Strategies for a Sustainable OpenLMIS Community (GOV)

Owned by Sandy Hawley (Deactivated)
Last updated: Oct 03, 2015
5 min read

Description: What resourcing options are available to sustainably support the OpenLMIS community and implementations? What funding or revenue options should the community pursue?

Leads: Mattias, Lindabeth

To Join Remotely: 

Join the call: https://www.uberconference.com/info3285

Optional dial-in number: 716-293-6106

PIN: 63498

Rapporteur/Notetaker:Sandy Hawley

Notes from Session:(SEE Tanzania handout from LB and Dykki's slides)

Sustainability Questions with Current Implementation

  • Being that the dev. is in JAVA- it's not a good talent pool for required skillset
  • How can we provide country resources to build capacity or at least maintain at a regional level?
    • what do we expect countries to maintain themselves
    • maybe Tier 2
    • do we expect T3 to step up/step in ? 

  • TZ/ZM provided stats - what happens after VIMS; what are their maintenance models?
    • some data from email  
    • not everything is transferrable
    • in Pakistan- all support is in country, provided by JSI in country staff
  • Where is the cut off- where do we need to focus our resources to work with countries on transition plans? 
  • VR- MOZ (500 facilities, 20 users, collecting at each facility- TZ has 5K users)
    • Country is not taking on tech support yet; in both cases its not used nationally right now - would need related capacity effort to support that. VR providesM&E/Tech support for Tier 1 and Tier 2
    • challenge; VR has funding for overall supply chain work to implement and maintain, but not significant .08 Developer- scale of implementation is not as large as TZA
    • when that project ends, if Gates doesn't continue to support, VillageReach would need to fund
  • Benin - ADP hired VR to do technical assistance; self funding relatively small amount - if it were to scale large, the sustainability would be an issue 
  • CHAI- starting with a group within MOH to do T1 and T2

Tiered Support

  • T1- can't log in, can't go to next screen
  • T2- advanced support did everything right, but not getting right answer
  • T3-  more of a bug; possible code corrections, could include features if we had maintenance fund--
    • this should be provided by the community (mixed feelings)
  • Several implementations have not been doing this - USAID could continue to support or Gates rep could provide some cover with ARC- what does that model look like?
    • 8% Dev. in Benin, X% Dev. in MOZ
  • T4- "system administration"
  • Country governance role
  • If its not a severe defect, and there is a workaround (Community Security Council) decide on product roadmap; defect would go into the cue

What to do now, key issues

  • The expectation was that this has gone longer than anticipated - budget is an issue - we don't want it to fail, but what can we do with drastically shrinking budget?
    • $16MM to 4MM in Zambia - HIV TA $500K only
    • oenus lies with country; conceptually they understand, but they are not in a position to pay/sustain
    • if they want support they have to pay for it
  • First generation deployments- don't we have the responsibility to take on T3 support , but we are not taking on T1 or T2
  • Community is in favor of upgrade path, they join the path as new implementations, but this will depend on whether the upgrade path and what they want (timing wise) are happening at the same time.
    • doing that upgrade is a significant cost 
  • Things that are in the community supported category should be paid for by communityEven if we wanted to transition, the skills are not available - NOW; need to onboard them, train them, that's another project
  • It shouldn't be, (was the project scoped with maintenance plan?) frame the products as much as you can, start using local resources
  • DHIS2 is defacto for a reason, they were originally a university program set up to build capacity, grad students capacity first, THEN system- ownership was integral
  • ramp up takes time

Regional makes sense- See also Dykki's Slides

  • Breaks dependency of someone inside TZ; single group could cover multiple countries
  • If there aren't many T3 issues, partner or local dev firm could absorb, Kenya, Ethiopia - contract for support of these entities, customization, system admin
  • Software, config, domain expertise - ARC doesn't exist yet-- but there has to be others who are doing this so we don't have to: GEMBY eHealth 
  • Cases: iHRIS
    • East. Central and Southern Africa Health Community- regional associations for MOH
    • West Africa health Organizations
    • paid by countries (partner/donor $)
    • Leader of Stakeholder Leadership Group - went to present to these groups at conference
      • peer to peer is powerful
      • more countries came on board, ECAH became part of team, all major E. African teams became bilateral - with USAID support along the way
    • Dissemination pilot to 13 regional leaders, sharing results  
    • Video: A West African Perspective on Open Source (YouTUBE)
      • great insights on value of os - against proprietary products where vendor controls price
    • Country rolled out system w/o partners - asked countries to go to donors and ask for community to support implementations
    • Product hasn't matured enough to deliver on the "download" promise - OpenLMIS  needs to go through that process
    • East Africa has a resource; that is why ARC is focusing on West
      • as it forms, OpenLMIS should partner, be involved closely
    • Nothing is more powerful than country engagement
    • Academic partners are key, especially in TZ - UCC, UD; foundation knowledge is there
      • MCSP is embarking on eHealth architecture activities
      • A lot of local capacity
      • partners need to work out who runs it (some competitive nature there)
    • World Bank, EU, DIFID - local teams have pulse on funding opportunities  
    • Is having countries run this the long term goal? If it is, have to take steps to get there from the beginning
    • Saas- a potential model, implied support contract that is part of the service cost; if the math works out, its sufficient
      • we might need to broaden our target , 3PLs involved, has to be more than MOH customers
      • how would impact last mile environments  would the negatively effect product
      • humanitarian space - Word Vision, Mercy Core- is anything different?
      • DHIS2- adopted by countries for M&E
    • Does  iHRIS still have support for core product? - at the time
      • country implementation
      • core product community
      • what do we need to do to stay relevant
      • some fail, some success- need foundation model
      • Global Goods, Apache, if something could come to exist in this vein that could aggregate funders around common investment

Opportunities

    • OpenLMIS move to a Foundation
      • stage approach - we are in 2nd phase
      • but this doesn't solve how this entity continues to get funded
    • Groups working on sustainable financing - that generate revenue to fund drugs, can we tap into those groups?
    • Private Sector
      • data sharing agreements with countries so we could share with 3PLs and pharma - ways to make it relevant for others, buy-in
      • what if countries sold directly to PS, some brokering would be required
      • is it possible to license it for profit if revenue comes back to community - NO- then its no longer opensource
        • hybrid approach means dual licensing, usually you are selling a license to a slightly different product with more stuff (proprietary piece of your code base with special features they are paying extra for those features; marking up costs to ensure others can afford it)
      • Google model- selling space for olmis (HIV drugs, etc) (smile) (this was a joke)
      • Long-term - start planning for democracy of olmis
      • Near term- millions $ commodities managed through this system- the cost of maintaining this system is minuscule compared to cost of products- the cost of visibility
        • at least have information about supply chain and M&E and how they relate, ROI, why its valuable, it's not a one-time implementation
        • 80% software costs is in maintenance
        • countries can't see the financial benefit - when they are fully stocked- it costs nothing. When they know they are stocked out, they see the cost it takes to maintain.
      • Countries don't appreciate or prioritize
    • Labs CHAI- trying to understand support and maintenance cost (transaction tax)
    • Who holds the money- CHAI reached agreement with suppliers (authorized service providers)- proactively look at health, provide service on demand
      • contract is between MOH and manufacture- they are in charge of their local agent in country (BID?)
      • if you get to a certain volume you get economies of scale
    • Commodities are different than other health sectors

 

Artifacts:

Session recording became too big to upload, please contact kevin.cussen@villagereach.org for a copy.

 

 

OpenLMIS: the global initiative for powerful LMIS software