Complexity impedes effective delivery
So said outgoing chair of UK’s Parliament’s Public Accounts Committee Edward Leigh in an “I’m outta here!” letter. Never mind the irony of publishing it as a PDF instead of an HTML page – it’s a good launching off point to think about Big Ticket IT.
Those four words sum up the problem with most big software, and he throws in a laundry list of other issues that can cause software failure, primarily, in my wording: management incompetence, laziness, and overly optimistic estimates (both of schedule and “value” in the finished project).
I have a personal, totally unscientific rule of thumb: the more budget there is in an IT project, the more likely it will cost a lot in both time and money. That’s a bit of Yoggiic tautology there, but when it comes to the new idea of open government, it’s a good seam to cut along when thinking about IT projects. When I hear about a multi-million dollar government projects to do, what seems, like relatively simple information presentation, I often scratch my head wondering where that money goes. Would a Twitter account, WordPress.com blog, or Facebook page do just as good?
Beyond actual technologies, simplicity helps in process and planning: often it’s better to deliver IT incrementally rather than all at once.
Saving open government from going pear-shaped
Of course, then there’s the issue of compliance and accessibility, and then you’re into the weeds. In some corners of IT, you’re discouraged from thinking too much about edge cases (“what if they have a dial-up modem?”), in more (government) complex IT projects it seems like edge cases are the main problem as you’re concerned about 100% of the market – better, that 1% who’ll complain.
Clearly, there are things worth spending money and complexity on. Though I haven’t checked it out personally – not living across the pond – from what I’ve seen of the form-handling in Southwark, success can be had it government IT projects.
That case and others points to at least two non-technological things that open government efforts need to double-down on:
- Understanding the usability needs of the “users,” the “citizens” and having enough respect for those users to give them good looking, but primarily highly functional and efficient processes. This is a difficult task that private industry, healthcare, and everyone typically fails at. So when there’s a form that you can fill out in minutes, without having to ask questions, you notice it, and you like the agency. For the government, you’re trying to keep your citizens from thinking you’re incompetent and looking to reduce your budget as punishment. It’s that simple: whenever the government wastes my time, not only do I want to punish it, that’s the only course of action I have. It’s not like I can choose another government like I could choose another phone company.
- If you’re not delivering cost savings, your success is only partial. This is especially true in American government where most people assume the government is a giant, recursive welfare system for itself. Unlike private industry – except for the military, perhaps – results are not worth paying a lot for, the sentiment seems to go. In theory, with all the fat in big, complex projects, finding cost savings should be easy – a “feature, ” a requirement for the project, in fact…