Thoughts on the GDS Services Week Open Show and Tell

I’ve finally gotten around to watching the video of this that was posted on 6th March.

I haven’t had the time to watch many of the Services Weeks videos yet but by all accounts is was another excellent showcase of great digital work being done by teams all across government.

Okay — the agenda for this video is:

• GOV.UK Accounts
• GOV.UK Starting a business
• Standards Assurance
• Design System
• Remote accessibility research
• GOV.UK Pay
• Collecting info from users

My first thoughts are that all of those are important but none of them feel powerfully innovative. The Service Standard continues to evolve (particularly with regard to data) but has been around for nearly 10 years. The Design System is a vital piece of reusable technology but is almost as old. GOV.UK Pay goes from strength-to-strength and is finally starting to make serious inroads into local government but, again, has been around for some time. “Starting a business” sounds like the famous Tom Loosemore presentation at Code for America in 2015 so it’ll nice if we’re finally catching up to that, as much of the ideas in his wider presentation feel as far away now as they did then.

The one thing that does stand out is GOV.UK Accounts which. As I have written about before, this is a very interesting 180 degree change from GDS’s original stance of citizen interactions with government being based on individual one-off transactions rather than long-running relationships. I’m generally a fan of this switch but there need to be many careful public discussions about centralised data storage of basic citizen data and under what circumstances that can be reused / shared (I’m very glad to see GDS already leading those discussions).

It’s interesting to hear someone say that “every team blogs about there work”. It feels to me that the volume of blogs coming, not just from GDS, but from government digital teams in general is significantly less than it was five years ago (I keep thinking of writing a script to check but my coding skills are very rusty). There are good fairly regular updates on something from GDS but only GOV.UK Accounts feels like it’s being discussed openly with any frequency. Meanwhile, over at HackIT (alas before the cyber-attack in November) we were posting weeknotes for every team and putting all of our show and tells on YouTube. We hope to start being able to do both of those again very soon. As I’ve been saying for many years — there’s no reason why every GDS team couldn’t model this kind of behaviour to encourage all government teams (not just Digital ones) to do the same thing.

GOV.UK Accounts

It feels weirdly preemptive to talk about Accounts when there isn’t any clear cross-gov digital ID strategy (in fact with many departments spending public money to develop their own non-joined up systems). This presentation feels like it glossed over the login part of the service. There’s some impressive iteration already and lots of good thinking going into cyber security and user needs. It also sounds like it could be a critical component of the upcoming work on Digital Attribute reuse (see the recently closed consultation and my response) — especially if GSD were to develop a GaaP component that recorded which citizen’s digital attributes have been viewed / shared by who and why (as already exists in the Estonian X-Road system).

GOV.UK Starting a business

It’s remarkable, but not sadly unexpected, that in 2021 we’re still talking about how teams overcome the friction of working across government organisations in order to be able to start to deliver joined-up end-to-end digital services for citizens. If the change in GDS leadership wanted to focus on something radical that could drive real change they put a team on looking at why that is so hard and encourage them to discuss openly the issues they find and what should be done about them. This kind of thing should be so much the norm that we don’t even think about it — but we’re still a long way from that. It’s great to hear that there’s a cross-gov “start a business” community but how would I know that such a thing exists? What other really useful communities of interest are out there? This is exactly the kind of thing that GDS should be maintaining on the Service Manual and encouraging people to communicate about in the #sevicecommunities channel in the cross-gov Slack (no meaningful posts in there for at least six months).

Standards Assurance

It’s great to hear the GDS wants more teams to do assessments. That said — it’s a decade or so since Service Standard assessments started and I still don’t know where to look for a list assessment results. For example, how many DWP digital services costing over £2M have failed an assessment in the last 5 years? Or how many HMRC services have had recommendations related to accessibility? That kind of data just isn’t available. This is even more crucial when it comes to GDS’s initial strong level of government that no service over £5M could go on GOV.UK (baring in mind no service was allowed to go anywhere else) unless they passed a Service Standard assessment. Is that still true? We know it’s painfully not been so for a number of covid-19 services but what about more generally? The data isn’t available. Considering how critical the Government Service Standard (n.b. not the Digital Government Service Standard any more) should be in ensuring quality assurance of government services the resources available to the team to do their basic work (never mind expand the use of the Standard to non-digital government functions) is limited to say the least. It feels like Cabinet Office continues to not take this quality assurance function seriously.

Design System

Not a lot to say here. The Design System is great — one of the best achievements GDS has made (and that’s saying something). It’s great that there’s a lively cross-gov community of people of people contributing to this as well as using it (900 live services and over 2,000 github repos pulling in the Design System code). A good example of tailored reuse is our own HackIT Design System. It’s also really good to be reminded that the Design System isn’t just about code that creates web-pages — there’s also a lot of advice as to how to design those pages to deliver the best experience for our citizen users.

Remote accessibility research

Interesting discussion — glad to see continued focus on this important area during the lock down. More of a “how you can” presentation than an overview of how something is going.

GOV.UK Pay

Really impressive stats showing off why GaaP can be so important — 500 services in 160 organisations collectively taking over £1M a day. It’s great that this is starting to also happen in local government — albeit slowly. This presentation covered adaptions made to GOV.UK Pay to support call centre staff working at home during the pandemic.

Collecting info from users

Nice to see some user research being done into broad stats of how information is collected from citizens. Only, we did that already when “GDS Submit” was a thing maybe five years ago. I’m pretty sure the outcome of the user research, especially regarding accessibility and the issues with the “long tail” of of undigitised PDF forms, hasn’t changed since then. The speaker says “There is no way to produce an accessible easy to use online form on a small budget” and I would definitely agree that that is the case today. As I’ve written about a number of times in the past, the solution for this is a GDS run “service builder” solution that any government organisation can use for free. Built on the Design System such a service wouldn’t even be all that hard to create except it’s currently basically impossible because departmental data is locked away behind individual organisational firewalls. I’ve been highlighting a lot of “data capture” issues in legacy local government systems recently but it’s in central government departments where this is perhaps the biggest issue and the major blocker to a GaaP “service builder” service. The API community is making slow inroads into this but there’s a very long way to go. Meanwhile, the fact the form-builder team in GDS is worried about not even getting funding to further investigate this really highlights the current lack of dedication to supporting GaaP beyond the existing successful products.

Summary

All-in-all there’s a lot of very good stuff here. The issue is that little of it is new and much of what is here is struggling to be supported with enough funds to make it viable. I can only hope that the new management at GDS is able to persuade the Treasury and other government departments that further investment is needed in the centre of government to provide the GaaP services we need and that money will be saved in the long-run by not duplicating these efforts over and over in different organisations.