Interview with Mark Belinsky and James Green

UNDP Chief Digital Office / DPGA / Government Toolkit

Role of the United Nations Development Program (UNDP) and its Chief Digital Office. Also your areas of responsibility. UNDP mission: "Eradicate poverty while protecting the planet."

 Mark founded Digital Democracy way back when.
 Went to Silicon Valley, launched and sold a startup there.

 UNDP Digital Office was built right before covid stuck.  The team
 has been growing fast for the past couple of years -- from 6 when
 Mark joined to a bit over 20 now.  Gaining traction within UNDP.

   - thought leadership
   - digital transformation w/in UNDP itself
   - facilitating digital transformation w/ partners around

 Mark works on innovation and scaling for that third, external
 facing piece.

 UNDP makes products and offerings to help them.  Tries hard to
 understand what's happening in the field; bottom-up approach to
 DPGs.  Adding layers of best practice, insight and support to
 what's happening.

 Org is pretty decentralized, though Mark is in NY office.

 Mark is one of the only technical ppl on the digital side, says
 technical capacity w/in the organization is not equiv to a tech
 startup of course (quite normal for many gov and NGOs).

 Right now are in the process of trying to be aware of

 "Inclusive digital transformation" but trying not being overly
 prescriptive -- e.g., no "technologies have to be open source".

 We are adherents/signatories "DIAL Digital Principles".

** "Inclusive Digital Transformation" Working on nailing down exactly what that means; will "...having no citizens left behind." ** Best examples of DPG success you're aware of? Ideally includes: real-world usage, locally sourced support, reuse.

 Don't know if we have that yet.

 No central database of landscape of digital projects.
 DO started using ML/AI to categorize projects

 longevity, local support, a lot of these pieces are unknown b/c
 they haven't been categorized yet.

**** Examples of DPG failure? Trying to identify the failure modes.

 *IMPORTANT*: A lot of technologies that are built and abandoned
 generally have very strong project design before the
 technologists come in, and then there is some forceful pushback
 if they are encouraged by the technologists to redesign.

 How do we create these DPGs in partnership w/ offices so we are
 establishing best principles from the beginning: MVP, iterative
 design, making sure what we're creating is actually needed,
 testing all hypotheses.

 Often there are questions around OSS: is it secure, what is the
 business model, etc.  W/ external technologists there may be a
 relationship not necessarily correlated with expertise; if DO is
 going against what hired local technologist says, the project
 owner has to throw up their hands and says I don't know who to
 trust but I have a relationship with this local person, so...

 *IMPORTANT*: Mark requests: a flowchart for decision-making!
 E.g., UNICEF provided a really good one for blockchain projects
 (leading usually to don't use blockchain of course).  Mark will
 send it.

 atrium.network/guide (see p. 9 / PDF p 12)

 The decisions for this flowchart would be: "Should you use OSS?
 Should this a DPG?"

**** Who asks UNDP DO for help, and if so, what do they ask for? Runs the gamut.

 In the 200 countries UNDP operates in, their operatives will send
 requests, e.g., about something that has been requested by the
 local government.

 Can we implement a Digital Readiness Assessment as a starting
 point?  Too often it's ppl asking for a faster horse.

 The Digital Readiness Assessment is almost ready -- intent is for
 it to be a DPG itself, of course!

 Right now it's being converted to a kobo toolkit survey (good,
 OSS), but then is analyzed in PowerBI.

 *IMPORTANT*: Ask Mark about this later.

** Staff turnover at relevant agencies in Pathfinder countries? Lack the cohesive knowledge (of where projects are) to say one way or the other. ** Digital sovereignty In many countries, "proprietary" also by definition means "foreign". Is this an issue that drives decisions? Separate out data-storage sovereignty from running-code sovereignty.

 *IMPORTANT*: Mark's experience w/ digital sovereignty is that
 sovereignty is a more understood concept than digital, so
 perceptions about s are projected onto d in ways that are not
 necessarily fair or valid.  E.g., data has to stay in our country
 on our servers, but those servers are totally insecure and easily
 hacked.

 Local mgmt and therefore sovereignty of encrypted data in cloud
 storage is potentially a stronger real s than location-based s,
 but there's no opportunity to discuss that point.  Even country
 offices of UNDP aren't ready to.

 DIAL principles: security is just one piece, don't prioritize it
 above all else.

** What would you like to see in *** IMPORTANT: Use countries from the Global South as examples. Avoid giving the impression of digital imperialism. (When Obama was in the White House, the U.S. was an okay example, but not anymore. And not UK, Australia, etc either.)

  Mark may have some examples for us this soon; he's tuning their
  ML to try to identify them.

