5 February 2019
Time | 21:00-22:00 East Africa 20:00-21:00 Central Africa 13:00-14:00 Eastern 10:00-11:00 Pacific |
Attendees | Akros: Ona: Craig Vital Wave: CHAI: UCSF DiSARM: Ali, Hugh |
Regrets | Vital Wave: UCSF DiSARM: Ona: |
Agenda
- Review of Previous Week’s Action Items
- Action item: End of month, Craig deliver narrative for version 2
Ona shared narrative for v2
- Action: Bid on reveal.io → waiting to hear back by Feb 23
- Follow up: Use case tests - Craig is working on this and will get over to VW end of Wednesday (Importing locations and coordinates)
This is done - just waiting on one for Nifi
- Follow up: CHAI to send the forms Botswana and Namibia are using
Sameen has the forms, Pierre has them too - please share
- Follow up: Craig to work with Connor to connect Jonathan re algorithm UI
To happen down the line, for R3
- Team Updates
- Akros
Thailand forms - need by Tuesday next week latest to not affect dev timelines for R2
- These need to be added to the data dictionary with field level detail and skip logic
- We are just waiting on Thailand
- *Puts the release date and schedule date at risk*
Field testing this week - share full results early next week
Subcontracts coming end of week
Akros to share Narrative, Configuration Doc, and Marvel by end of week
- Ona
Update
- V1.2 released yesterday with sync-based bug fixes
- Cleans up bugs found by Akros and VW during 1.1 testing
- Shipped documents to Akros:
- Narrative (Annie to add link)
- Major UI Changes for Review
- Configuration Document
- Field Testing configuration:
- Added the basemaps to the Android client
- Added points for three new districts to the mSpray dashboard and Reveal
- Added locations to OpenMRS and user accounts
- Craig is finalizing everything this and testing today
- Scoping continues
- Filling out data dictionary
- UI mock-ups shared with Akros (including video)
- Sam is scoping domain objects and high level engineering changes with the team
- Conor is beginning work on the Web UI on Monday
- Development on the family module is beginning tomorrow
- Need support on forms:
- We need to create about 8 forms as defined in the high level narrative
- GeoWidget work:
- The team is working on integrating the GeoWidget’s sample application with the Common GeoRegistry
- There are only a few items required
Form configuration - can we confirm when this is scoped for?
- Not ready to discuss this. Craig needs to discuss this with Matt and the engineering team.
- UCSF
Update
- Making progress on thinking through infrastructure and how users would make requests to the API. Started playing with code
UI - is a way to test and access algorithm without code
- Understanding is that it is not user-facing
- Could still use just won’t render the full picture in the UI but could export a file with the full output
- Craig: how often will this be run? In Reveal planning interface we have fields that say we have xxx LOE based on 5 team members hitting 5 structures each in a day
- Web service needs to be responsive to potential frequent requests. This is an intensive algorithm
- ***Annie: Good to get information on user expectations on processing time and number of points that they want to cluster/think through the flow of where clustering fits...how often they want to generate cluster, what kind of visual aid to understand that a cluster is processing. Also thinking about hardware investments for this. We could make this faster if there really is a need.
- Job queue would be important architecturally so we send a request to the queue, query for updates against the process and query for results once they are available→ get something written on what this looks like so Ona can look at for performance etc.
Two f/u questions on specs
- Minimum number of points per cluster?
- Every point would be put into a cluster, even a cluster of 1. BUT if you want to exclude clusters based on a minimum, that is possible
- Second round of clustering possible where you merge small clusters?
DiSARM M&E in Namibia and Botswana
- DisARM team starting evaluation of tool in Namibia and Botswana. Evaluating
- User satisfaction of the tool, issues with interface, which pieces most useful
- Feedback on training of tool
- Comparison between DisARM and paper-based systems - how disarm has improved, accuracy, completeness, saved time.
- Only paper based compared to disarm (district level)
- Also looking at how one person might be doing both paper and disarm (individual level) - does disarm minimize errors?
- Collecting data in March (1-2 months). Info to share!
