Tom Burton: can open source cure Canberra’s Kremlin web?

By Tom Burton

Tuesday October 7, 2014

Some times small decisions matter. In a single tweet the Australian government’s chief technology officer John Sheridan last month announced a special government version of the Dutch open source web publishing system Drupal had been selected to power the Commonwealth’s websites.

Called govcms, the software will be available to all agencies as an out-of-the-box content management system. It will also be hosted using cloud infrastructure provided through Drupal’s United States service company Acquia. Acquia won the two-year agreement to provide Drupal as software as a service through a public cloud, to the nearly 1000 agencies that are part of the federal government.

Agencies will still be able to choose their own CMS, but Sheridan’s anointment of Drupal as the Commonwealth’s CMS of choice is profound and will have a deep impact right through the entire public sector — federal, state and local — as well as strong ripple effects across the entire local enterprise technology sector.

Drupal is your classic digital disrupter; a robust, free, open source software, backed by the savvy and well-funded Acquia. With the weight of the Commonwealth government behind it, we can expect the small but active ecosystem of local developers to grow quickly as the major development houses move to be able to service the platform.

This decision will also arguably impact the entire Australian digital ecosystem, including the enterprise corporate and not-for-profit sector.

Organisation websites these days are fast becoming the pivot of most enterprise IT stacks. Until now Australia has been a laggard in embracing true open source software with the giant software providers — Microsoft, SAP and Oracle — keeping an iron grip on enterprise systems and the prevailing technology environment.

This has been especially so in the federal government, where fears around security and integration, coupled with antiquated procurement rules, have thwarted efforts to embrace alternative technology solutions, even where cheaper and arguably better exist.

Microsoft, for example, dominates the desktop environment through its Office and Exchange suite, despite the availability of economical cloud solutions such as Google’s Apps service.

In both the US and Britain, Drupal has been strongly embraced by governments, not just because of its initial cost advantage but also because of its innovation benefits. With over 30,000 developers playing on the open source platform it offers lots of opportunities for application development, which can then be shared among agencies.

Acquia is no trivial player — last year it was the fastest-growing private company on Deloitte’s Technology Fast 500 list. Based in New England, Acquia is being ripened for an initial public offering and already has heavyweights such as Amazon as key investors. Acquia was set up by the Dutch founder of Drupal (Dutch for “drop”) Dries Buytaert as a service company to support (and monetise) the free software. Drupal got its big break when some digital evangelists close to the incoming Barack Obama administration chose Drupal to power the site. This endorsed Drupal as a serious player in the government space, an open source solution for a world traditionally beholden to proprietary systems.

A slow slog to better digital engagement

In Australia, the Gov 2.0 movement never really got beyond a promising report back in 2009. But while a number of states such as Victoria and New South Wales are now actively pursuing open government principles, the Commonwealth’s digital engagement efforts have been relatively low key and led by the more adventurous agencies.

An enterprising Sydney development firm called Previous Next saw the opportunity and created a special government version of Drupal with accessibility and security features baked into a distribution known as aGov. The aGov software is what Service NSW is using for the front end of their signal service portal now big rolled out across the state.


And aGov is what will now power the sites of any Commonwealth agencies that take up the CMS software, hosted by Australian-based data centres Acquia has sublet to publicly host the application.

A CMS is the publishing system that powers a website, the engine if you like. At its simplest, content like this is created and then the CMS makes the content available at a location (a URL) such as a home page. The CMS wraps a templated design and navigation around the content and any other features or applications such as a comment box, related content or form.

An aim of a CMS is that it enables non-technical people to publish. But while the front-end user interface is simple it hides a lot of technology grunt, with CMSs using databases and lots of program smarts to dynamically bring all the pieces on a web page together and serve it to the requesting user in split-second timing. In the early internet days these publishing systems were multimillion-dollar investments and made websites a scary project for agencies and corporates alike.

Drupal and other big open source systems, most famously WordPress, have completely changed the web development space. These days it is more than tenable to build large-scale enterprise sites using so-called free publishing software, a lightweight design theme and a set of modules or plugins that add features to a site.

It is this ease of publishing that has given every organisation the ability to engage directly. This in turn has quickly seen the emergence of the corporate newsroom — a place like a traditional media newsroom, where the publishing and engagement activities of an organisation are now carried out.

Sheridan’s initiative now removes any technology excuse agencies had for embracing the new digital world. For agencies looking for a simple, quick public execution, the software will be fine.

Drupal does have a sophisticated way of thinking about content and there is a definite learning curve, so for agencies just wanting a micro-site WordPress is famously easy to work with.

