sure that you guess secure connections and, you know, manage is the, the transportation layer of of the protocol. Yeah,

Max Carlson 48:09
yeah transformation security and all of that. But, you know, I think the place to think about this would be in these adapters. Right, I mean this would be, This would be where you could adapt a semantic model out here for consumption by a building block. I’m less certain how to adapt a semantic model in an API gateway because it seems like if something’s calling into GUP stack, we want them to be using the Go stack ability API’s otherwise you’re kind of missing out. Right. So, in this sense, I think, when the date when the call when the requests are coming into gov stack I think it makes a little less sense here.

Ingmar 48:49
Yeah, yeah, I understand.

Max Carlson 48:52
Okay, what else, what else should we talk about. Okay, well, so I did get, I guess I’ll stop sharing for now. I did get a missive from. Let’s see. So, Mario wasn’t able to join and but you wanted to give an update so I’m gonna give an update on behalf of him. So they discussed the, the case of cash payments for unconditional social benefits transfer use case, and they agreed to consider the option, but the option is different from the other options and that there’s no payment being routed, but it would be, depending upon the policy in place for managing such cash payments. So in that respect, they’re going to check in with a payment Working Group representative from the World Bank to get their inputs on best practices from the World Bank for such payments and include them for that option. So that’s kind of a, you know, their progress, other updates, they are going to include bank to bank transfer and mobile money payments as well as part of the payments, building block or building blocks, right, that could be multiple building blocks. So yeah, just to, just being VJs proxy here. So I guess I can’t really answer questions on that one. Um,

Ramkumar 50:18
all right, I think it’s really the, the key thing here, I think Max is that last time when we discussed we were trying to push for inclusion of the cash payment remember. Yes, very much so. Now they did consider that as part of their process. Yeah, I think the thing is, they were kind of mixed up in terms of the requirements because of cash pay this mobile cache kind of scenario that was there in the spec. So they were kind of mixed up saying back will entail so my other things in terms of how do you secure it, how do you track it, and so on so forth. So the conclusion has been those are not bad, they’re they’re beyond the scope of the use case. Payments group is only focused in terms of the payment part of the process, it’s not going to bother about how the auto sector of the truck and how to track the. Yeah, the payers and so on so that being excluded, now they’re comfortable, to just say okay, this must be one more just the payment process. Instead of mobile voucher for example, how do you do it on a payment voucher. Right So key point is that they have to read the use that, put that use case into the bucket. In the same vein.

Max Carlson 51:41
Yeah, and I think it’s reasonable to say that the government is not going to be managing a cash trucks, right, it’s more about the, the communication between. So first of all when the government decides to pay somebody then that has to get debited from the government’s account and put into somebody else’s account, and then later, there needs to be a reconciliation, because it could be debiting 50 people’s worth of payments, and then the cash truck needs to come back at the end of day and tell the government. Okay, I actually paid these five people, but these other people haven’t paid yet, so that there can be some accounting happening for that. So that’s great.

Trevor 52:24
Just, I think they also might want to consider it to the cash payment in terms of voucher. Yeah, because you know I think that, you know, if, if you had a government that realistically wanted to distribute cash

Ingmar 52:48
they

Trevor 52:50
doing it via third parties, I mean, ATMs is one. But a better one is actually supermarket chains. I’m not sure whether you’re aware that supermarkets love customers taking cash out of their shops, because it’s really expensive to handle you have to pay the banks to turn up with a cash truck and take all the cash away. So, if for example you were able to dispose you know like in most western supermarkets, When you buy your goods with your debit card. You can also ask the cash, as well. You can say, Can you give me like, 30 euros, and I buy my shopping, and then you pay for it on the debit card, and then they give you the cash back. It would be interesting to see whether they were able to do some kind of sanitation of that bearing in mind Horn of Africa. It may be a really successful way of distributing cash because, you know, supermarkets, and French, or French or supermarkets have extremely good penetration into into remote areas, because people have got buy stuff somewhere, right, and you might not have, you know, major supermarket in every village, but you do have them in every town. So, I still feel that there’s a strong use case for not only using, you know, government bureaus and cash trucks which is actually a pretty inefficient doing thing. Voucher remediation for cashiers and SMB like a supermarket is I think a really practical way of dealing with that.

Ramkumar 54:39
Right. Okay, I see therefore there are some, there’s one more aspect we need to take into account max that we started with a bunch of use cases you had in dial, and then started mapping in Vietnam into the ultimate use cases that are coming up from within the groups. Not sure that we are mapping them back into your main list all of them. So it might be good to add these those documents to flag those new use cases. So you get one consolidated list finally at the architecture.

Trevor 55:23
I think it’s a good idea. What I would suggest is prior to going to all of the effort of updating the use case is just to, just check with them, if, if this was something that they had considered in terms of vouchers, because it may already be in the back of their mind, or it may be something that they say listen, it’s not practical because of x y and Zed reasons, because actually you have to get most of the vendors to buy into that. And, you know, you also need some way then reconciling between, you know, in this case the supermarket, and, and the government sector is not as straightforward. It was a more question to say so. Had they considered it, or he was oh, it was something, when, when they’re talking about cash payments that he was just understanding whether that was in the back of their mind.

Ramkumar 56:21
Okay, I can pause it, but it would be a great idea, as you can just put in your whatever you said now and a few blank lines out there so I articulated correctly. Otherwise, I’ll be getting what I understood from what you said. Yeah,

Max Carlson 56:44
so, in, just a reminder, architecture group CC architecture at discourse, cosec global. And then it all show up in our discourse.

Ramkumar 56:56
Okay. So, same problem, I’m just saying that how are you crack. How are you, mapping out all the use cases that are coming in within the building blocks. How are we getting them.

Max Carlson 57:08
So I just posted a link to the logical process blueprint. And I think, you know, All use cases should be at least linked to from there.

Ramkumar 57:21
Right. Okay, so, maybe,

Max Carlson 57:24
even if it’s even

Ramkumar 57:29
if it’s in payments, three use cases coming up from it building block. Yeah, out there in your list, and how much of them map to the old ones common or new

Max Carlson 57:42
well so I mean, my suggestion would be that we treat them as new use cases, because they’re going to be variants in any case, and we just add make sure that at least there are links in this document because then all the groups can see them.

