Thank you, Liam

I first met Liam Maxwell in February 2010. It was at a Tory-dominated event called “The Network for the Post Bureaucratic Age”.

Not one of my usual stamping grounds.

William Heath and Tom Steinberg had urged me to go, but it was only the chance to hear a keynote by my hero Heather Brooke that persuaded me to go.

I found myself standing at the back next to a man who joined me in nodding fiercely as William Heath launched into his familiar IdealGov rant about the abject failure of the civil service to embrace the opportunities of the internet age.

We got chatting. Once I’d got over the shock of his being the Head of IT at Eton, I was impressed.

Here was someone with a well-informed, and very healthy anger about the waste and inefficiency of government IT. He knew the scale of self-harm and wasn’t afraid to shout about it.

In 2016, the kind of ill-fated multi-year, single supplier  IT-led programmes which were the norm in 2010 are no longer acceptable within HMG.

Liam changed the weather.

That didn’t always involve making friends. Such change rarely does.

Liam and his IT spend approvals team coming to work for GDS in April 2013 was a crucial moment.

The joining up of those demonstrating new ways to transform services with those able to stop ‘bad’ IT spending provided a powerful carrot and stick with which to challenge and change behaviours across central government.

Liam does good stick.

And boy, from 2010 – 2015 did the government IT crowd need his stick.

Which is probably why now is good moment for him to move on from the role of CTO for government.

Digital government has entered a new phase.

Hundreds of teams all across government are learning to work in new ways: agile, iterative, user-centrered service design, supported by new platforms like GOV.UK Pay, properly managed data via registers, and using the 1000s of suppliers and cloud services available on a pay as you go basis through the Digital Marketplace (itself a legacy of Chris Chant & Denise McDonaugh’s stick and carrot double act) –

Those teams want to work this new way. They include some amazing people. But they need support from the centre. It’s a more collaborative time. As Stephen Foreshew-Cain (GDS’s new boss) says, GDS has got their back. Less stick needed.

Liam’s moving to a more political, special advisor-like role, to which he’s admirably suited – and I don’t say that with any sense of snark. Political savvy is vital to changing government.

Bringing someone with his knowledge into this more political court is long-overdue. The technology rhetoric coming from senior politicians is often embarrassingly ill-informed.

So I look forward to hearing Number 10 et al taking more practical and enlightened technology policy positions;

Less of the damaging ill-informed knee-jerking about encryption. (As any teacher knows, you can’t ban maths.)

More about the need to stop smooth-lobbying incumbents messing up the level playing field for disruptive tech innovation.

And much more about the importance of a level playing field to let *everyone* in UK gain full benefit from this digital revolution.

And I’m sure he’ll continue to support GDS from his new perch.

Good luck, Liam. And thank you.

The UK owes you one.


“Distribute power as widely and irretrievably as possible”

From Danny O’Brien’s 1997 email describing a trip to SF to meet Louis Rossetto, founder of Wired magazine. At the time Danny and Tony worked on Wired’s UK edition, as did I.

                                                                – * –

San Francisco is such an optimistic town, and, while Wired itself is actually a pretty mundane place to visit, its very ordinariness filled me, briefly, with the belief that Wired UK was possible. I loved meeting these people, loved the culture they sought to represent. It was only on the flight back that I got worried.

On that flight, I got into a blazing row with Tony Ageh; the only row that I've ever had with him. Tony isn't efficient. He's not really wired. But he's a very charismatic man, and I'd say a visionary too; he's easily a match for Louis in that.

We were talking about something that Louis had said, something that made Tony uncomfortable. Louis had talked about Wired being for 'the digerati' - 'guys who'd been teased at school for being geeks, and who are now earning millions while their schoolmates flip burgers'. Wired UK, Tony insisted, was going to be for everyone - even the burger flippers. I stuck to the Louis line. I said, these people are defining the future - you've got to write a magazine for them. So why, said, Tony, can't everyone define the future? Isn't that what it's really about - everyone at last, defining their own future? Can't this magazine help create a place where everyone can take these tools and use them for themselves?

What Louis is doing, he said, smacks of revenge - of comeuppance. He's turning a global change into a vengeance tale. And so it went on, me insisting the value of representing the thoughts of the wired, Tony claiming that the function of any magazine he produced was to distribute power as widely and irretrievably as possible. We almost literally came to blows, rocking and pushing on the plane even though we were only an armrest away. The argument was never resolved; eventually we fell asleep, awkward and furious.

