Disruptive Proactivity

How to get to a high quality National Information Infrastructure (NII), starting from here, is an interesting question. As the idea of an NII moves forward, similar questions of how to engage large bureuacracies come into play, in different ways for different organisations.

The Open Data User Group has seemingly run quite hard into the bureuacracy of the various NHS organisations, trying to find out where the direct care facilities are (principally, GP practices, dentists, pharmacies, and hospitals). There’s not really a hard problem, but the NII is a new idea for an existing institution to try and get its head around.

“We have an API, why can’t you use that?” being their entirely legitimate response if they don’t understand the basis for the question being asked. It seems that a mutually comprehensible answer to that question wasn’t found over the summer.

To achieve it’s aims, ODUG requires negotiations a vague understanding of the bureucracy processes, the needs of the NII, and encouraging large slow moving organisations to change. Only two of those three are required, but that adds a requirement for a clearly explained question.

Having HSCIC/NHS England relicense the API, with the consequent change in commitments to the other organisations who prodvided the data, might not be the path of least resistance.

Taking a step back, what is the problem?

Core item: Where are the GPs, Hospitals, Pharmacies (and other bits), and what do they do? When are they open?

Surely, if there was an interface developed between the NHS Choices API, and, say, OpenStreetMap, which shows where the public database was wrong, would be a public service, that most organisations would be happy to correct a public record that is wrong.

Generally, NHS organisations are happy to help the public (or, at least, support the public helping itself). That is the sort of warning that a script can do, looking for divergences

When that exists, and there is then an extract of OSM, under OSM’s terms, that shows a direct benefit to the world, with tacit institutional support, maybe those processes could get formalised to make that dataset, in that format, an officially supported product of the Health and Social Care Information Centre or whichever organisation can take ownership of it.

However, that will only happen if the organisations either understand why doing something is important, or know that there will be pain if they don’t do it. Causing pain to bits of the NHS bureucracy is extremely hard – they possibly have access to rather strong painkillers (which seem to have hallucinogenic side effects).

No single approach

The NII is a novel idea. It is instantly comprehensible to those with a data background who understand systems and data linkage. It is somewhat of a more complex idea for others.

Asking a single question over and over is not likely to be an effective way to a rapid response. Asking different questions in a slightly different way may get an outcome faster.

There are some things that are easy to change, and some things that are difficult. It takes discussion and some level of understanding to work out which is which, and how to get there from here.

Or you can keep asking for everything up front, and see how that goes. It does work, especially when you have no10 behind you, it just doesn’t always work.

But Carl has 10 Rules for Radicals.