Ramkumar 57:56
And they’re not mapped the same way as you had mapped with case template.

Max Carlson 58:01
Yeah, well, you know, that’s fine I guess as long as we end up with a web with a web sequence diagram that shows the flow between building blocks. You know, I think that’s okay. So it, you know, these, so I think, let me just share my screen again right so so that everybody’s, everybody can get a visual and see what I’m talking about. So, what I’m hoping for is that at least, if there are interactions between building blocks that there is a link in this document the logical process blueprint that I just posted in the chat that there’s a link in here to that use case, you know, whether or not it follows this specific format or not, like,

Ramkumar 58:42
I, you know, basically means that they have to deposit their use case links in your document. I’m just getting a bit.

Max Carlson 58:50
Yeah, exactly, just so that at least, people can, you know, can see it, right. So if it’s

Ramkumar 58:59
at this point I don’t think anybody knows this.

Max Carlson 59:02
Okay, well I’m just, I’m just saying it now so everyone knows it, who at least is on this call right then, if you have flows that go between building blocks, like your use cases or flows that go across building blocks, please put links to them here, so that the other groups can review them. Right, if it’s an internal flow internal to your building block, it only, then you know you can just leave that, you know, in your building block definition that’s fine. Does that make sense.

Ramkumar 59:39
Yeah, They should know. Maybe mail will help this link with this link.

Max Carlson 59:46
Yeah, so I’ll just make a note, you know, one of the key decisions, is that we need a

Ramkumar 59:56
backward migration of use cases. Groups that cross building block,

Max Carlson 1:00:06
or that use multiple building blocks should be linked from this Google Doc here.

Ramkumar 1:00:18
Yep. So I just wanted to bring up one point, I think a couple of times, I kind of come up I was trying to discuss that. The registration group with Inmar Trevor is here and Ema is here. Yeah, yes, yes. So I was just trying to bring up this one point where our discussion is the concept of document oriented storage. Right. And so I am I have discussed this and we are just bringing up this point because traveler, had this model, referring member who was going back to this model where there was a document template, and then you were feeding into the document template as part of some workflow, different portions and so on so forth. So in that model what I was thinking, is that the various fields inside the document are not necessarily same record, the same, the same formula is the same register. So correct me if I’m wrong, because I am thinking that a document might get populated from parameters that are stored in different registers that may not always have everything in the document come from the same one single table, for example. So is that a valid assumption, or is that not a valid assumption

Trevor 1:02:08
you phrase actually a very good point, and is an often a misconception. Okay, so firstly let’s take the digital documents. If I create a legal digital document like like a prescription. The prescription will have say my name is the patient. Let’s keep it like that and we’ll have the doctor’s name is the person that issued it. Now that document is a snapshot. So it’s not a link so genuinely speaking what happens is if you’ve got say a read say like a registry of patients in one system and a registry of doctors in another. In the digital version of the document. It should not store like the doctor ID, it should actually take the whole record. so it will take all of the doctor’s information and place it on the document, like the address and the practice number and the phone number, and it must be stored as a snapshot on the digital document. And it does that so that if for example you had a physical document, and as you know like you scanned it, it should match with if you just store, like a foreign key. And you say this is the doctor’s registration number, and you can always refer back to the registry Well, if the doctor’s phone number or address it changed in the digital document will differ from the scanned document so that the reason that it’s a valid point that you may because not, it’s not obvious, right. But when you have a transaction document, you must copy all of the master data or the registry data onto the business document because it reflects what was done at a given point in time. And what made you know and that changes over time. So, in terms of the digital document management, that’s why you copy registry information. Does that doesn’t

Ramkumar 1:04:12
answer your questions. So essentially, in doing so, what we’re saying is supposing I’m actually writing a prescription, a portion of that prescription is populated by some data lying in some register, Maybe the patient registry, patient, and a portion of that is lying from, maybe I am as part of the prescription, writing out medicines that have to be taken, and maybe that portion is coming from some medicine database. And then I am fixing my signature out here, and that is coming from some other database, my digital signature. Then there is a, what would you say a footnote, which says, Do this, don’t do this, blah blah blah, disclaimer is coming from another database. Together, all these fields, it is called prescription document. Right. So what you’re saying is, when I compose this prescription document, bringing these relevant portions from wherever these other registers are and compose this document. And when you store this document is the entire document goes in, in one single one single storage. There are two different ways of looking at it. One is, I’m assuming, this would be going into some no SQL kind of database where you don’t really care whether it is a table or a document, you know, as long as the name and value field. So I’m assuming that is, if that is not the assumption if it’s supposed to go into some SQL table or something, that it’s more complex than what I’m saying right now. So, it looks like it should therefore be stored as one object, right, this this prescription will be stored as one object with embedded contents that have been imported from other objects, other registers or other tables of whatever files. And then this object stored as an independent one line entry. In, a new league newly created document instance. Right, but

Ingmar 1:06:34
why is this important. We thought we could find this one. It could be, it could be, it could be structured database, it could be json file, it could be anything.

Ramkumar 1:06:53
Well, Let it be so. Now let us take the same case with the structured database, I’m having all these things in some structured database, and maybe database that structured database doesn’t have a file embedded it might have a blob, or it might have a pointer to a file. Either ways, whichever way it is stored. All I’m saying is that any other document that has created the prescription. In the next instance when I want to access the prescription, two equals three, registration building block. The client application which wants the prescription must be able to come and say, you know, in my, give me this prescription, and you are able to get all the data related to that prescription, right, which means that internally you have a mapping of how well and where all that data is stored. Is that a yes.

Ingmar 1:07:46
Yes, because, well, the prescription in general is put together and from different data that is true. And then when it’s, when it’s registered in the register of prescriptions, then that’s a different register it contains information about the prescriptions, and it should have some kind of a structure as well. For example, you need to search by the prescription number you need to search by the owner of the prescription, and so on and so on so but it is already a new register.

