GDS's visions of the future

Master of ceremonies, Mike Bracken

I must admit, I was a bit surprised to receive an invite to what was billed as the launch of the Government Digital Service – but was, more accurately, the housewarming party for its new offices at Holborn. I consider myself a ‘critical friend’ of the project, but it’s clear that some people focus on the ‘critical’ part. I had visions of one of those American police sting operations, where they tell all the local fugitives they’ve won the Lottery.
Looking back at the tweets afterwards – from people who were there, and those watching from afar – I was surprised at quite how big a deal people were making of it. I observe these matters more closely than most, I admit; but I didn’t hear a lot I hadn’t already heard before. Some of it several years ago.
What was more important, by far, was who said it. And where.
Francis Maude in casual mode

Leading off the sequence of rapid-fire speeches and presentations (slides now on the GDS blog) was Cabinet minister Francis Maude: note the open-neck shirt, the relaxed saloon-bar lean against the side of the podium. This was not your typical Address By The Minister. Citing his pride that this was happening on his watch, Maude made a somewhat unexpected statement: ‘where a service can be delivered digitally, it should be, and only digitally.’ That sounded like a step beyond the notion of ‘digital by default’. Had I taken that down correctly? Yes I had; helpfully he said it again. And again. Fair enough…
The honour of following the Minister fell, perhaps unexpectedly, to Ryan Battles from Directgov. In fact, this was a recurring theme throughout the morning: it felt like every opportunity was taken to credit Directgov, how much it had achieved, how strong its satisfaction ratings had been.
Tom Loosemore

Tom Loosemore’s slot was probably the most eagerly anticipated: a first public sight of screens from the Single Domain ‘beta’ build. He opened with a tribute to Directgov, and said he now appreciated how difficult it was to get things done in government. And via an extended jigsaw metaphor, he demonstrated some of the new site’s key principles – most notably the use of ‘smart answers’ javascript-based forms which asked specific questions, and gave specific answers. (I’ve got a story of my own to tell about such approaches… another time.)

But for me, Chris Chant’s comments may prove the most significant of all. He described how the GDS’s IT had been set up, using the kind of instant-access, low-cost tools you’d expect of a technology startup. Mac laptops, Google Apps, the open-source Libre Office software suite, and no fixed telephony. (OK, so maybe the Mac laptops wouldn’t be low-cost to buy initially; but they’re more developer-friendly, and almost certainly lower-cost to support.)
I think that’s when it all fell into place for me. The day wasn’t about demo’ing the current work-in-progress on the websites. It was about presenting GDS itself as a vision of the future. It’s an office space which looks and feels like no government office I’ve ever been in: and for many, it’ll come as quite a shock to the system. (Not least the ceiling-height photos of Francis and Martha.)
It’s taking a pragmatic, rather than the usual paranoid and overbearing view of IT security; and a modern approach to ‘desktop’ computing. Which of course is the only sensible thing to be doing in this day and age… although that hasn’t been enough to encourage government to do so in the past.
Ian Watmore

Ian Watmore’s comments confirmed this: one of the first things Mike Bracken had asked for upon his appointment was ‘a building’ – and this was it. Perhaps appropriately, Watmore observed, it’s a former church. As might be expected of a Permanent Secretary, his remarks seemed the best-prepared – although, as he freely admitted, the previous night’s football results must have been quite a distraction for an avid Arsenal fan.
Some visionary remarks from Martha Lane Fox, about technology providing a route out of poverty, brought the procession of Big Names to a close. It’s hard to imagine a more illustrious lineup of speakers for such an event; (almost) all of them speaking without notes, and with conviction. These were the people at the highest levels of the department in overall charge of public services, all speaking as converts to the benefits – to the user, to the civil servant, to the taxpayer – of the new tech-led approach. There’s absolutely no questioning the backing for it.
And by their very mode of operation, GDS is setting precedent after precedent, about what is allowed, and can be done in a Civil Service environment. Others can point to it as an example, and ask difficult questions of their own IT and facilities managers. If they can do it, and apparently save something like 82% by doing so – why can’t we? Or more to the point, how the hell can we justify not doing so?
Things are changing.

