No need to say it, it goes without saying, it should be obvious to all but,
just in case it isn't obvious to all,
IDA is dead.
IDA, now known as "GOV.UK Verify (RIP)",
is the Cabinet Office Identity Assurance programme.
And it's dead.
"If Verify is the answer, what was the question?"
"If Verify is the answer, what was the question?"
It's April 2016 and some time in the next few days or hours someone has to decide whether to declare that GDS's GOV.UK Verify (RIP) system is live.
Whose decision is it? And what does "live" mean?
"Live" must mean something. We can't have another débâcle like GDS's 25 transformational exemplars, when eight exemplars had gone live 800 days after GDS gave themselves 400 days for the project but, if you added in the nine exemplars that were in public beta, then that meant 17 were live, call it a round 20.
Let's assume that "live" means no longer in beta testing, it means that GOV.UK Verify (RIP) is now in operation, it's being relied on, the relying parties are committed, the safety net's been rolled up and stowed away, GOV.UK verify (RIP) is on its own, it's out there in full view, there's no alternative, alea iacta est.
Relied on by whom?
- Not the Scots. They've got their own identity assurance system for access to public services.
- Not local government. They don't need GDS, who are still working on their model for local government.
- Not the NHS. They've explicitly rejected GOV.UK Verify (RIP).
- Not DWP. They've got enough problems with Universal Credit. And they remember the promise that identity assurance for 21 million claimants would be "fully operational from Spring 2013".
As we repeatedly discover, "on-line access to public services" means on-line access to HMRC.
What are the chances of anyone deciding that HMRC should depend on GOV.UK Verify (RIP)?
GOV.UK Verify (RIP) still can't verify the identity of a company. So it can't be used to collect corporation tax, for example, nor the bulk of PAYE, NI and VAT. For that, HMRC will have to continue to rely on the Government Gateway.
But GOV.UK Verify (RIP) is meant to be able to verify the identity of individuals. What are the chances of anyone deciding that HMRC should depend on GOV.UK Verify (RIP) and switch off the Government Gateway at least for individuals?
The enormity of that decision makes it clear that it won't be taken by DCMS, who have somehow inherited nominal responsibility for the national
GDS is in the Cabinet Office. The permanent secretary, John Manzoni, is also the CEO of the civil service. He starts to look senior enough to be involved in the decision. So does Matt Hancock, the Cabinet Office minister.
But only involved. HMRC also would have to be party to the decision. And the Treasury. Which means not just the permanent secretary at the Treasury but the Chancellor of the Exchequer as well. So now we're into the Cabinet. And the Cabinet Secretary, Sir Jeremy Heywood.
This decision is going to have to be taken by a team of senior ministers and their top officials.
It's an unenviable job.
The decision team might look at the criteria established a year ago in GOV.UK Verify [RIP]: Objectives for live. There, GDS set out six objectives that have to be achieved before GOV.UK Verify (RIP) can go live:
|1||Readiness for services to adopt GOV.UK Verify (RIP)||GDS claim that there are nine public services currently available through GOV.UK Verify (RIP). The true figure might generously be seven. GDS sometimes claim that there will be 50 public services available through GOV.UK Verify (RIP) this month but they also row back from that figure to some smaller, unknown number. Who knows who's ready for what?|
|2||Demographic coverage: 90% for services using GOV.UK Verify (RIP) by April 2016||Relying, as it does, on credit history information, passports and driving licences, GOV.UK Verify (RIP) has had particular trouble registering the very young and the very old. There is no sign that those problems have been solved. According to GDS, the solution lies in using more personal information to identify people. They have never said what additional personal information they have in mind and we have no idea whether people would consent to the intrusion. Who knows what the demographic coverage rate is?|
|3||Success rate: 90%||63%. Allegedly. That's the authentication success rate. Meanwhile, the authentication completion rate languishes at 33%. Neither figure is anywhere near the 90% minimum |
|4||Everyone can use GOV.UK Verify (RIP) to access services||No. Some people refuse to use it. Among the willing, some don't have the IT skills. Among those willing who have the IT skills, they don't all have access to viable broadband. Not everyone has a live-looking credit history and a passport and a driving licence. GDS's assisted digital initiative has never made any headway. There is no hope whatever that everyone can use GOV.UK Verify (RIP) to access either the public services or the private sector services GDS have been vainly lobbying.|
|5||A range of high quality certified companies for people to choose from||Their quality is a matter of opinion. The nine "identity providers" promised fell to eight when PayPal pulled out. Verizon have been hacked and are currently out of action which takes us down to seven. Of those seven "identity providers", only three are certified and no-one's ever heard of two of them, trustworthy though they may be. All the "identity providers" want to collect colossal amounts of personal information and share it with other companies all over the world out of the owners' control. Why would people choose any of them just to check the points on their driving licence?|
|6||The product and service are scaled, resilient and operationally ready for live||Who knows?|
Quite rightly, GDS have never had to take a decision of national importance. The decision team are likely to have very different criteria. But even by GDS's own lights, the decision now can't possibly be yes, GOV.UK Verify (RIP) is ready to go live. Just look at the table above. The decision team have careers and reputations to consider. And there is something called "the national interest", to which they will not be blind.