Hours later, I woke up. We were above the Arctic, and through the frosted porthole, I watched the Northern Lights, something I had dreamed of seeing since I was a child. I didn't wake Tony. I made him miss it all.

                          - * -


Whenever I hear that most duplicitous of phrases “the sharing economy“, I think of Tony Ageh and his vision for digital technologies designed to distribute, not concentrate, power.

Sadly, he was right about the revenge of the West Coast digerati. Code and capital begat power. And power duly corrupted.

Ten rules to help design a digital, democratic government

The peerless Richard Pope has published 10 rules for anyone redesigning a democratic government to make the most of this digital era.

Richard’s 10 Rules for designing a digital, democratic government are:

  1. Split data from services. Hold it in organisations with appropriate accountability (central government, local government, professional bodies) and make the quality of the data independently verifiable.
  2. Services can be provided by any layer of government, and by commercial or third sector orgs. It’s OK when they overlap, complement and duplicate.
  3. It is possible to interact with multiple layers of government at once while respecting their organisational and democratic sovereignty.
  4. Build small services that can be loosely joined together however citizens like. Do not try and model the whole world in a single user experience, you will either fail or build a digital Vogon.
  5. Put users in control of their data. Millions of engaged curators are the best protection government has against fraud, and that citizens have against misuse.
  6. A user not having to understand government does not mean obfuscating the workings of the system.
  7. The system should actively educate people about how their democracy works and where power and accountability lie. Put transparency at the point of use.
  8. Be as vigilant against creating concentrations of power as you are in creating efficiency or bad user experiences.
  9. Understand that collecting data to personalise or means-testing a service comes at a cost to a users time and privacy.
  10. Sometimes the user need is ‘because democracy’.

They condense so much wisdom about good service design, accountability, democracy, security and platform thinking.

They were developed while Richard and others were designing a prototype of a natively digital government. I was lucky enough to demo this at Code for America in September (video below):

My sole quibble is with Rule 3:

3. It is possible to interact with multiple layers of government at once while respecting their organisational and democratic sovereignty.

I’d argue there’s often no such thing as “organisational sovereignty”; there is only “democratic sovereignty”.

For UK central government, civil service institutions such as government departments are not independent legal entities. Their existence is entirely in the gift of the democratically appointed Secretary of State, who is free to determine the shape/size/makeup of the institution(s) needed to execute her/his mandate.

Funnily enough, this is not something the UK civil service talks a lot about – not least to the Ministers it serves…

Blockchain vs Democracy (aka ‘software is politics, now’)

The Economist succinctly nails the issue I have with “federated trust” distributed ledgers such as Blockchain / Ethereum:

The idea of making trust a matter of coding, rather than of democratic politics, legitimacy and accountability, is not necessarily an appealing or empowering one.

To be clear, append-only, immutable, verifiable ledgers (distributed or otherwise) are properly exciting as a new form of digital civic infrastructure .

There are some good posts on the GDS blog exploring their potential.

It’s the ‘federated trust’ bit of Blockchain that worries me, more for reasons of democratic accountability than wasted electricity.

Democracy has evolved as the least bad way we know of managing the tensions between collective and individual needs. You mess with it glibly at your peril. Or rather, at our peril. Careful democratic thought required, as well as careful code.

As Richard Pope says, “Software is politics now“.

Update 14 Nov 2015 :  Jeni Tennison, Deputy Director of the Open Data Institute, has written a very fine blog post explaining far better than I can why using blockchain to store personal data is fraught with issues.

Keynote at Web Directions 2015, Sydney

Tom at Web Directions

Run by the wonderfully energetic John Allsopp, Web Directions 2015 is a cross-disciplinary conference for those blessed with living the Southern Hermisphere . It’s been going since 2004, and was held this year in Luna Park, a surreal 30s-era funfair in the shadow of Sydney Harbour bridge.

I was invited to give the closing keynote on digital transformation of public services, entitled “Enough Lipstick on Pigs“. I’ve put the presentation online; video to follow.

Presentation to Code for America Summit 2015

Here’s a link to the presentation I gave to this year’s Code for America Summit.

It explains why I believe a new institutional architecture is required to support a natively digital nation, and what form these new institutions might take.

It’s basically me presenting the ideas of Richard Pope.

The video of the talk will be more meaningful, but people afterwards were nagging me a url of the presentation, so this’ll do for now.

[update] Here’s the video of the talk and demo.