Ramkumar 1:08:24
So where are we, therefore, are we are we thinking as to how we can create those document templates which define prescription is equal to these things, build is equal to these things, or is it something that the application should decide, is there some of the use is there a, is there a service that we use on the registration building block of defined documents said like this

Trevor 1:08:50
is the generally there is no cross, cross industry standard like a reference architecture that defines this, and this is actually one of the problems that you have with industry based reference architectures, is that they don’t, they don’t think of things in these terms, it’s essentially something that, that’s quite unique to the SAP stable but there is a very valid point, if, if you have any of the building blocks, there is acting as a system of record, ie, this will become a legally binding record of what was done, then you would need to make sure that all of the relevant master data was copied on to the original document, and there are certainly been legal cases and before I joined this API, she had a customer that had this problem. When, when I build them a system. I kind of, it was a prebuilt application system it was actually a sage product. And what we did is we kind of generated these quotations. And we stored them as documents, and we didn’t actually store the address information at the time of generation in digitally. Now, what happened was that their document management system had a, an issue, and they were trying to regenerate them. And we because we didn’t store address versions and store the address version on the document. We couldn’t actually regenerate some of these legal documents. So, it is, it is certainly a design consideration, if you look at accounting systems and you look at any and I’ve worked with a couple, and you think, sure, a lot of this data is redundant. It’s got the, you know the address on the customer record it’s got the address on the doctor’s record he’s got an address on the prescription record or the quotation record or whatever. Why is it redundant. And the reason for this is that when you generate a legal document, you must have a snapshot now. A lot of companies do that in terms of putting it into a digit, like a PDF and they put it into a document management system. What it then does is it turns it from a structured data source like a database into an unstructured data source, which can be incredibly difficult to then

Ingmar 1:11:21
sort of

Trevor 1:11:22
mine at a later stage. So, these are the trade offs, but I think that the point that you’re making is whichever form of digital document it takes, whether it’s a table level, or whether it’s like a blob or a JSON file or an XML or whatever. It must be a faithful representation or a snapshot of what happened at that given point in time, because the registry information changes.

Ramkumar 1:11:47
Yes, I’m saying that therefore, there should be a way to templatized a document.

Aleksander 1:11:52
Yeah, I mean,

Ramkumar 1:11:54
obviously if you’re storing it. Now the question is, should that templatized ation become a capability offered by the registration building block because somebody who’s building an application can then come to the registration building block and declare, here is my description. This is how it looks like. Here is my bill, this is how it looks like. And thereafter, the registration building block knows what needs to be stored or retrieved from wherever as a snapshot in order to assemble a prescription, or an assemble a bill. So the capability of defining this should be, should be an integral part of the restriction building block, or should the capability be some end application because then it becomes a problem. Every application is expected to have the capability.

Ingmar 1:12:49
Well, I definitely see that the right place to, to, to have the template builder is the registration building block, because it’s, it’s collecting all the data from different sources, putting it on the form, visualizing it for the user. And at the same time if a user wants to. If user wants to generate the document out of a PDF, or organization or whatnot, then it’s just a button click or is maybe it’s even background process that will just generate the document upload it to the right place and so on, so on. So I see it as a resume. They needed functionality in the registration

Ramkumar 1:13:35
in some legacy situations where they’re either using some OCR to translate a printed record scanned or printed record. all those things fall in the same relevant and same for when there is a template that you can populate that. So who is going to define the template, and we can say here it is. This capability is there, you define the template according to whatever you know whatever documents you need with the capable cables.

Ingmar 1:14:07
So here’s, here’s what I wrote, you will see it as a system that enables to convert these paper templates into electronic templates. I don’t see much need to have OCR, reading, because that’s the whole world of its own. But just to generate a document from template and data, that’s definitely the registration building block domain. Yeah,

Trevor 1:14:35
I don’t know if any of you remember in the distant past Microsoft walks this thing called Rosetta net. Yeah it was. It was around the year 2000 So, the kind of concept behind this it was all kind of XML based was that you had a way of interchanging information, and their view on it was whether it’s a, you owe a system, important because you can have things like an application doing an acknowledgement of a receipt of a prescription, you know, to the medical supply, so the forms are the digital documents on only things like Word documents PDF documents and those kind of things. They’re also things like EDI II documents or system to system documents, and it’s, it’s a kind of a meta structure that kind of sits below it. Interesting enough, Rosetta and it was a complete disaster because they couldn’t standardize it, and again it comes back to these industry standards and reference architectures that they, they never aligned with each other, I can tell you that open IMS will have nothing in common with something like bi n which is the banking industry reference architecture, and they, you know, were one talks about, you know, a, a patient a doctor and an insurance company. The buying models, we’ll talk about, you know, customers and merchants and, you know, beneficiaries any con cop maps and then that’s that’s what makes this kind of thing really difficult. However, the concept that you do have industry standards for templates is true and I think you’ll see that in the open IMS, they will say that these are typically the semantic structures of a certain type of document, as I’ve said before a prescription for the medical industry in every other industry is called a purchase order, right, because that’s what it is, when you get a prescription, it’s a purchase order, and the doctor gets paid or the, the, The medicine gets in, it gets paid based on it, depending on which way the is private or public healthcare. So I think that that’s what what makes these kind of document templates, difficult. Generally, the way that sa PC does it is you have a header, a header is that if you look at any kind of quotation, or even a prescription, which is a purchase order, it has a header, which has like the patient, the doctor, the surgery date, you know, maybe some kind of status represented by a stamp, and then it has items which tend to be the product information. But that’s not always applicable in some cases a document, literally is just a header. So for example, you know like a letter or notice or a legal document is just basically a header with text, so it gets very very difficult to sort of template eyes these documents. And all I can say is that within the industry, new have had it, you know, PDI in the manufacturing industry has been around for 30 odd years, or 40 years now, but in other industries. They have not managed to create these kind of electronic document standards. And I think that this is where it is problematic because if you want to interchange the document, so if you want to take the registration document, and pass it to another entity, then you have to do some kind of transformation mapping and extraction, and if you convert the document into, you know, certain types of formats like blobs of tape pegs and things like that, then you break it. So it was a very simple question but it’s actually, it’s not simple as this like these arrows, they weren’t actually primary key foreign keys, it was actually the movement of the data onto the form

Ingmar 1:18:41
SPRL standard. For example, and good good examples to have a standardized XBRL, but the the ontologies, or the, the gap systems in each country, this, this is, this is different, a little bit different in each country but the standard is the same.