The decision team might decide to take the traditional Whitehall route and delay the live date. But delay it for how long? A month obviously isn't long enough. Three months? Six months? GOV.UK Verify (RIP) has already been in development for four years and in beta for two. What reason is there to believe that matters will have improved in six months time? How many more "identity providers" will walk away in the meantime? What miracle or magic needs to take place? What needs to change? And if someone knows the answer to that, why hasn't it already been changed?
The decision team might even take the radical option and decide to cancel GOV.UK Verify (RIP) now. Bow to the inevitable. The first cut is the deepest. Stop-loss. TSR2. GDS set out to avoid the creation of a national identity register,
The decision team could take the Government Gateway away from DWP, who have been poor custodians, and give it to someone else to improve, not the Home Office, maybe HMRC. Or they could give the identity assurance job to Scotland. But not Estonia. Or – think back to Taurus and Crest – they could get the Bank of England to sort something out, probably with the banks, who have on-line identity assurance systems and experience coming out of their ears.
There's face-saving to consider, of course. And not just nationally – blatant log-rolling has seen GDS celebrated abroad and credited with the creation of copycat operations in the US, Australia and now Argentina which in turn validates GDS as the hip option for the politician who wants to be seen as transformative and modern. It remains the case that, the sooner the decision team act radically, the less the loss of face.
Hard to believe but the decision has been taken.
We know that because Neil Merrett told us yesterday GOV.UK Verify [RIP] on course for live service switchover this month.
GDS caught up today: "We and the certified companies are now working towards our next milestone - going from beta to live later this month".
Verizon have now limped back into action, registering new victims of GOV.UK Verify (RIP).
GDS say that "having a range of high quality certified companies for people to choose from is one of our objectives for live". They have these objectives, none of them have been met, please see above, but GOV.UK Verify (RIP) is to be inflicted on the public nonetheless.
Talking of all eight "identity providers", GDS are "sure their solutions are secure". Barclays, on the other hand, say "the internet is not completely secure ... we cannot guarantee the security of your data ... we will ... try to prevent unauthorised access". They can't both be right.
GDS will "continue to help GOV.UK Verify [RIP] work for more people". That's not going very well, as we were saying only this morning:
What is the probability today that GOV.UK Verify (RIP) can verify your identity? According to GDS:
- GOV.UK Verify (RIP) tends to exclude individuals with a low income, people outside the managerial and professional classes, the unemployed, the very young, the very old, urbanites, women and Northerners.
- And for everyone else, even theoretically, it's still miles away from the 100% identity verification rate you might, if you're old-fashioned, associate with public provision.
GDS want to go live even though they know that between 20 and 30 percent of low-paid individuals can't register for a GOV.UK Verify (RIP) account. So much for putting the user first, these people will be excluded by default from public services.
"... no reason for GOV.UK Verify [RIP] to be used ..."