Ten things Alphagov gets right

Late on Tuesday night, the password protection was lifted from – and the most eagerly anticipated web project ever produced by government, arguably the only eagerly anticipated web project ever produced by government, was finally revealed. And it’s… well, it’s quite a shock to the system. Or rather, ‘The System’?
It’s important to recognise what Alphagov is, and what it isn’t. It is an illustration of how the ‘experts’ think government should present itself online. It is a pre-pre-release product: they aren’t just saying ‘you might find problems’, they’re more or less guaranteeing it. It is not a finished product – in terms of information content, browser compatibility, accessibility, etc etc. It isn’t a live site: much of the content is a snapshot in time. And it’s not a definitive blueprint of how things will be: it’s a challenge to the status quo. Some of it won’t be workable; some of it won’t be palatable. But it’s time to ask some difficult questions.
Rather than pronounce one way or the other, here’s my list of the ten things Alphagov – as a product, and as a project – has got right. (That doesn’t necessarily mean there’s a complementary list of 10 mistakes, by the way.)

  1. The fact that it happened at all.
    Don’t lose sight of the achievement it’s been to get this going in the first place. Those involved haven’t been working for free: my guess is, the project will have cost several hundred thousand quid (update: pricetag of £261k ex VAT confirmed via Twitter), at a time when jobs and services are being cut (although of course, there’s a view to long term savings). It’s been shielded from the Civil Service: more of a skunkworks, probably, than the ‘official’ skunkworks. It’s been staffed by a phalanx of individuals and small operations, working with open source tools and technologies, and hosted ‘in the cloud’. This is not how ‘we’ do things.
  2. Delivery 1, Perfectionism 0.
    The team were brave enough to publicise a go-live date in the mainstream media. And to within a day or so, they made it. Sure, it was rough round the edges, probably rougher than they actually intended. But they were absolutely right to get it out the door, and worry about the fine detail later. That’s the luxury of being an alpha, I suppose: the opportunity to concentrate on what really matters.
  3. It challenges the norm (while it can).
    You know what they say about the ‘first 100 days’? That’s roughly how long Alphagov had – and they’ve used it to good effect. They’ve shown healthy disrespect for ‘the way we do things’, as they should. They’ve pushed boundaries, broken rules, and thought the unthinkable. But that grace period can only last so long: in fact, this public release probably marks the end of it.
  4. Focus on search.
    For many people, Google is the internet. Alphagov recognises this on two levels. One, it presents itself primarily as a search engine – with the sophisticated ‘auto suggest’ function being a particularly welcome addition. Two, it’s very search engine friendly, with very clean HTML markup, and meaningful and keyword-loaded URLs. It’s also nice to see them indexing other government sites in their own search, although the results are frankly a bit patchy.
  5. Tools not text
    Perhaps the greatest leap forward demonstrated by Alphagov is its preference for online interactions, as opposed to text documents. So for example, instead of a maths textbook explanation on calculating holiday pay, you get a web page which asks a couple of questions, and gives the answer. The page listing bank holidays doesn’t just give you a written list of dates – it gives you a link to a .ics file, which can be imported into your calendar (Outlook, Google, iCal, etc).
  6. Location based services
    On similar lines, it’s fantastic to have a geographic lookup function built in. So for example, instead of telling you to contact your local police station, then chucking you at a list of every police station in the country, it points you to the only one you’re actually interested in. (Well, near enough: the data for where I am seems to be a bit off.)
  7. Jettisons the old, embraces the new.
    Alphagov is surely the first government project to revel in its (very strongly-worded) disregard for government browser guidelines. Whereas ‘proper’ projects are effectively obliged to spend time ensuring things look and work OK in Internet Explorer v6, they’ve used that time more profitably – demonstrating how the use of more modern features, such as geolocation, could really be beneficial. How many times, I wonder, have great ideas for on-screen interaction been killed by the Lowest Common Denominator?
  8. Single government view.
    From a user perspective this one’s a no-brainer, but it still remains the most potentially explosive: absorbing each departmental web presence, and putting a common identity across them. They’ve handled this beautifully, albeit rather cynically. The departmental ‘sites’ retain a certain individuality, if only through the use of a defining colour – red for the Treasury, blue for BIS, and so on. And the Ministers, whose vanity could kill the whole idea, get great big pictures. But for most people, these departmental presences simply won’t be there, until you go looking for them. And that’s how it should be… as long as we can trust the team and the technology at the centre, to be responsive to departments’ needs and desires. (Sadly, the ‘alpha’ won’t tell us that.)
  9. Straight talk.
    I love this page: Does my child need a car seat? You get your answer at the very top of the page, in extra-large bold letters. The sentences are short, decisive and jargon-free. And there’s no missing the safety advice at the bottom, with its mock highlighter-pen effect.
  10. Transparency throughout.
    From the very start, Alphagov has been active on Twitter, picking up well over a thousand followers. They’ve given cute little insights into the team’s activity, they’ve answered questions, they’ve generated a bit of excitement. Shortly before launch, they launched a blog (with our help), pro-actively announcing and explaining some of their more radical approaches, and posting in their own names – not to mention direct links to their personal Twitter accounts. They’ve had (more or less) an open-door policy for people inside government wanting to visit, and see what was brewing. And now it’s live, they’re taking feedback via public routes: comments on the blog, Twitter / Facebook responses, and a Get Satisfaction account… and acting on it, too. Truly exemplary.