- AOB
- VW - Oslo meeting last week
Great and productive
Reviewed each user story (priorities, potential ambiguity)
- Many modified to clarify functionality
Few big buckets of work discussed in detail: task management portion of DHIS2, usage analytics, etc
Going to go through same process with CHAI for user stories with Reveal (Pierre and Annie to do tomorrow)
- Maybe we should split Reveal user stories into Android and Web/Core
- Mostly splitting between the two? Or not? → some were split but many were already separated into one or the other (in UiO this makes sense because there are many different product groups and understanding backend need for user story to see what is actually achievable from a timing perspective)
Rescoping and descoping based on case notification and investigation → likely many involve Reveal tool
- Roadmap Check-in
Alignment of R2 functional roadmap and Roadmap user stories
- Annie, Pierre, Craig to discuss
12 February 2019
Time | 21:00-22:00 East Africa 20:00-21:00 Central Africa 13:00-14:00 Eastern 10:00-11:00 Pacific |
Attendees | Akros: Annie Ona: Craig Vital Wave: Derek, Pierre CHAI: Anne UCSF DiSARM: Ali |
Regrets | Vital Wave: UCSF DiSARM: Hugh Ona: |
Agenda
- Review of Previous Week’s Action Items
- Action: Bid on reveal.io → waiting to hear back by Feb 23
- Action: Akros to share Narrative, Configuration Document, Marvel mock-ups by end of week
Done
- Action: Akros team to get information on user expectations on clustering algorithm
In progress
- Team Updates
- Akros
Mocked up ento forms
- Forms needed for FI
Intervention Type | Status of Reveal Data Dictionary |
LLIN distribution | Mocked up from knowledge, don’t have sense of Thai forms |
Breeding Site/Larvicide | Mocked up from knowledge, don’t have sense of Thai forms |
Mosquito Collection Point | Mocked up based on IRS routine intervention, don’t have sense of Thai forms |
RACD/Mass Blood Screening | Mocked up based on Thai forms |
IRS | Mocked up from knowledge, don’t have sense of Thai forms |
Behaviour Change Communication (BCC) | Guessing this is a yes/no, don’t have sense of Thai forms |
Case Confirmation | Abbreviated mock-up based on Thai context, need validation from Thailand |
- Still waiting on hearing back from Dr. Prayuth. Pedro is looking at the data model to see if we can get any more specific.
- How much of a delay will not having the forms have? We will have placeholders in for now; we could release un-updated forms and then do another dot release with more specific forms. The indicator reporting is going to be the biggest thing that is slowed down, DHIS2 integration (not relevant)
- Reveal release 2 might have a few release - a “big” one with the client view and then a staggered release into April
- Integration budget approved
- Workplanning session two weeks back → likely these timelines will shift since we are still waiting on Thailand, would be good to pin down dates for first visit
- March visit for CHAI team to touch base and confirm timelines
- Visit after that will be larger team
Field testing was delayed one week - will share documents first thing next week
Desk testing updates - now on Rv1.3
Demo video in progress
- Demo walk through for Bots/Namibia context
- Is goal of demo to sell either country on Reveal?
- Screen by screen walk through (video?)
- plus documentation to walk through, catered to their forms
- Access to test demo test instance
- High level demo useful for other countries
- What is the difference between Botswana and Zambia forms
- Unit of capture
- Data fields
- It is fine to demo just with the demo fields updated
- To do: Annie to cross compare fields for Namibia and Botswana
- Would want to make decisions in Q1 to pilot in Q2 - would be best to do this before the data review meeting in early March. Prioritize Namibia. Botswana timing will fall around DisARM M&E timelines which are tbd.
User Story review
- Annie and Pierre reviewing this week
- Pierre documenting notes/justifications in test cases -> How would we like this presented to all?
- Pierre to set up slide deck with slide per user story with thoughts/notes/motivations for any changes
- Changes - descoping or editing language
- ONce we come to consensus, move to roadmap as source of truth
Subcontracts sent
- Ona
Rv2 scoping update
- Scoping is expected to be completed by Friday this week.
- Roger and Craig have clarified mock-ups that were shared.
- Finalizing forms this week
- FI screens mostly done, need to define flow → how do users get in and out of family module?
Development
- Reveal Web UI is actively under development
- GitHub repository is now out → project on ona
- Supported and released Reveal v1.3 - Sync based bug fixes (to do: share apk)
- Also configuring new operational areas for testing
- Working on integrating family module this week
- Adding head of family, adding members to family, button to collect form against family member (form itself still in works)
- Archive individual or archive family
- Working on adding different location types to the map
- Residential structures, larval breeding sites (point, but polygon possible), mosquito collection point, temporary location (community meeting point)
- What about temporary sleeping location (e.g., that could be linked with a permanent residence)? Could be necessary in the future, would have to add icon, entity, add to maps. May be down the line.
- GeoWidget:
- Scope defined for next release
- Drawing arrows between index cases
- Core GW requirement → Reveal queries database to return location ids in date time order and then passes to GW, then GW draws arrows
- Drawing radius around my location at varying zoom levels
- As user zooms in, the 75m radius is zoomed at the scale/zoom level
User story alignment for Rv2
- No major updates here
- UCSF
API update from Hugh - will check in next week
Finalizing M&E protocols with CHAI, Sameen getting in-country feedback, orienting NMCPs to protocols. End of April/May results.