Stuff that matters, done the right way

There are so many committed teams delivering digital services all over government, it feels unnatural to highlight just one – but when it comes to doing stuff that matters, at scale, and under pressure, the Department of Work and Pensions (DWP) team developing the Universal Credit (UC) Digital Service is impossible to ignore.

Here’s a video demonstrating the UC Digital proof of concept developed by DWP and GDS following the 2013 reset of UC:

Not an easy ride

Their mission is to simplify the welfare system, and to help more people to move into work, while supporting the vulnerable.

From the moment the UC Digital Service team was established in summer 2013, they’ve been at the sharpest of sharp ends. The entire UC programme was being reset. The UC Digital Service was given the herculean task of developing the strategic “end state” service, to support the full diversity of people who might claim UC at some point during their lives.

It hasn’t always been an easy ride.

And as the UC Digital Service scales up from a few hundred diverse claimants in South London to cover the whole of the UK, there will doubtless be more bumpy moments ahead. This stuff is hard.

Doing the hard work to make things better

But, when you spend time with this incredible team, it’s impossible not to be impressed with just how much they’re getting right in designing a service that will give this radical policy the very best chance of success.

Let me give you a few examples:

  1. They’re not building a website; they’re creating a new service to deliver new policy. That requires a new organisation, with new ways of working and a new culture, as much as it requires great user journeys or quality code.
  2. They have strong backing from the top of DWP, from Ministers and Permanent Secretary down. The Universal Credit SRO Neil Couling provides exemplary leadership (he’s at every show and tell), chair of the UC Programme Board Sir Robert Walmsley is applying appropriate governance with refreshing alacrity, and the digital support from Kevin Cunnington has been crucial.
  3. They make sure they really understand and meet the needs of UC’s users. All users. Even the most vulnerable. Especially the most vulnerable. And they never forget that DWP frontline staff are vital users too.
  4. They don’t pretend they get everything right first time. They watch how users interact with the service, and then they iterate the whole service to better meet the policy intent. Sometimes they iterate code, sometimes process, sometimes training, sometimes policy – it’s all one service.
  5. They’re not scared to deploy the real-world user insights they’re gaining to challenge policy decisions or existing custom and practice, sometimes long-standing. Doing the hard work to make things simple.
  6. They swiftly and cheaply prototype multiple different ways to meet the policy intent, ditching those that don’t work in practice in favour of continuing to iterate those that do. That’s the future of policymaking.
  7. They’re confident enough to start small, recognising that the value of early work often lies in learning about the reality of users and their needs, and testing the most important assumptions.
  8. They know that the unit of delivery is the team, and that each team must contain the right mix of specialist skills and experience. Those with deep frontline operational experience are highly prized for their often priceless insights.
  9. They scale up carefully and organically. They started with one team, and are now up to six, but have grown at a steady pace so their culture and quality hasn’t been compromised.
  10. They invest the time in hiring high-quality specialist skills, and are happy to seek support from the centre when they need it, be that from GDS, CESG or elsewhere. From its inception, the UC Digital team welcomed the support GDS can offer, a close collaboration which continues to this day.
  11. They make sure all those joining understand the intent behind the policy. Indeed, they constantly remind each other of the point of UC – to simplify welfare, help people gain sustainable employment, support the most vulnerable.
  12. They are agile. They update the service early and often. They are highly disciplined in their agile planning, they prioritise ruthlessly, and their testing and integration is continuous.
  13. But they’re not dogmatic. Where it makes sense, they’ll accommodate other project methodologies – for instance, when integrating with some parts of DWP’s legacy IT estate.
  14. They don’t try to do it all themselves. If they can use an existing tool or capability they will – though they know when not to compromise the coherence and quality of the service. They’re not afraid to do less.
  15. They know that security, like user experience, is the responsibility of the whole team, and requires defence in depth, awareness of emerging threats and agility to respond swiftly.
  16. Openness. They’re open with each other about what is and isn’t working, and honest with stakeholders about what is and isn’t possible. They know the value of such a culture, and fight tooth and nail to protect it and maintain it.

I could easily add more.

The challenges continue

It would be wrong to suggest that the UC Digital Service team is getting everything right – indeed the demonstrations they give during their regular cross-government show-and-tells rarely go unchallenged.

The challenges that lie ahead as they build on and replace the existing ‘live’ UC service remain profound; daunting even. It’s vital the team are given the time and space to iterate towards a mature operational service, able to support a reality that is often messier and more complex than appreciated.

But if anyone can do it, they can.

Follow Tom on Twitter, and don’t forget to sign up for email alerts.

via Government Digital Service » Tom Loosemore