Trevor 1:19:04
Yeah, so this is the kind of the difference between the kind of the protocol and the semantic the semantic tends to be you know like what one person calls a citizen, the other one calls it a patient, and the third one calls it a customer and the fourth one calls their beneficiary, but I’m still me. Right. So that gets very very difficult to to align particularly in some cases where, you know, a customer can be a person or an organization, but in another definition citizen, I can be a citizen but a company can’t. So when you’re mapping, you know, customer information which is a really good it’s one that is very common, actually, you know one system is expecting the customer to be patient, and another one’s expecting the customer to be like either a company or a person, so you map the customer information to the patient information, and then you realize, it actually doesn’t always work that way. And it gets even more complicated than that. So the, the underlying protocols are I think are out there, but the logical semantics are not, they tend to be industry specific. And that’s why, usually when you’re looking at something like AppStack they tend to be kind of citizen, or to retain registered business based, and that’s a good kind of government sector view of the world, but then when you try and integrate government systems into health systems, it becomes less clear. You know, it never matches up nicely, but at least there’s a generic structure here that should cover most cases, and then the other thing that you often get is just general containers and containers is something you can chuck anything into, which is what that kind of folder concept is, and by the way the container where you see

Ramkumar 1:20:47
like the looks like

Trevor 1:20:49
a paper file like the patient case. Most process engines or BPM engines or workflow engines have case. And in that case you put anything you want. So they don’t care what you put into it, and they often the challenge that they have is the structure that gets put into the case they can’t always get site off, particularly if they are unstructured data, so you if you had like a workflow engine with a patient case. The case usually comes with the workflow engine, and then you put inside a scanned document of a JPEG, and there’s a business rule that says if patient over 18 Well then your workflow engine just falls over and stops. So, you know, you always have to say whether you have it as structured digital data or unstructured data. You know, you can’t really define that upfront, you have to do on a case by case basis.

Ramkumar 1:21:45
But then, to define a document template. You could well define that this document has unstructured data in some place right. Yeah,

Trevor 1:21:56
absolutely. I mean, generally say again SMP for example, you have attachments on the header and you have attachments that you can put on each line item. And you also have attachments you can put on the wall structures, the attachments you can put against the customer and attachments that you can put against the product, so you can put the attention to everything. And then the customers say yeah but now there’s too many places to put the attachment, I don’t know where to put it, it kind of becomes a little bit unclear, but that’s why when I’m doing process model, I always refer back to pieces of paper, because it helps to people conceptualize what the template should look like when you digitize something a lot of people lose the structure that is required in order to run a business process, but when you show pieces of paper if you look at a form, it usually has a bit at the top that way you know you put all the main accurate details, and then you have, you know, should we say process or product related data and then it has you know for official use only, which is where you put, you know, rubber stamps and signatures and internal reference numbers, And that’s actually how you should be representing your, your, your digital document in a structured format. So that’s where you know, like open formats like JSON or better than, shall we say more structured formats like XML because there are constraints within XML so the, the JSON protocol allows you to have these kind of nested structures. The problem comes is when you want to kind of interrogate them for running a business rules sometimes it gets a bit ambiguous or a little bit difficult to identify.

Max Carlson 1:23:37
So, yeah, so I mean, since, you know, all of this data in our system, I’m hoping is going to be, you know, eventually represented in JSON right so that’s kind of the goal. And so one thing that I think we should talk about and I think Ingmar you talked about this a little bit. A little while ago is the notion of having references inside that JSON so you can have a reference from one document, essentially to another, right. So the use case of building a patient case, right where you want to store a snapshot of, you know, all of their details at that point in time, could be in theory just as simple matter or quote unquote simple of keeping the existing schema that defines what that case looks like with the references to, you know, the child and the mother and all of that, and you want to serialize their name and everything as it was at that point in time right we can agree on that, you can just create a copy inside the JSON using some sort of opinionated format, and then still keep the reference Id like ICD 10 So you’d be able to look at what it was at that point in time that way but you could still follow the trail and go look up the code later so I mean, Yeah.

Ramkumar 1:24:52
From a storage perspective from a storage optimization perspective, Max. There is another way you can do it and I and we can discuss that the point is seen, if we had a policy that any document that is modified does not overwrite does not overwrite the previous document it just becomes a new version of the document, then what happens if somebody changed my address. Well whatever document has my address. Yes, there is a new version of the document, but wherever I gave reference linked to whichever other documents that are using my address, that link will always point to my character version, the version of the snapshot.

Max Carlson 1:25:39
Yeah, that they had, so Exactly so. What I’m

Ramkumar 1:25:42
suggesting is that as in all these documents so much as replicated storage which you provided.

Max Carlson 1:25:48
Yeah, it’s necessary. It’s necessary if you want to store a snapshot in time. So what I’m suggesting is that we adopt a convention where documents stored in the system of record registries must copy all the data into the document to provide a record as a snapshot of time, and should include references to all foreign keys to objects copies to objects copy, yeah. And if you do that, you can pretty much

Ramkumar 1:26:16
understand the last piece, what does that reference two asset copies,

Max Carlson 1:26:19
so this would be like the doctor.

Ramkumar 1:26:23
The first one you’re gonna need the second, you are anywhere copying all the

Max Carlson 1:26:27
data you are you still want to be able to refer back to that doctor, right. So, you know, it’s just, yeah, so you need to you need some sort of thing that says the doctor ID is blah, which is kind of how this template started out anyway. I mean, that can be represented pretty well in JSON schema and open API 3.0 or 3.1 rather, so

Aleksander 1:26:50
you might want to evaluate what will happen in future so that’s that’s references.

Ramkumar 1:27:04
Okay so that is the new so if you go back there in my doctor’s latest.

Ingmar 1:27:10
Well it’s, I think, I think it’s pretty, pretty important to actually track down. For example, which doctor, offering which vendor prescriptions, so kind of a reference number in Estonia, we’re using this unique identifier of a person that is always linked to all that, all the subjects, whatever document. So in other cases when this unique number is not available, then some other kind of for France

Ramkumar 1:27:47
to do no I’m only. I’m only suggesting that if you are giving references. Then what are we copying all the data, and if you’re copying all the data, why are we again giving references, that is not clear to.