** <2021-08-20 Fri> Karl to ask Mark about progress on this. [ref:ff1bfdf0] :KARL:DUE: * Map DPGs to UNDP DO's DTF categories if we can. UNDP DO has a Digital Transformation Framework they've developed (the Digital Readiness Assessment is structure on top of that).

  https://digital.undp.org/content/digital/en/home/work/transformation.html

  See detailed diagram in [[file:archive/Digital Transformation Framework Full.png][Digital Transformation Framework (full image)]],
  which Mark shared from [[https://www.dropbox.com/s/3cwyfs4uepody56/Digital%20Transformation%20Framework%20Full.png?dl=0][this Dropbox link]].  Would be great to
  have a mapping of DPGs to the DTF pillar elements and Sectoral
  Opportunities, b/c this would help UNDP fill in the gaps.

**** Language barriers. Mark feels digital literacy is bigger barrier than language barrier per se.

 Same way definitions of "broadband" are terribly outdated,
 definitions of DL are terribly outdated too.

 Q: What is your definition of DL?

 A: Comfort working in a virtual environment.  Comfort utilizing
    digital tools (say, Dropbox and SharePoint -- virtual
    whiteboard, virtual filing cabinet, virtual war room).  How
    deeply can you use tools (e.g., Excel formulas).  And perhaps
    project management: collaborative/team management tools like
    Asana etc.  And personal tool management: ppl who are literate
    have lots of tips and tricks -- to have a tip is a sign of DL
    that Mark would like to see in everyone.

 Digression about mute buttons.  UNDP is a Microsoft house, so
 often dealing with Teams.  Had been using Miro.

 Q: So why is UNDP a Microsoft shop?

 A: Mark evaluates Microsoft tools in comparison to the many other
    things he's used, but for many people Microsoft tools have
    been all they've used -- so its problems are their norm,
    that's what they think technology just inherently is like.

    Also, DO is also tech office, and decisions about org-wide
    tech usage come from the CTO not the CDO.  So there's this
    dance that DO does around what it recommends vs what it does.
    CTO's focus is entirely internally-facing of course.  So when
    DO is building externally-facing technologies, do they use the
    internal back-end?  It's always a complicated question.
    Result can often be that something is done "off book" in order
    to just Get It Done w/o waiting a month for approval from the
    technology office.

    Note that this is a situation that many DPG-using /
    DPG-producing agenies may also be in sometimes.

**** James's notes from this interview. [ref:ed31d635] (James had to leave the interview partway through. Karl found his notes in 'belinsky-notes.txt' and moved them here in r18353.)

 What's your role in all this Pathfinder stuff?
 Mark:  you lost me with "Pathfinder"
 
 James and I know each other from the tech for good space.  Helping to guide
 efforts of governments to pull in and explore FOSS.  This office at UNDP arose
 right before covid struck.  We went from 6 when I joined to a bit over 20 now.
 
 Their remit:
 
 * Thought leadership
 * Digital transformation of UNDP
 * Digital transformation of partners around the world
 
 Mark works on innovation and scaling.  We're a remote office, mostly not based in
 NY.  Many of us have never been to the office.
 
 They adhere to the DIAL digital principals, which helps keep them on an open
 source path. https://digitalprinciples.org/
 
 As a development agency, we see our value-add as "inclusivity", which means "no
 citizens left behind".  No tech for tech's sake
 
 Signs of DPG success?  We don't have that yet.  What is the landscape of digital
 projects?  We don't have a way to track or note that in our systems.  We started
 doing ML on proposals to detect that a proejct is open or not.  So now we have a
 list of digital projects.  We're seeking info on DPG success:  longevity, local
 support.  They're unknown because they're not part of the original proposal in
 legacy projects.  In other cases, we're just not at that point in the process of
 collecting it yet.
 
 Are there DPGs that haven't worked out?  There are a lot of projects that get
 designed before tech people get involved.  Then there's pushback when the techs
 ask for re-design.  How do we get tech involved earlier?  Establish best
 principles from the beginning (mvp, iterative design, requirements gathering,
 testing, reuse of existing projects).  Often there are questions around open
 source-- is it secure, what's the business model?  With external techs, there
 will often be relationships, and it's hard for the Digital Office to correct
 anything because external techs are closer to the project and have the
 relationship.  And then they'll find that the external tech isn't doing best
 practices (e.g. the code is on a laptop, not in an online repo).  And
 procurement processes make it hard to replace that person.
 
 Mark wants a flowchart for decision-making.  UNICEF provided a great one for
 blockchain projects (pg 9 of https://atrium.network/guide).  Guide should
 address: Should this be a DPG?  How do you make the decision of what tech to
 utilize?
 
 [James had to leave this meeting, Karl kept going]