- Action: Evaluation questions - Ali to share protocol.
Annie to loop in Ali into contract conversations
- AOB
- RVLv1 Sign-off
Run through all user stories with CHAI → two that are outstanding
- Action: List of error messages needs to be sent to VW/CHAI
- Demoing the sync up to mSpray dashboards, Pierre ran through this today with Kelvin and running through with Anne tomorrow.
- Great job team!
19 February 2019
Time | 21:00-22:00 East Africa 20:00-21:00 Central Africa 13:00-14:00 Eastern 10:00-11:00 Pacific |
Attendees | Akros: Anna, Annie Ona: Craig Vital Wave: Derek T, Pierre CHAI: Sameen, Anne UCSF DiSARM: Hugh |
Regrets | Vital Wave: UCSF DiSARM: Ona: |
Agenda
- Review of Previous Week’s Action Items
- Action: Bid on reveal.io → not an option
Porting to that site by end of week. Will share for feedback
- Action: Akros team to get information on user expectations on clustering algorithm
In progress - Derek P. has reached out to ministry stakeholders
- Useful to understand which of those requirements are critical versus nice to have, ie which would prevent the user from using the algorithm versus nice to have.
- Stratify need
- Demo,
To do: Annie to cross compare fields for Namibia and Botswana
- This is done. Need to define next steps
- Craig - these are in the data dictionary now - can you take a look and see what the effort would be to pull this into an apk.
- What are we trying to sell? Is the Android piece enough (data collection forms)? Or do we need to take this into the dashboards for the demo.
- Sameen - we don’t need to prioritize the dashboard. Android components so that could be user tested, fill out the form and see an operational area that is similar to their expectations. Each village has a targeted denominator.
- From an imagery perspective - for demo Zambia imagery is fine.
- Contextualization of tool, renaming of areas, so they can see it aligns with workflow.
- To do: Sameen to send hierarchy.
- Immediate needs for demo:
- Update android client data fields (form)
- Renaming of Zambia operational area to familiar village names
- Anything else needed for demo prep? No.
- Fixed denominator will be required for Namibia. What kind of timelines do we need to take into account for customizations
- We need to get a sense of what configuration timelines are
- To do: Akros to get sense of what additional customization would be needed for DiSARM and getting that to Ona
- Country would want tool ready for internal testing in June/July
- Sameen working on timelines to get decisions by end of April.
- Will need to be a discussion if we select both countries
- What needs to be done for DiSARM this year? To what extent will countries be using the app?
- Should also know in April.
Akros walkthrough document
Akros video
- Both of these are the generic demos at this point
- Supported and released Reveal v1.3 - Sync based bug fixes (to do: share apk)
Done
- Action: Disarm Evaluation questions - Ali to share protocol.
Done
- Rv1 signoff
Action: List of error messages needs to be sent to VW/CHAI
- Craig working on template to give to Sam to explain.
Action: Demoing the sync up to mSpray dashboards, Pierre ran through this today with Kelvin and running through with Anne tomorrow.
- Found a bug in geoJSON. Kelvin identified this and a nifi fix that will be fixed tomorrow
- Team Updates
- Akros
Mocked up forms
- As far as they can get for now, will be modifications once have full feedback from Thai team
FI level setting
- Need for three years historical data?
- Out of scope ETL for past three years of data
User Story update
- We would like to present proposed user story changes to VW and CHAI
- This will likely take 2 two-hour sessions
- To do: First can send a spreadsheet and document for review
- Will share by end of week
- DiSARM
Waiting to confirm the scope of work, neat initial work on algorithm and UI side
- To do: Akros to respond to SOW clarifications.
- Ona
Still working on scoping the Android client
The data dictionary is nearly complete with follow-on task actions and forms defined. When we add a location or register a location, we auto-generate tasks based on the campaign type that’s selected.
The family module is integrated internally and we are working on modifying it to fit the new mock-ups.
Working on Add Location form right now
Need help with:
- Focus Investigation reporting that will need to be in the Web UI
- What new objects are needed. Some summary data, how its stored/accessed/edited → need to start these conversations now.