Ingmar 1:28:00
So we’re talking about. It’s, it’s, it’s, we have been thinking like this for for quite many years and trying to achieve this, For example, very good example is a population registry, so every every registry has information about people the population and and everybody’s copying it so at some stage, we had this idea that, okay, let’s not copy the date, because the person’s name may change. Let’s just keep the reference number. And that should be enough. But for scalability, and for availability. The it’s reasonable to still copy the information, just because you may need to search your own registry for for first name and last name, instead of doing the search to the population registry every time that you need to search. So you want to alternate abilities if needed.

Ramkumar 1:29:13
Yeah. My question is so if you’re, if you’re deciding to copy it to the local version. They still need they need the link. Yes, so wherever that document is, And if so, why do you need the link.

Trevor 1:29:27
It’s just for traceability. You know, if for example I mean ID numbers is a good one, since I’ve been in South Africa, I had three guys, three different ID numbers with the Department of Home Affairs, because as I’ve changed my residential status the ID number has changed. So, you know, you always say, Well, when I registered for my seeker with this bank. What was my ID document at that point in time. Now the Hydra document might be my primary key in, in the citizen registry, but you always want to copy that on to the actual document. At the time of the event so that as my ID number changes, because the third last digit, by the way on a South African ID is your citizen status. So, as I went from a permanent resident to a citizen, they change that number. So, the. That’s why you can’t always rely that and it’s a good idea that you you register it and store it against a document and ultimate is only like, 30, images, so it’s not a big space over, you know, says,

Ramkumar 1:30:45
Here is a snapshot, and by the way, here is where I got this snapshot from.

Trevor 1:30:49
Exactly, so that you can audit it just as another point, I will just also mention this within SAP, what we are finding, particularly in government sector actually, is that when you have a process that works across entities, government entities, there is a trend moving towards blockchains for these. So that’s why I don’t want to be too prescriptive about templates and and protocols. And the reason for this is that they are finding more and more in government sector that they want a single globally shared. What’s the word immutable immutable document so that they can use it as a reference. And the nice thing about blockchains is they’re highly immutable, you can’t change them. The reason that I mentioned this is there are, I don’t know the UK Government at the moment is building a high speed rail link between the north and the south of England. Sad SAP currently is helping them put the whole kind of supply chain network together so when blockchain was launched about 10 years ago, you know, it was very much Bitcoin and Kryptos and all these kind of things, in reality that’s not where it’s being used for the, the, kind of, two use cases that are the highest uses of blockchain are firstly, the medical industry or the life sciences industry. They actually are using blockchains to to track and trace the COVID vaccines and prevent fraud and false copies, because we’re on the supply chain that uses blockchain, and the other sector that’s using them is actually public sector, because they are finding that when they do, you know like the publishing of a contract to build a railway or a power station or a dam. You’ve got various government agencies who want to see it, you’ve got the tech. You’ve got oversight committees, you’ve got local councils, you’ve got departments of infrastructural trade or transport was taking it. And, and I think that, although I haven’t kind of pushed it as a point when you’re talking about systems and records and documents that this is something that is becoming very popular. Particularly when projects are function, funded by, you know, the World Bank and, you know, investment banks and development banks, those kinds of things it’s quite a hot topic, and it’s not to say the blockchain Wando is more what they call distributed ledger, which is part of the blockchain to Dotto standards where you can put basic business logic in the document. And it basically becomes immutable, so all of the information must be copied onto the document, and the, the, the, the, becomes the block, and then the chain is a hash key from the preceding document. So, if you use the preceding document it changes the hash breaks the block, and then your immutability is questioned, So that’s that. I just thought I’d, I don’t want to kind of, you know, disrupt the whole meeting but certainly I’m seeing that with the work that I’m doing with development banks and central banks and is more of a workflow thing that they’re interested in not crypto. So, we need to bear that in mind,

Ingmar 1:34:34
we’re actually using in Estonia, the blockchain already for many years, order to guarantee the integrity in business registry in land registry. See that records has have not been changed by accident or by by somebody tampering the data. So that’s useful Lundie verifiable credentials is coming in so hard so always as well.

Trevor 1:35:05
And, again, is that in the public sector, that you’re getting more of those kind of blockchain use cases.

Max Carlson 1:35:15
Well, Estonia is actually kind of a pioneer in that. Right. They’ve been deploying blockchain for the for registries for a while just that, you know for because it’s much, much harder to forge a document or change it later if it’s part of a chain like that.

Trevor 1:35:35
Yeah, government sector is the, you know, government agencies that are using that blockchain to, to get the immutability, is that correct,

Ingmar 1:35:45
yes. I’m working in Estonian government. The full time. The business registry and land registry are the biggest registries that I’m responsible of the development and maintenance. And we started using Blockchain when the blockchain wasn’t hyped yet. So we started implementing in. It wasn’t open Bitcoin blockchain but closed private blockchains got a fine company offered this based on in government, and has been working since then. And we have been tracking our data I’m still waiting. Some whistles going up when the bots are finding some data that integrity is broken. So, either by by by some administrator making mistakes by chasing data in the database incorrectly or, or, or some hackers. So far, so good. Only only some technical issues, things but other than that it’s working pretty fine.

