Platformland November 2021
Welcome to what I’m half thinking should be rebadged ‘primary key monthly’.* This month is quite identity heavy again, but there’s also libraries, postcodes and digital trade to get stuck into. It’s moved over to Substack because people kept on asking for something more bookmark-able.
🇮🇳 India is creating a new addressing system. Digital Address Codes will be linked to coordinates and organised via non linear geospatial grids to account for population density, and allow for non location addresses. The approach paper published by Department of Posts explains the maths behind how the schema was decided upon and lists targeting of benefit payments and e-commerce as the policy drivers. Could it also be a response to the likes of Google’s Plus Codes?
🇦🇺 Australia is attempting to use its MyGovID to seed an ecosystem of identity providers (see link to new legislation in the previous Platformland). New company directors must now use it to register for a Director Identification Number. December will bring an app that includes a digital wallet for credentials.**
🌍 G7 Trade ministers have made a pronouncement on digital trade. It includes a commitment to ‘single trade windows’ (a euphemism like ‘one stop shop’ or ‘portal’ that’s unique to the trade sector - really we are talking about software, data standards and APIs). This feels ripe for a digital public goods approach. ‘X-Road/MOSIP but for trade’?
🇬🇧The UK is building an app as part of its GOV.UK offering that will provide a common identify verification component for services. The team developing prioritised getting real users over the extended discovery and alpha periods that have become the norm in the UK civil service.
🍏 At WWDC21 Apple documented how their APIs for verifiable credentials work. The whole video is worth a watch, but Apple has basically adopted the Smart Health Card format (in turn based on FHIR and the W3C Verifiable Credentials standard, minus the blockchainey bit).
Software is politics
Researcher Ian Brown has posted a translation of China’s proposals for regulating interoperability of platforms. It includes categorisation by size, but also by category (e.g. financial services, compute, search, house rental, messaging). It provides an interesting checklist for policymakers anywhere to ask ‘who runs this here?’.
The EU now accepts the Covid certificates issued by the UK’s National Health Service(s) and other third countries. It’s a nice reminder that in favour of cross border interoperability being 97% politics, 3% tech and standards.
This is a great summary of what politicians can learn from Universal Credit (the UK’s digital welfare system). Including “There should not be a sequential process of top-down “Policy Decision” and then “Technology Implementation”” and the need to understand the underlying platforms of the finance system.
The practice of platforms
Occasional links to blog posts where people talk about what’s different about the job of making platforms…
Kat Stevens and Jonathan Harden from the UK Government Digital Service talk about how they migrated GOV.UK Pay, which is used by government services to accept payments from the public, to Amazon’s serverless offering, AWS Fargate.
Michelle Edgerley blogs about building a register of government organisations in New Zealand and the importance of genuine ownership.
Robert Karanja of the Omidyar Network provides a refreshingly non-hyperbolic overview of the state of digital ID:
Digital IDs can make it easier and safer for individuals to bank, vote, travel, obtain government services and safeguard their social media profiles and interactions.
But here’s the problem: Digital IDs can also be used to restrict freedoms, increase surveillance and actually make it harder to do many of the things they are supposed to facilitate.
And restates the case for standards and norms around digital identity platforms and meeting real user needs:
That’s why the global push to harness sophisticated, cutting-edge technologies to create digital ID systems can benefit from a focus on good policies, transparent processes and accountability. The ethos of the tech community is famously to “move fast and break things.” With the #GoodID movement, we can reconcile that impulse when it comes to digitizing society with the needs, experiences and rights of everyday citizens.
What if you apply platform thinking to Libraries? How do commercial platforms disintermediate civic interactions? Are libraries a prerequisite for government digitisation programmes? (It’s long, so recommend jumping to section 8).
Joe Tomlinson, Jack Maxwell and Alice Welsh document how the use of digital credentials to prove post-Brexit immigration status in the UK, with no non-digital route, may be unlawful under anti-discrimination law.
It is difficult to imagine a more important document than one which proves the entitlement to reside in a particular state: it is used as the manifestation of the ‘right to have rights’. In practice, this means that those individuals can only access and share proof of their immigration status online, including when they want to rent a property or apply for a job.
They note that the choice for a digital only approach is a policy choice as the agreement between the EU and UK only says that proof ‘may’ be digital, and there is no law requiring it to be digital only. Then go on to illustrate how, statistically, this choice effects disabled people, older people, and Gypsy, Roma and Traveller people. They give examples to illustrate the types of problems people may face, such as:
P has applied for a job. The employer calls P to arrange a time for an interview, and asks him to ‘bring along documents for the right to work check’. P brings his passport and letter of confirmation of status to the interview, but the employer says that he needs a ‘code’ from him. P does not have a smartphone with which to access his status. The employer offers to let him use one of its computers, but this is no assistance. P cannot remember the ID document or email address associated with his EUSS application and he does not know how to find and navigate the website. P misses out on the job.
This paper is interesting from a Platformland point of view in a few ways. Firstly, digital credentials are clearly on the rise and we’ll likely see more situations like this. Digital credentials in the hands of users can empower, but they when are associated with a mandate from government you can’t pretend there is choice for the end user. Secondly, this is a great example of academics unpicking how a digital service works (the paper includes example user journeys, for example). This should prompt the question: how can digital teams be more transparent about how services work so they can enable this kind of scrutiny. Thirdly, the authors make the case that a mixed (digital and paper) approach might alleviate the situation. If the problem of minting and issuing codes was solved once across government could it make that easier and cheaper?
* After all, most of government is mostly service design, and most government service design is about making it easier or harder to join databases together, and that space is the space of platforms. (I’m beginning to think the Artisan Integers movement was onto something after all).
** Is there good user research / HCI studies into if ‘wallets’ are a good paradigm for users? It probably is, but …