“… with few exceptions the Australian public sector is a wasteland for innovative web design and engagement.”

The technology, however, is not really the big issue. In a world where social media has turned communications on its head, digital engagement has quickly become a very industrial exercise and now goes to the heart of any organisation’s brand and reason for being.

And this is the challenge for government. Websites have quickly become the critical main channel for most government agencies for their stakeholder and citizen engagement, not to mention managing the many citizen/customer contacts and applications that underpin them.

But for over a decade too many government agencies have simply used their websites as a public archive. Whatever information architecture existed in the beginning long collapsed as content has been dumped into corporate websites.

Unlike the commercial world, most of these sites were owned and operated by ICT units, with limited input from communication and marketing specialists. The end result can be seen in the vast number of web 1.0 sites, with their typical left menu structure, vanilla design and overcrowded home pages. Some have a carousel to promote some content, but beyond the homepage there is little hierarchy or obvious discovery paradigm, leaving users struggling to access what they need.

The emergence of social media and the need to engage in a far more sophisticated way has pressed some changes, but with few exceptions the Australian public sector is a wasteland for innovative web design and engagement.

And while government is stuck in a previous era, commercial websites are now far more than just a publishing exercise.

I attended a Sydney conference last week for one of Drupal’s main competitors, Sitecore. Founded in Denmark, it’s a serious mid-tier publishing system. What struck me about this conference was that the publishing side was barely mentioned. It was all about digital marketing, with lots of references to person-to-person or human-to-human engagement.

This change comes from the realisation it is now relatively technically straightforward to service individuals’ personal requirements, on any device at any time. The same technologies that serve contextually relevant ads and enable advertisers to follow users around different sites are now being applied to content and other services.

The problem for government is that it is light years away from this user-centred view of the world. Marketing is largely a foreign concept for the public sector, other than the big delivery agencies and what remaining businesses are still in public hands. While the commercial world is aggressively embracing highly sophisticated digital outreach techniques, government is still getting out of its first generation archive sites.

This is not a CMS issue, and to the extent agencies simply use the free and (relatively) easy rollout of Drupal to put lipstick on the pig, we will continue to see important public institutions struggle — and indeed fail — to engage and influence stakeholders.

How to do it wrong (and right)

Take, for example, Treasury and the Reserve Bank of Australia. Both are critically important agencies — if you like the central platform for the control and management of our economy. Despite recent downsizing, the combined intellectual DNA of both agencies remains impressive, yet the public manifestation of this combined work is woeful.

Research work is routinely published (late in the day) with no supportive promotional campaign or engagement plan. The content is long and over written, locked up in old-school PDFs, with little attempt to break down into richer media formats such as infographics, images or video. The font style and size is tedious and the overall user experience is something the Kremlin would be proud of. Materials are simply posted with little attempt to engage other than the very specialist economist audience that once were the traditional stakeholders of these agencies.


The Treasury and the RBA websites frankly are more suited to a primary school and reflect an approach to modern engagement that comes out of the dark ages. Both agencies are critical to the delivery of important and sobering work that should be framing much of our policymaking for the future.

Agencies absolutely have a choice if they want to be part of the modern communications world and engage their communities with the entire intellectual might and insights they can muster. Yet their websites suggest the task of remaining publicly relevant is proving beyond them.

Witness an excellent recent paper from the RBA, which found the mining boom had not caused the much-anticipated Dutch disease. The Dutch disease is the problem of a boom in exports causing a rising currency and a collapse in trading industries — especially manufacturing. It had been widely assumed this had occurred off the back of the resources boom — but the RBA paper carefully deconstructs this and shows that, in the main, Australian industry has done well out of the 2000s mining boom.

“… the antiquated websites and old-school engagement practices that go with them reveal a serious lack of capability to begin to engage in the modern digital world.”

Such an important finding has profound implications for government policymaking, but the paper was released without fanfare on Friday afternoon and, apart from a couple of specialist media articles, has disappeared without trace.

The whole struggle to sell the budget is another good example of the problem. Selling a budget is not simply pushing the May budget documents. Anyone who has seen a presentation by Treasury secretary Martin Parkinson would not be complacent about Australia’s economic growth and fiscal challenge. This is the message — with supportive materials such as infographics and video — that should be pressed, using every smart campaign play in the digital engagement toolbox.

The simple point is the antiquated websites and old-school engagement practices that go with them reveal a serious lack of capability to begin to engage in the modern digital world. Little wonder these agencies are losing their public clout and having difficulty even framing questions of major import, without being hammered in the contentious world of social media.