Ramkumar 1:37:03
Yeah, I mean, we make a recommendation that the backend should be blockchain, preferably and I know

Trevor 1:37:18
I wouldn’t be forcing it. Because it’s not, it’s not accepted by a lot of things in this, there’s still a lot of perception issues. I think it’s more the other way round, that we must make sure we don’t execute it. It’s a really powerful

Ramkumar 1:37:40
capability of the registry. Yeah, I

Trevor 1:37:44
mean, for example, I don’t know if you know K if w, then closely linked to the M Zed and di Zed case who are the German Development Bank, and they have created this excellent open source project called through this presentation. Yeah, yeah, so they created this thing called true budget, and true budget is an open source blockchain solutions that they use in Burkina Faso, I think they use it in Brazil I think they had a use case and I think in the Horn of Africa they got one customer that’s also, and it’s their government divisions, and they don’t use it for transferring the funds they use it for is for storing where the funds have been spent so when they do a drawdown from, from a project that’s funded by K FW. They put a request with the invoice. We’d like some metadata around who we need to supply was he the directors of the company at that point in time. He has put into the blockchain along with the amount of the drawdown and is used like a workflow actually, that’s, that is the primary use case now that’s being used in blockchain is actually a workflow. Yeah,

Ingmar 1:39:02
sorry. Actually, the the the throat system itself, it kind of changed on the logs as well so everything going through the throat is kind of change with the hash values so later on, whoever is using the throat inflammation can you use this information in court by saying, This is what I received at that time. And it’s immutable, so that nobody can argue on that. So, yeah, yeah.

Trevor 1:39:36
The other thing just bear in mind is not one of the problems that might have with the blockchain. And it’s like, for example, Bitcoin is a, is a blockchain wonder old technology with USM is a blockchain too. So what’s the difference here. What they found with Blockchain ones is that they used to replicate the whole database securely. So even if you had the data you couldn’t see it unless you could be cryptic. But what we’re finding is those databases are getting quite big, and each of the nodes requires a lot of processing power to keep the blockchains up to date which is why Bitcoin is using like a nuclear power station to keep this kind of infrastructure going because you need to replicate all of the data in the blockchain to Dotto, this is these are known as a shared ledger technology so they don’t distribute all of the data to all of the nodes. What they do is they selectively distribute data based on who the subscribers are so therefore the blockchain today databases are smaller than the blockchain one but databases and this is another objection that they had in the early part of blockchain is that they tended to consume a significant amount of information storage, and were therefore difficult to search and to, to mine but that’s changed a little bit to.io and so on the big data technology. So, again, is where to be cognizant of that. Because if one of the problems that you do have with blockchains is that they don’t. Certainly blockchain one doesn’t scale very well, but it does certainly have use cases in cross company workflow applications, and that seems to be the primary use case that we SAP are finding now.

Max Carlson 1:41:15
Yeah, well, I think, I think it is worth pointing out that Estonia’s whatever their solution is I mean it’s been working for those applications for 10 years now, there’s something like blockchain. Yeah, but it’s like seriously like blockchain before it was cool. Yeah, yeah. We didn’t call it blockchain. Yeah, no, but it’s kind of kind of cool. Okay. All right, so, um, let’s see. Alright so, how’s everybody’s progress on the PowerPoints. Oh one more thing I wanted to cover right, is this, uh, I wanted to review this document with people, um, this. Sorry, this workgroup cookbook document. Um, just to see if it if it kind of, or I wanted to at least point people at it, because we’re looking at onboarding new groups and I know some of you are considering onboarding new members. So, this document is designed to ease that process. And it kind of talks about the stages in the project, you know, we’re in state we’re in step one, right. But then here’s what happens after that, you get a reference building block from the specification, then building block reference implementations are assembled in sandbox environment and demonstrated, then we actually try to modify configure and deploy a solution. Once a country is selected one or more reference solutions. So that’s kind of a rough idea of the, of the, you know, how the project is flowing, but we’re just, we’re basically. Specifically we’re focused on step one, right. Okay, so, uh, let’s see, and then getting started, right. Make your own copy of this building block definition fork a copy of the GitHub repository for diagrams. Review the arc architecture blueprint non functional requirements for requirements, and then review the, and I guess we should add a security building block specification, right. So, any comments on this so far. Um, you know, would this be helpful, do you think, to, to give to your, you know, new members say when you’re onboarding,

Ramkumar 1:43:53
and also some, some orientation. Although discourse has not been really used by all the groups, but if you’re going forward with the discourse and they should also know what GitHub and discourse. Yeah. Okay.

Max Carlson 1:44:10
Um, I mean I feel like we’re in, we’re, we’re making progress with discord, so I’ll just add a note to do that. Maybe it’s a bit too premature, right.

Trevor 1:44:32
To do this, but what, what we often do in as a part of a change management process is not only write these as a document but to make them as a, as a video. Yeah. So, they often do, like, they take a group of people that are being on boarded. Right. They provide a detailed document like what we’ve done. They then often provide a one pager, Like a cheat sheet of all the key things. So once people understand what they are, which is what you’re doing here, you’re kind of doing a detailed, very explicit statement of what all the bits are, and then you complement that with a cheat sheet with all the key, you know, resources and where to find them. And then when they first present that to the team of onboarding they then record that session, and it becomes one of the resources that they can use moving forward and you know you can post on YouTube or wherever you want really, really matter, or within one of the repositories, but what we tend to find is having the combination of audio and visual. It helps people to, to absorb it, some people prefer to read and other people prefer to listen. Yeah so, as I say it’s a bit, maybe a bit premature because we still kind of finalizing. You know the details but you maybe want to consider that as a next step so that it’s more consumable for people. Yeah,

Max Carlson 1:46:04
I’m just gonna put a comment here.

Trevor 1:46:14
He does. He does recordings and,

Max Carlson 1:46:18
sort of, I mean it’s kind of janky, but it can. I don’t know if ours is set up to do the recording,

Trevor 1:46:26
or we can do on teams as well as the other way, Because that that does

Ramkumar 1:46:30
mean online right so if you, if you had a session like this, we’re explaining this and it’s recorded. Yeah,

Max Carlson 1:46:38
yeah, sure. Point taken. So, okay, and then there’s this process that we’ve been doing right. And so I wanted to talk about this right we’ve been looking at the current sprint, writing a use case for the current sprint, you know, writing alternate use cases and then basing on basing that I still got this old process of, you know, using the logical process checklist, but we also and then we have this, you know, which I think maybe has fallen out of favor I don’t know. And then there’s creating this cross block diagram, right, that showing the interactions between building blocks and then this has like your link to that. So you can see it right, right there. Um, okay. And then there’s just a suggestion for working inside your group and then how to share it. You know, so please if you have any like other comments or feedback on this, please a rebate. Um, you know, if I’m missing something. Um, cuz yeah, we need to turn this into a video and presentation pretty soon for new members and, you know, And for new groups. Okay. Uh, sorry I didn’t mean to hijack the conversation there. So I guess we can review progress on PowerPoints. Anybody have any progress on PowerPoints. I’m

Taylor 1:48:16
not on our end, I haven’t switched our we have our PowerPoint, we haven’t switched over to use new

Ramkumar 1:48:25
format right to all the groups, and also the team, Team format that came from Sarah have circulated to all of you, I think we are looking forward to closing this as part of the review. This month’s review meeting. So that way, good different groups can take this up, whatever is remaining and finish it off it is to first cut worship. Okay that’s sent off to.