- To do: Derek to share mockup with CHAI side. Anne to solicit feedback
- Craig will send the Android client mock-ups over the next couple days to Akros with a walk-through. We will need the review process
- Craig wants to flag the size of this scope and identify that we will need a number of releases with an MVP delivered by the end of March and a number of features in a dot release
- AOB
- 8:30: Integration conversation
General Info
- Get access to a staging environment of DHIS2 for the country
- (possibly) share any high resolution imagery with DHIS2
- Need to align the org unit hierarchy with Reveal
- Need to export the GIS info from DHIS2 and import it into Reveal
- DHIS2 build
- Possibly add a different GeoLevel to capture operational area below the current lowest version
- Forms need to be created
- Tracked Entity Instances need to be setup to track locations within an operational area
- Operational areas need to be defined
- Families, relationships and family members need to be created as tracked entity instances
- Relationships need to be made available between the location and family TEIs
- Data Mapping
- Ensure that the org unit hierarchy aligns
- Map the field values from the form to
Namibia
- API build - who doing from DHIS2 side
- DHIS2 build
- Data mapping
Botswana
- API build - who doing from DHIS2 side
- DHIS2 build
- Data mapping
26 February 2019
Time | 21:00-22:00 East Africa 20:00-21:00 Central Africa 13:00-14:00 Eastern 10:00-11:00 Pacific |
Attendees | Akros: Annie Ona: Craig Vital Wave: Pierre Dane, Derek Treatman CHAI: Anne UCSF DiSARM: Ali, Hugh |
Regrets | Vital Wave: UCSF DiSARM: Ona: |
Agenda
- Review of Previous Week’s Action Items
- Demo, Craig to estimate when this can be done based on conversations last week
Immediate needs for demo:
- Update android client data fields (form)
- Renaming of Zambia operational area to familiar village names
Team has scheduled in over next two weeks to share by the 8th of March
Anne to confirm with Sameen over next week. CHAI team happy to handle demos for now
- To do: Akros to get sense of what additional customization would be needed for DiSARM training documentation and getting that to Ona.
Derek P to have this to share at end of next week (March 8)
- Rv1 signoff
Action: List of error messages needs to be sent to VW/CHAI
Action: Kelvin identified bug in JSON and a nifi fix for syncing to work
- This is fixed now.
- To do: User story scoping - Annie and Pierre to send a spreadsheet and document for review to Anne DONE
- To do: Akros to respond to SOW clarifications to DisARM
Done
- To do: Ona to respond to contract
- To do: Derek P to share FI dashboard mockup with CHAI side. Anne to solicit feedback
Annie to share with DiSARM → done
Input by next week’s call
- Team Updates
- Akros
- apk v1.4 Username: DSMECOP, Password: Amani123
- What is the process - VW has a feedback form for capturing issues
- Akros webpage will have a contact email for an issues with installation
- We probably need a support channel that is responsive - there is a risk around making this so public..maybe we have a contact line that is a better sales pipeline
- To do: Annie to circle back on this before putting it live
- To do: Google Issue Report Form - to discuss when this should be implemented
Focus Investigation Dashboards - feedback needed by this call next week
To do: Imagery update - better update to come by end of next week (March 8)
Testing feedback/final release outputs
- A final tested and edited release x version
- An up to date bug tracker with all relevant issues addressed (internal)
- A post testing edited script for our grant partners to follow
- An updated 'new user stories' document
Annie on leave March 2 - 16
- Fully offline first week
- Partially online second week
- Anna to back-up and manage while Annie is out
- DiSARM
Feedback on algorithms discussions
- Contracts being received by UCSF grants team
- Looking at ways to structure and document API
- Derek reached out to people on algorithms
Update on Reveal Seattle trip
- Concept note ongoing
- Will circle back to VW to get feedback at a later stage
- Craig lives in Seattle!
ASTMH
- To do: Anna to submit symposium tomorrow
- Ona
Scoping:
- Final day of UI mock-ups on Android client
- Data dictionary is set
- Working on the data model for Focus Investigations and the relationship between Intervention and sub-interventions (FI vs. Campaign vs. Case)
- Received the FI reporting scope today and are iterating on it
Dev:
- Finalizing Family Module
- Updating Circle
- Reveal 1.4 release and providing data support
Demo Prep:
- Namibia and Botswana APKs and Operational areas are scheduled - Target delivery is Friday March 8th
- Cambodia CGR - On our radar, but haven’t yet assigned the team members. Target is to deliver by end of March (29th)
- Server side OpenSRP and CGR with Nifi integration
- Stretch goal is if new point is created in Reveal that it would queue to be added to CGR
- This is a POC integration
- Cambodia is focus for CGR right now, opportunity to discuss using tools like Reveal as way to interact with CGR.
Questions:
- Marvel updates? When will client-side Rv2 mock-ups be ready to share?
- Target is end of Day Thursday
- Release 2 dot releases - can we schedule these out?
- Yes, it’s on our to-do list, but has been deprioritized so we could get the developers unblocked on a number of things.
- Target to have this would be mid-week next week.
- AOB
- Rescoping of user stories for Reveal
Want to put some time on the calendar