So what happens next? It’ll be fascinating to watch. The geeks have thrown down their gauntlets. It’s time for the civil servants to consider how their information and services could fit into the new mould. And for the public to compare the Alphagov approach with the established Directgov/departmental model. Which is better? There’s only one way to find out.

Loosemore leads Lane Fox vision work

Confirmation on the Cabinet Office’s blog of something that’s been known within the Whitehall webby world for a little while now: Tom Loosemore (ex BBC, Channel 4, Ofcom, Show Us A Better Way, etc etc) has been ‘asked’ to put together an ‘alpha’ version of what a Single Government Domain website, as proposed by Martha Lane Fox, might look like.
My feeling was that, although Martha’s principle was sound, I feared for its execution. With the ‘right people’ involved, it could be made to happen; with the usual people involved, however, it would almost certainly go the same way as previous attempts.
Tom Watson MP took a very similar view of things. Writing at Labour Uncut last November, he said:

As Martha rightly points out, to achieve the changes required to make engaging with HMG online a simple, pleasurable experience requires a massive change in culture and technical expertise. And Francis [Maude] is also humble enough to know that he’s going to need the flair and talent of Britain’s best web people. He needs the A-team.

… and indeed, one of the names Tom (W) went on to list was Tom Loosemore. It was a suggestion I entirely agreed with: indeed, I’d mentioned Tom as an ideal candidate for the CEO Digital position (although he himself didn’t agree!).
Tom and his team – which also includes FCO’s Jimmy Leach as the designated Editorial Lead, and has called on various ‘usual suspects’ from the gov/web field (including yours truly, briefly thus far) – have been working out of a deserted floor of COI’s Hercules House offices for a couple of months now, starting with a thorough analysis of traffic and search data from various sources, to identify exactly what the public wants from its government.
Subsequently, there’s been quite a lot of activity over at ScraperWiki, showing a combination of political material, consultations and general public information. There’s an alphagov account at Github. And intriguingly, there have been a couple of FOI requests made via What Do They Know, in Tom’s name, to get JobCentrePlus-related information out of DWP.
Some early visuals – rather bold, post web-2.0 you might call them – have been shown to senior Whitehall webbies, but it’s far too early to offer a judgement on them. The Cabinet Office blog includes a pledge that the team ‘will be making public their progress as they go’, and ‘will report when the first iterations are public’: which, I believe, should be in a matter of weeks rather than months. Meanwhile, you’ll probably want to start following @alphagov on Twitter:
[blackbirdpie url=”!/alphagov/status/52762689625268225″]
Whether or not you like the thought of this initiative, or its hush-hush approach (thus far), there’s no escaping the fact that Francis Maude had given his provisional approval to the notion of unification; and to get us all to a definitive ‘yes’ or ‘no’, we need exercises like this to test out what it will/would actually mean. And you’d struggle to put together a team with better experience, skills and insight to do so.
PS Don’t forget, Francis Maude and Ian Watmore are in front of the Public Administration Select Committee tomorrow morning (Wednesday). I suspect this may come up. Follow the action live at from 09:45.
PPS I couldn’t resist a cryptic tweet this morning: ‘Tempted to register – it’s still available, and might come in, you know, handy.’ Amusing to note that it’s since been claimed by Richard Pope (aka memespring).

