That's how Thomas à Beckett died
Ministers make bad technical architects, hashtag just sayin
I just read the State of Digital Government Review and the blueprint. I have thoughts.
They did a consultation and 131 organisations responded and their input seems slight undigested. The review has 5 root causes and the blueprint 5 kickstarters and 5 big challenges. I tried to track root causes to actions but its a bit wooly.
I have questions
Lets examine these 5 root causes systematically. The first is leadership.
Leadership
There is little reward for prioritising an agenda of service digitisation, reliability, or risk mitigation. Organisational leaders are not paid, promoted, or valued for doing so. Digital does not shape and drive the organisational agenda.
This seems contradictory. The blueprint has a plan for ‘a government app’. That means a fundamental technical decision that will shape and drive the organisational agenda has been made.
Choosing an app, over many apps, over a website, over not using the web but drop-in centres or whatever is a technical decision. To get to that decision lots of things have to have been done. These would include things like
a statement of objectives - what are we trying to achieve?
a survey of data in use and data overlaps - what inefficiencies are we trying to remove - and what organisational changes will come. In the context of Whitehall that would involve a schedule of legislative changes.
a mapping of users journeys in the context of data and objectives and an understanding of the citizen cohorts the new system (of whatever technology) is going to serve.
After this (and a whole bunch of other stuff) does the question become what technical choices are being made. And we would expect to see evidence of this work in the blueprint and next steps. But we don’t.
So who made the call for an app? They are the senior technical leader in Whitehall, the only technical person with a pan-optical vision of digital services in government. And why an app? The answer to those questions will reveal a lot about the state of the nation.
But first some context
25 years ago I was on a government working party in Scotland and the civil servants told us that the minister wanted to hang a plaque and our recommendations had to include a wall to hang it on.
Way back when during the Welfare Reform days that led to Universal Credit a date ‘emerged’ when the Welfare Reform programme would deliver a new social security system. Nobody has been able to determine who or where that date came from. It was an undead, an incanted golem, and the political expectations it created nearly destroyed the programme.
“Why doesn’t government have an app?” is the sort of plausible question a Minister might be asked over Sunday lunch. It’s the sort of question the Minister might ask in the private office. It’s how Thomas à Beckett died. It’s how senior politicians end up being technical architects, by mistake.
As a former Research Fellow in Scottish Government I am quite a long way from Cabinet Office, but App chat was heard in the northern reaches last year. This week I did some discrete enquiries among pals more in the know. The new app could be a more focussed thing that has been political upsold to be more than it is - a plaque on a wall. Or it might be a full-on golem and a scape looking for a goat.
Strong opinion, weakly held.
I would be delighted to be proved wrong, HMU.
The real leadership failure is elsewhere
Generally I think GDS is on the right track. The Foundations of the Digital State’s main proposals are for a weak centre with a strong mandate that does co-ordination without communication using standards. GDS is a bit too <we tell you what to do> for me. I won’t go through all the arguments again (just the read damn report already).
But there is a root cause to be addressed. Prestige in the civil service comes from closeness to the Minister - and working on bills is the peak of that. The Cabinet Office Guide to Making Legislation doesn’t mention technology. The Scottish Government’s Parliament and Legislation Unit’s Bill Handbook doesn’t either, the Bill Team training at Scottish government doesn’t touch on it - I know, I took the course.
If you train your senior civil servants that tech doesn’t matter, then when the Minister comes the accidental technical architect they neither can nor will speak truth to power to them (ie say no).
Addressing that basic root cause would be a priority for me, and I have a plan for Scotland.
Converting this to Westminster would be trivial. Y’all should do it.