The Privacy and Consumer Advisory Group (PCAG) have devised guidelines under nine headings for the privacy aspects of GOV.UK Verify (RIP). The Government Digital Service (GDS) claim to abide by all nine while actually abiding by none of them.
Hat tip someone, the minutes of PCAG's 10 February 2016 meeting have been published. It would be a pleasure to write several thousand words commenting on the matters arising but let's concentrate on item 2.4:
In five years time the Office for National Statistics (ONS) will conduct the 2021 UK census. We have been prepared ever since the 2011 census to see new methods used.
GROUP BUSINESS - ONS Census (Terry Makewell, Jo Neagus - ONS)
Following a presentation on the potential use of GOV.UK Verify [RIP] for the 2021 Census, Group members recommended that there was no reason for GOV.UK Verify [RIP] to be used on this occasion given that the Census is a count of households rather than individuals and that it would be inappropriate to do so. However, the Group further suggested that GOV.UK Verify [RIP] could be used for other surveys run by ONS, but that this would be a separate discussion and the potential role of Verify would depend on the specific surveys under consideration.
Digital methods. 2021's will be a digital-by-default census. The census will be a service resting on one of GDS's data-sharing platforms, underpinned by a canonical population register compiled, surely, by GOV.UK Verify (RIP).
No. The decision has been taken. The ONS will do its census. It won't rely on GOV.UK Verify (RIP). For whatever reason, GOV.UK Verify (RIP) is out of the running.
GDS work tirelessly to improve everyone's lot. They tinker around with the front end of GOV.UK Verify (RIP) like top hairdressers trimming and shaping and layering, always seeking to retain a certain look while everything under their expert hands is actually alive and constantly changing.
There was Todd Anderson, for example, telling us the other day in GOV.UK Verify: [RIP] Technical delivery update, 11 April 2016 that "to improve GOV.UK Verify [RIP] and make it better for end users, since our last update we’ve … added new journeys to the hub to reflect the new features released by the certified companies".
Keen-eyed stylewatchers will have spotted some of the primping done on the registration dialogue for new victims of GOV.UK Verify (RIP). For example this screen ...
... used to ask victims to confirm that they are aged 19 or over. That was before 12 April 2016. Now it's 20. And that makes it "better for end users".
Think about it.
According to the Office for National Statistics (ONS), the breakdown of the UK population by age and by sex in mid-2014 was something like:
Source: Office for National Statistics
Ages above 105 are not included on the population pyramid.
The ONS's data sheet estimated that there were 790,575 19 year-old victims in the UK in 2014 and that there will be 864,872 by 2039. Let's say there are roughly 800,000 of them at the moment.
What Sweeney Todd Anderson and his team have done is to reduce the GOV.UK Verify (RIP) universe by 800,000. Just like that, they've excluded 800,000 people from the possibility of getting on-line accounts which would allow them to transact with government.
GOV.UK Verify (RIP) already had a problem reaching its target 90% coverage. It's languishing at the moment around the 67% mark. GDS have just chopped another 1.2% off GOV.UK Verify (RIP)'s reach.
These are enormous decisions to take. And they're being taken -- in the modern, transformed, agile Whitehall way -- by the hairdressers of GDS.
An ex-maths teacher writes:
A The Government Digital Service (GDS) say that the demographic coverage for services using GOV.UK Verify (RIP) must be 90% if the system is to go live in April 2016.N The ex-maths teacher's contribution isn't very impressive, is it. GOV.UK Verify (RIP) is no use for age verification of the young. We already knew that. Otherwise, five year-olds don't need to view their driving licence details or apply for rural payments, for example, so it doesn't matter if they're excluded.
B At the same time, GDS are trying to dissuade anyone under the age of 20 from registering:
The Office for National Statistics (ONS) estimated that there were 15,259,986 people under the age of 20 in the UK in mid-2014 and that there would be 16,647,588 of them by mid-2039. Those figures represent 23.62% and 22.41% of the ONS's estimated UK population, respectively.
B implies that it is impossible for GOV.UK Verify (RIP) to achieve a coverage of 90% and A implies that it is therefore impossible for the system to go live in April 2016.
Despite professing its virtues, GDS seem to be strangers to data science.