Public IT project hell: let’s make government work for us

Why does it cost $235 million to integrate a few IT systems?

Johannes Ernst contrasts the Yahoo/Facebook deep integration announcement with the US government’s announcement that they will spend $235 million on integrating incompatible healthcare IT systems, and asks some pertinent questions:

I assume we all agree that an environment in which leading-edge companies innovate on their own to the benefit of their customers is better than one in which the government has to spend large amounts of money to drag along kicking and screaming “participants” — as it is so common in health IT. How do we turn US healthcare IT from the latter to the former?

One might equally substitute education, or local councils, or law enforcement. It’s a widely-accepted truth that public IT endeavors suck, and that enforcing data standards across disparate public bodies is like herding confused, angry cats into a very wet bag. It’s also true that commercial web services have been very good at integrating for the good of their customers, often without any money (let alone $235 million) changing hands.

I do think there’s a false distinction that’s been made here: public bodies and government departments tend to be swamped in a sea of bureaucracy that prevents them from moving or changing as nimbly as many commercial companies. (Of course, as companies begin to become institutionalized through age and size, they also become less nimble: take Microsoft and IBM.) Many of these restrictions are necessary for the simple reason that they’re using our money, and some regulation is required to ensure tax funds are being spent wisely and benefit the wider public good. We don’t want people to just walk off with it.

Our tax dollars at play

It’s also widely-accepted that our tax dollars are not spent wisely, and often don’t benefit the wider public good. Public bodies are full of inefficiencies, in part because of the bureaucracy involved. I’ve certainly worked within university environments where entire departments of people could reasonably be described as incompetent, but had integrated themselves so well into the system that they had become a required port of call in the bureaucratic workflow. I’ve also seen fully private companies formed using university money and resources earmarked for public research, and government grants essentially spent on beer and travel. These are the kinds of inefficiencies and sanctioned fraud that must be stamped out.

Public bodies and private companies are different in one major respect: their stakeholders. It is a legal requirement for shareholders in a company to have access to the company returns, board minutes and so on (although a wider cloak of privacy is often necessary). In a public body, the stakeholders are the public, yet we often don’t have access to details like financial statements, minutes and decision-making rationale. In Britain, an attempt to get government departments to work like commercial companies has resulted in a ridiculous system where departments must pay each other and the British taxpayer often doesn’t have a legal right to the information they produce.

The public is the board

Ultimately, in a democracy, the public should be the board of directors. Genuine public oversight hasn’t been possible before, but transparency and accountability are now possible via the Internet. We don’t need political parties and administrations to be our eyes and ears any more; we need them to be our hands, and act on our behalf. We need to be able to see the inner workings of public bodies: not just the numbers, but the actual internals and decisions. With genuine public oversight in a way that ensures the bodies know they’re being watched, and governments obligated to maintain these bodies for direct public benefit in a way that’s responsive to the public, costs should go down. It’s not perfect – and Switzerland has recently shown us the dangers of having frequent public referendums – but given the spending, inefficiency and fraud inherent in the system, we can no longer trust the government to do this on our behalf.






One response to “Public IT project hell: let’s make government work for us”

  1. Ben Ward Avatar
    Ben Ward

    You make some worthwhile points (bureaucracy, inertia), but moving a few Facebook statuses and some photos about isn’t exactly on the same level. Perhaps there’s as much complexity in Yahoo & Facebook, but the green-field profit-driven nature of these projects means they’ll be easier because the underlying systems can be changed without a knock-on effect.

    It also doesn’t help that people reading ITIL books try to implement high security before actually making things work. All that integration you spent ages on? It all vanishes behind a firewall within a few months. “Large pieces, not joined at all”, if you like.

    Transparency is important, stakeholders having visibility is important, but considering that most IT projects already have more stakeholders than you could shake a stake at, having the public repeatedly burst in and ask for an Inland Revenue iphone app would probably not improve things.

    We in the web world tend to underestimate how complex and rigid systems & structures are within large organisations. The web thrives on openness and technologies which can be seen, implemented and replaced in little over a yearly cycle. By contrast, the rigidity and brittle interconnection of enterprise systems and the closed nature of their development world (often due to data integrity & privacy concerns) will always cause this disconnect between the two worlds. Perhaps coaxing them out and working with them would produce better results than insisting they work like us.

Leave a Reply

Your email address will not be published. Required fields are marked *