Interim govt CEO Digital appointed

Although I don’t believe there’s been any kind of formal announcement yet, I’ve had it confirmed from various well-placed sources that Chris Chant, the Cabinet Office’s programme director for cloud computing (salary £125-129.9k), has been named as ‘interim’ CEO for Digital – the all-powerful position recommended by Martha Lane Fox in her review of Directgov:

a new CEO for Digital in the Cabinet Office with absolute authority over the user experience across all government online services (websites and APIs) and the power to direct all government online spending.

I don’t know Chris at all, but it’s hard to imagine a richer CV – HMRC, Government Gateway, Defra, London 2012, and currently, the man tasked with finding massive IT savings in the cloud. Everyone I’ve spoken to is complimentary, using words like decisive and bullish: this is good. But I understand he’s a Spurs fan. So, you know, swings and roundabouts.
It’s unquestionably a technical rather than a comms / editorial appointment… which shouldn’t come as a surprise in the current context of budget cuts and technology-driven opportunities. (And incidentally, I’m told that’s very much in line with Ian Watmore’s perspective on things.) But in that regard, you wonder what distinction there is between this new CEO Digital role, and the government CIO position recently vacated by John Suffolk… not to mention the various other IT management bodies around Whitehall, and within departments. I suppose it depends how he chooses to wield his ‘absolute’ power.
I don’t see much evidence of an online footprint: he was briefly active on Twitter over the summer, but hasn’t posted anything since August.

You say you want a revolution…

When the Lane Fox review proposed ‘a new central team in Cabinet Office in absolute control of the overall user experience across all digital channels, commissioning all government online information from other departments’, we may all have misread it. There’s a must-read comment on Steph Gray’s post reviewing the review by Tom Loosemore, who played in a role in the review. He writes:

The *last* thing that needs to happen is for all online publishing to be centralised into one humungous, inflexible, inefficient central team doing everything from nots to bolts from a bunker somewhere deep in Cabinet Office. The review doesn’t recommend that. Trust me! It does, as you spotted, point towards a model which is closer to the BBC – a federated commissioning approach, where ‘commissioning’ is more akin to the hands-off commissioning of a TV series, rather than micro-commissioning as per a newspaper editor.

With that one contribution, the proposal is cast in a very different light. And maybe it isn’t as crazy as it all sounded this morning. Sighs of relief all round.
I do firmly believe that there are appropriate role models, and appropriate technical solutions, to allow government web publishing to be consolidated satisfactorily – for all concerned. Yes, the BBC is a fine example, but Whitehall already has its own examples. BIS has done it – see a PQ answer, coincidentally published the same day as the review, in which it rightly trumpets the savings made by bringing all its partner organisations on to the same platform, under the same ‘unifying navigation bar’. (And among my own clients, Defra is on its way to doing it.)
It’s easy to imagine a WordPress1 multisite install, with Super Admin rights kept to a central team, a few custom (and customisable) themes, and a carefully selected set of plugins. Child site administrators wouldn’t be permitted to add new code; but would still have considerable freedom within those permitted boundaries. The themes (and indeed, WordPress itself) would enforce certain technological standards and methods, including – you’d imagine – a family bar across the top of each screen, a standard layout, the same font, good and consistent accessibility approaches, etc etc.
So in principle, yes, it’s absolutely possible… and could meet all four Lane Fox objectives. I dearly hope it can. But I share the mild scepticism of Steph Gray’s concluding paragraph: ‘There’s much to like in Martha’s report, and a real opportunity to make things better. If it’s done right.’ The thing is, twice in the past decade, with DotP and then The Club, it’s been tried – and since everyone still isn’t on a shared platform, you have to conclude it was done wrong.
This new vision will be dependent on the formation of a centre of genuine expertise, whether ‘absolute’-ly powerful or not, in the Cabinet Office. The same Cabinet Office who produced the report as a 5.7MB graphic-based PDF file, whose text was neither searchable nor selectable (for which, to give him his dues, Tom Loosemore has apologised – ‘no excuse’). The same Cabinet Office whose own news ‘child’ website failed to even mention the review’s publication (since corrected). So, some way to go, you’d have to say.
[blackbirdpie url=”″]
And I wonder whether it ultimately requires a dramatic culture change of abandoning separate Departmental identities, including separate press operations. Francis Maude sat on the report for a very long time, before eventually publishing it (apparently) unaltered. Did he find it a tough sell to his ministerial colleagues? That would certainly explain why the response has been to form a new ministerial committee.
The Directgov CEO is walking out. So is the government CIO, explicitly named as the one to lead the development of the shared platform. Not to mention the retiring Director of Digital Engagement. So, to whoever finds this landing on their desk… I wish you good luck. And I assure you, there remains a lot of goodwill ‘out here’.
1 I’m sure it would be possible to do likewise on any number of other CMSes, but I doubt any are quite as good as WordPress at providing flexibility to the child site admins.