I think this is a good thing to be dragged by the 22nd or 24 I know when is the next review meeting this month, all hands meeting.

Max Carlson 1:49:12
You know, I forget is the 20th or 21st I forget exactly.

Ramkumar 1:49:17
So we have to go. Maybe it’s good for groups to converge on that.

Max Carlson 1:49:24
Yeah, I mean it doesn’t have to say much. I mean, could just say like here’s the team here’s our progress. I mean, that that goes.

Ramkumar 1:49:33
Yeah, we can there need some something to populate it. Yeah, for me to flex the current status. Okay,

Max Carlson 1:49:46
format, and template. Okay, great. All right, so I you know, to be fair, the architecture team needs to do that too and I haven’t done it, we haven’t done it so don’t feel bad if you haven’t done it yet. Okay and then everybody’s using the building block definition template right down two significant changes and decisions and a log under a header, and your building block definition template so, you know, for example, information mediator, at the very bottom has this decision log with the dates and the description of what happened. Right. And then, who was involved in the decision, right. So this way, somebody who’s coming on later can understand how they arrived there, so that’s really important. So, a reminder to provide source for all image diagrams either dry Oh web sequence diagrams, right so they can be edited and evolved, and if better them into GitHub. Okay. You can do a link right there should check. Well, yeah, yeah. So, uh, yeah, so I’ll just put this, I’ll just put this in as kind of like a key decision. Each groups in here, groups will provide sorts. Hey, unless anybody has any objection to this. Yeah. Okay, I think that’s pretty clear. Uh, so and I guess this one I’ll just provide a link. We just had a comment. Okay, great. So, uh, yeah. All right, and I’m going to put this back sorry. Okay, so, uh, alright, so we wanted to review last week’s key decisions and see if they’ve been addressed quickly. So security highlighted this need, so I did talk to I talked to honey about this, the need for a networking group, and that’s specifically to deal with things like firewalls reverse proxies that we’ve identified as key components but, you know, uh, so information mediator put together this document, you know, just wanted to highlight this and this shows kind of the flow right from a firewall load balancer to reverse proxy, You know, and that’s talking to a web browser. Right, and then how the building blocks, and are connected with API gateways through the security servers to talk to other building blocks, and so on and so forth. So this is a more clear, more clearly articulated version so anyway, short story, we need firewalls right, we’re gonna need reverse proxies. And then, you know, web servers probably it, we also need so we need networking infrastructure we need a networking. Building Block workgroup. Probably so. I think that’s been, that’s been surfaced to honey. Um, let’s see. So, any questions on that just interrupt me I’m just trying to finish this

up, feedback on that. Sorry, honey. A honey said, you know he could see that, you know potentially being a, you know key point. But it also comes down to, you know, when you look at that, it, the requirements would have to be really high level right because we’re talking about being able to deploy either on the cloud or on premises, right, or some sort of hybrid cloud situation. So, um, that’s a, that’s kind of where this group would come into play right is coming up with a spec that basically makes sure that we have something that will work across all those right cuz if you’re on Amazon, then they’re probably going to have a reverse proxy for you like they certainly do with SSL termination and they have a firewall and blah blah blah, right. So, okay, so, um, but somebody needs to handle that and it’s a big, you know, it’s kind of a big job. So, uh, we showed the inflammation mediator diagram, that’s all gonna be, that’s linked to in this week’s as well as last week’s document payments. Looks like that’s been resolved, they’re going to handle cash in some sort of fashion, but not cash trucks grips are going to review the data for. So you guys should review this draft use cases, you know, look at discourse that goes back to Global has everyone looked at that. So let me just show that quickly, um, we’re using it in the architecture team, and we’ve got all of the links to our high level documents right here, right, and then there’s the sub takut topic of meeting notes and then the meeting notes get posted there. And then people can see what the meeting notes were, and then they can comment, anyone can go in and comment right so it’s a little bit more open. So the idea is that we will, will have this available for other groups so you know each group has their own topic in here. And, for instance, you could put your, you know references to all of your documents in this page here, right, and then have you can have discussions and subtopics. So, um, yeah. And so, just wanted to point that out to folks, um, take a look at it if you haven’t already. Okay, and then. So, please, it’s a reminder to you know fork a copy of this GitHub link so here’s the link that we needed up above right. Yeah. Yeah. Yeah. Okay. All right so, and then, uh, huh. You know, I guess, I would give a demo and overview of how we use GitHub for editing diagrams. Does anyone want to see that really quickly. Probably not. You guys are kind of like GitHub savvy right. But, I mean, I’ll show you like a quick, quick example, right. So right now I’m not authorized. And this is on APA diagrams dotnet right so now that I’m authorized, I can like say move this down here. Right, it says unsaved changes click here to save. So I can save it. And this is a commit message, right. I moved the cloud. And then it commits it back to get. So it’s that easy, right.

Aleksander 1:57:10
So, yeah, but you’ll need some photo binding to your copy in GitHub I will do it

Max Carlson 1:57:16
right so in this is on after diagrams and net. So, in order to bind it, you would import from GitHub, right and then I’ve got like a million things in here so go stack working group, building block API, right, so then you could go in here, there’s diagrams, going as API gateways. And then there you go. Right.

Aleksander 1:57:40
Yeah, if you already have it but I will create your one output.

Max Carlson 1:57:47
Yeah, so a good question.

Ramkumar 1:57:50
So, deposit to GitHub, first time.