Drupal is a good starting point to build some of the agility and user centricity that is required for one-on-one digital engagement, but the risk is agencies continue to see their web channel as a brochureware exercise rather than a serious industrial platform to support all their (increasingly complex and challenging) external relationships.

Critical is the need for agencies to build a platform as a total solution for their activities. Given the poor capability this requires a staged approach. But typically the road map starts with:

  • Strong user research;
  • An enduring and meaningful segmentation of users;
  • A design-led user experience exercise;
  • An adoption of a taxonomy to link all the related content;
  • A marketing outreach platform linked to the content engine; and
  • A set of applications to support these interactions from self-service help to easy-to-use forms.

At one level the technology is the easy piece. A pre-requisite is having an external vision that sees their agencies as relevant, confident players in their ecosystem. This will require a major change in culture at the senior executive service level where managers will need themselves to consciously engage with their stakeholders and citizens in a modern and relevant way.

This is definitely not the case now. Without a serious step-up, the rollout of agile websites will continue to be a tech play, rather than a serious statement about engaging with the new digital world.

About the author
1 Comment
Newest Most Voted
Inline Feedbacks
View all comments
Dan Craddock
7 years ago

A great read, Tom.

I applaud AGIMO’s decision to go open source and standardise. However, notwithstanding their feat of navigating the bureaucratic minefield it would’ve taken to achieve this, I can’t help but feel that announcing a standard CMS for government would’ve felt more groundbreaking in 2009. Much has happened in government digital in the last five years, not the least the establishment and success of the UK Government Digital Service (GDS) and it’s GOV.UK platform.

Despite several envoys to and from Whitehall to learn from GDS, including one by Minister Turnbull, seemingly little osmosis has occurred thus far. Where is our own centralised team of the best and brightest digital/UX people, located in and given authority by the Department of Prime Minister and Cabinet (DPC), in the same way GDS is by the UK Cabinet Office?

Sheridan mentioned in his keynote a forthcoming ‘Digital Service Standard’, but who will enforce it? The UK’s Digital by Default Service Standard is a success because it enforced by the GDS, who have the authority and mandate to guide, assess and approve key online services before they go live. I sincerely hope AGIMO is given sufficient clout if oversight remains with Finance rather than DPC, as I’ve seen no evidence to date that devolving digital governance to siloed departments and agencies, no matter how well-documented and intentioned, actually works.

Where is our own GOV.UK? The code for GOV.UK and the CMS on which it’s built – ‘Whitehall Publisher’ – are both freely available on GitHub. Effectively, GDS has done the heavy lifting for every government in the world, and made freely available not just their code, but their whole playbook. Given the unprecedented level of user research that went into the design of both and their subsequent success, did AGIMO consider adapting either? New Zealand has successfully forked and adapted the GOV.UK code for it’s website, saving huge design costs and time. It beggars belief that in such tight fiscal conditions Australian jurisdictions didn’t consider doing the same.

Despite being a “dedicated centralist”, Sheridan has admitted to me on Twitter that there are no plans at this stage for an Australian GOV.UK because “We have a different model of govt & a different web approach.” Quite what this means, I’m not sure; our system of government isn’t called the Westminster System for nothing.

As for a different web approach, sure, MyGov represents a head-start on the UK in the identity management space, but how does this ‘different web approach’ extend to every department and agency needing to create and maintain its own website? The machinery of government is complex and fluid; this shouldn’t be exposed at the user/citizen interface level, which today is predominately via the web. Anyone who has established a small business lately will have lost count of the number of Commonwealth and state websites they transacted with before being able to operate legally.

The number of Commonwealth government websites last quoted to me was 1,200. At the time of writing, that represents just under one website for every 133 APS staff, and one for every 19,000 Australian citizens. Are there really 1,200 government web teams and/or private local digital agencies good enough to create and maintain the kind of design-led user experience you described in your article? Either way, is this the best use of public money – especially in light of GOV.UK’s scaled success? Even if we got to the point of a consistent design template being used on all Australian Government websites, wouldn’t this again beg the question – why not one website? GOV.UK provides a platform the whole of the UK Government; it allows its digital and communications staff to get on with the job of actually creating and iterating content and transactions, without being side-tracked by endless, wasteful internal debates on branding, design, and accessibility, or wrangling external web consultants of varying quality and motives. What exactly is so different about this former British colony that precludes us from adopting this approach?

Having worked in government for ten years, I accept that none of these initiatives would be easy; GDS never said they were either, but they got there. And remember, that was in a culture immortalised by Yes Minister and The Thick of It! I also accept the Commonwealth may be in the planning stages of some positive initiatives, which aren’t yet public. Let’s hope the promised new e-government strategy due by year’s end will reveal greater progress.