Lane Fox report published; does Cabinet Office share her revolutionary zeal?

Martha Lane Fox’s review of Directgov has been published this morning – as an 11 page, 5.7MB graphic-based PDF file, making it impossible to search or select text. (Thanks to various colleagues on Twitter for confirming it wasn’t just me.) Its key recommendations, pretty much as anticipated:

  1. Make Directgov the government front end for all departments’ transactional online services to citizens and business, with the teeth to mandate cross government solutions, set standards and force departments to improve citizens’ experience of key transactions.
  2. Make Directgov a wholesaler as well as the retail shop front for government services & content by mandating the development and opening up of Application Programme Interfaces (APIs) to third parties.
  3. Change the model of government online publishing, by putting a new central team in Cabinet Office in absolute control of the overall user experience across all digital channels, commissioning all government online information from other departments.
  4. Appoint a new CEO for Digital in the Cabinet Office with absolute authority over the user experience across all government online services (websites and APIs) and the power to direct all government online spending.

The document is entitled ‘Revolution not evolution’ – but that’s certainly not the tone of the Cabinet Office press release, which describes the proposals in the most anodyne form imaginable. Where Martha talks about recruiting a ‘CEO for Digital’, and giving him/her ‘absolute’ power, the press release talks about an ‘Executive Director’ – note the immediate switch to Civil Service speak – whose job will be to ‘drive change and bring together existing teams working in this area’. And the press release’s line about ‘asking Directgov and Business Link to create a plan of what would be involved to converge the sites into a single domain’ seems two or three steps removed from actually demanding that it happens pronto.
The most provocative proposal in the document is surely the plan to consolidate everything into Directgov:

A new central commissioning team should take responsibility for the overall user experience on the government web estate, and should commission content from departmental experts. This content should then be published to a single Government website with a consistently excellent user experience.
Ultimately, departments should stop publishing to their own websites, and instead produce only content commissioned by this central commissioning team. There is no need for a major migration of content from existing departmental websites, they should simply be archived or mothballed when essential content has been commissioned and included in the new site.

But Francis Maude’s letter in response – quite rightly – takes a very cautious view of the work involved, and its implications… and almost seems to be kicking it into the long grass.

I agree in principle with your proposal that over time Government should move to a single domain based on agile web shared web services. However, as your report makes clear, this will be challenging for Government and I will need to consult colleagues before we make a final decision about how to proceed. To take these and other cross government issues forward, I intend to set up a new Ministerial Working Group on Digital reporting to the Cabinet Economic Affairs Committee.