Max Carlson 1:57:53
Yeah, so this is the process where you would go to this page specifically, right. And then you would say, Okay, I’m ready to fork it, so I’m going to create a fork of it. Right, so this is going to create your own copy of gum stack Working Group building block API to your account. Right, so this kind of requiring, okay, okay,

Aleksander 1:58:15
yeah. As much as I know why cannot push changes to a throw for throw fork. Yeah, you

Max Carlson 1:58:23
can pick what you make it, you make your own fork and then you can push changes to it. But I pushed

Aleksander 1:58:29
to my copy, in that case,

Max Carlson 1:58:31
you’re exactly. Exactly. And then if you know this gives you a way to so once you’ve done that, then this gives you a way to effectively create your own copy but then if, say, in here, there’s a new updated version of this API gateway diagram, you can update your version as well, and it kind of harmonizes making those changes. So, Um, so does that make sense.

Aleksander 1:59:07
All the drums are here in one one repo, not not different folders for different building blocks or something like that. Well, I

Max Carlson 1:59:21
mean, you can structure it however you want, I mean you can just think of this as a starting point, right, so this is a, you know, intended to be a starting point that has a, you know, you can make your own copy of pretty pretty easily. Right, so, um, So, anyway, and then you can check in your stuff you can change the name of the folders, do whatever you want.

Aleksander 1:59:48
And it’s just sort of okay, but I will lose connection with other building blocks. I think that maybe, maybe it’s reasonable to have already here in this room, different folders for different building blocks.

Max Carlson 2:00:04
Well, I mean, maybe except the idea is that somebody can take this example API, right, and then modify it. Using come up put their own Swagger definition right in there and rename it to whatever they want, right, you know, same with the JSON schema, like here’s an example of a JSON schema. Right, so we give you these kind of examples sequence diagram. Right, so that you it’s the intent is for this to be a starting point for each group.

Aleksander 2:00:38
Right, yeah, but it would vary, but if you change from this example, drawing, or something, and if I don’t save by a new name, it will be disturbing for others. If in case I decide to merge.

Max Carlson 2:00:59
Well, yeah, I mean it might be, I mean, it’s kind of up to you right so for instance if I create a fork of this. It kind of depends on what organization you fork it into, but if I, if I create a fork of it, then I’ve got my own copy. And you can either create a copy of the file, in which case it won’t receive upstream changes, you know, or you can, you know just edit the file in place, in which case it would receive upstream changes. So, that’s kind of a fine point. It’s a good, good thing, good, good, good thing that you’re raising there. Okay, okay. So, yeah, so I think, you know, in some cases it’s It’s good, right. But I think I guess what we would encourage is that people, you know add to these diagrams, I mean these diagrams are still pretty useful to look at, maybe, you know, at least it gives you something that you can start with, in a, in, in draw app that draw an eye out, because that’s literally what I just opened up in here right. So, yeah, so I think it is that kind of helping answer the questions I mean I think we need to go through, you know, maybe next week when we have all the groups together, and just do like a quick overview of this, and answer questions.

Ramkumar 2:02:28
Does that would that would that make sense to people. Yes,

Max Carlson 2:02:34
I mean the thing is I don’t I think it’s an optional thing like we can’t force people to do it but I think it’ll be a good practice for groups to do it. If you can, so as far as I know information mediation has their own fork of this repository already right at least tailor made one, maybe he had to go.

Taylor 2:02:54
No, we, we created the our GitHub repo before this one.

Max Carlson 2:02:58
Okay, let’s find around.

Taylor 2:03:01
Like, I’ll try to move through all the structure of this one a little bit but

Max Carlson 2:03:08
yeah I wouldn’t even worry too much about it I mean this is just intended as a starting point.

Taylor 2:03:13
There’s no Yeah no I see it’s awesome, that’s awesome. I think I’ll just borrow the folder structure just so that everybody can sort of not get confused when they look at one of these building block repos, right. Yeah, I like the idea of standardizing it

Max Carlson 2:03:26
right, and then they’re like, you know example flows to, you know, which might, I don’t know, may or may not be useful. So just as starting points. Okay, cool. So I just want to make sure people are aware of that. Let me just highlight this a little bit more in here so that I can, you know, um, you know, so working inside your group. As you know,

Ramkumar 2:03:56
your team I might have to jump up, yeah that’s fine the meeting.

Max Carlson 2:04:09
Yeah, I think we got to go. I gotta go as well so I think that’s, that’s about it so I’ll try to put together an overview for the monthly meeting of this. Yeah, just kind of an overview and a live demo of how to use GitHub for editing diagrams. Yeah.

Ramkumar 2:04:34
Yeah, I bet it’s good to know. Probably as us. We are not what four months down the between. Yep, probably can kind of fall some milestones that we, we have to reach.

Max Carlson 2:04:50
Yeah, could be, I mean I think it’s sort of like, right now, so I guess there is this other question right, which is do we want to work on. Do we want to advance the sprint, like add more Sprint’s, add more logical process outlines or more use cases, or in flow diagrams or work on this more. I mean I think we’re kind of out of time today. But, you know, specifically, eyes. Parshin one,

Ramkumar 2:05:15
we’ve all boop submit version one of their document templates to architecture. And that’s getting reviewed. That is a good milestone. And then, we can still see from that point of view, cleaning up those definitions is one part, but then are there any use cases that are left out which are not covered by the existing connections is a good thing to check right and help us kind of see, Maybe they’re already covering things in the other 20 our use cases. Okay. Not done where we have to add those. Okay, review by the

Aleksander 2:06:01
picture group. Okay, great.

Max Carlson 2:06:08
So I think that’s, that’s a good way to think of it so, so everybody please submit revisions of your building block specification. When you feel like it’s ready. You know as soon as you possibly can so we can review it. Okay, good. I think that’s about it. I think we’ve covered everything. There’s, there’s some outstanding stuff. Right. You know I need to look on work on this for information mediator. Remember you can, you know, you can request a formal review, everybody needs to work on PowerPoints, probably. I think this is pretty much accurate. Okay. Great. Anybody have any other questions, concerns, comments, they want to raise. Before we break

Unknown Speaker 2:07:00
now. Okay, thanks, thanks. All right. Thanks so much everybody. Take care. Thanks, everybody.

Aleksander 2:07:08
Bye bye. Bye.

Transcribed by https://otter.ai