Notable by its absence from the review is NHS Choices. Martha’s ‘shared service’ vision shows Directgov, Business Link, ‘departmental teams’, a Central Newsroom (CO/No10) and ‘digital engagement teams’ all feeding the Directgov brand/domain – but there’s no reference to the third of the three supersites. I spotted the other day that NHS Choices is being openly reticent about getting involved in the G-Digital project: the one-page overview on the G-Digital site notes that ‘NHS Choices are represented on the G-Digital Project Board and are considering how they can best utilise the project.’
And whilst the Maude response talks about ‘simplifying the governance of Directgov’, there’s no specific reference to the fate of its management board.
That’s my report on the publication itself; I’ll reflect on the proposals later. In the meantime, here’s what Steph Gray thinks… and he’s bang on.

Lane Fox plans for government web revealed

A document published on a non-departmental site appears to have lifted the lid on Martha Lane Fox’s plans for UK government web publishing. The document, published as an unrestricted PDF, is a review of the website of the organisation in question. But given the ongoing Lane Fox review, its author provides a helpfully concise summary of what may lie ahead.

[Lane Fox] is recommending that Directgov should expand in scope to become the government front end for all transactions, with the ability to mandate departments to meet standards they set; she is also recommending the establishment of a central team in the Cabinet Office in charge of commissioning all online government information, led by a CEO for digital to direct all online government spending. There has been no formal response from the Government to her proposals, but it reflects an overall trend for centralisation and standardisation of government online information and services.

A copy of Lane Fox’s letter to Francis Maude, dated 16 October, was attached to the document in question; but, sadly, has not been included in the online copy. It does at least indicate that the plans have already been widely circulated around the Civil Service.
For the record, the PDF in question appears (at the time of writing) in the first few pages of Google search results for ‘martha lane fox directgov review’.

What are you suggesting, Sir Bonar?

There have been some intriguing tweets from the well-connected, albeit fictional, UK data sharing czar, Sir Bonar Neville-Kingdom in the last day or so.
[blackbirdpie url=”″]
[blackbirdpie url=”″]
[blackbirdpie url=”″]
There it is again: the notion of greater rationalisation around Directgov. Hmm..?
In other news: Guido Fawkes is getting his teeth into the number of former Conservative Party (and indeed, Lib Dem) staff now finding themselves with Civil Service jobs. One stand-out name on the list is Rishi Saha, whose appointment was (finally) covered by the Mail a few days ago. Newsnight’s Michael Crick quotes his job title as ‘deputy director of communications in the Cabinet Office (and effectively head of digital communications, in charge of the websites run by the Cabinet Office and Number 10)’ – but he isn’t named in the Cabinet Office’s recent orgchart.

Lane Fox review hints at further rationalisation

Martha Lane Fox’s review of Directgov appears to have taken a slightly wider view than simply how well everyone’s favourite orange website works. Speaking at a conference in Birmingham, Cabinet Office director of digital delivery Graham Walker said:

We’ve been doing a review of Directgov and most of government on the web. We can see that there is a need to massively simplify it, with a lot more rationalisation and to improve the user experience.

Most of government on the web? More rationalisation? I’ve been hearing rumours that Ms LF’s recommendations may include a stronger role for Cabinet Office in departmental / policy publishing, much as it already has (through Directgov) in citizen-facing material. Are we looking at a single super-site for departments?
On a superficial level, that’s a return to the Dark Ages of 1994, when departments used to send floppy disks in the post to Norwich, where someone from CCTA would mark them up into HTML, and FTP them over to the Government Information Service (aka It’s also a return to the expensive and ultimately unsuccessful notion of The Club / DotP, which would have seen all departments running on the same (bespoke) CMS as Directgov and DH.
But a lot has changed in the past year or two – and it’s possible to envisage more modern publishing infrastructures, based on a managed multisite approach. You can look at what we’ve done on WordPress for Defra as an example: a centrally controlled environment, with the root level defining aspects like primary navigation and plugin selection, but with a high degree of flexibility and freedom given to the various child sites.
So yes, I can see why such a move would seem desirable – Directgov has been a success, at a very high price of course, and consistency of presentation and UX wouldn’t be a bad thing. (Indeed, I’ve written here previously, in praise of greater presentational consistency.) And yes, I believe it’s now technically realistic, in a way it never was before.
The precedents give plenty of reason to be pessimistic, though.