Tim Almond

From the Daily Mail…

But it is already three years late and over budget, and last month Britain’s leading computer scientists called for an urgent inquiry into the crisis hit scheme, after warning it may not work.

I doubt that. A computer system for something like patient records and appointment bookings is not that difficult. In terms of data model, it’s not too complex. The traffic levels aren’t that significant. The storage requirements aren’t very high.

Think about it. Even though there are 50+ million of us, lots of us don’t visit the doctor or hospital for years. Most people go a few times a year. Very rarely do we actually have surgery done. Now, how often do you look up something on eBay, Amazon, Google or Yahoo? But they manage it.

If you are Google, with billions of pages and gajillions of searches, you have some major technical problems which mean that you have to resort to things like writing your own File System that will optimise performance. For the rest, technical problems are rarely an issue. A little tuning here and there, but it’s all well established practice.

The real problems with software development are before the coding stage. It’s about having users who want it. Analysts who spend time with the users working through how they work, and how it will fit. Breaking down business processes BEFORE you write a line of code.

My experience is that projects that have those things work. Coding problems later are a lot easier to fix than analysis problems, which are usually fixed by ugly code changes (instead of revisiting the analysis stage).

From what I understand, a lot of GPs aren’t too keen on the NHS IT programme. That means it’s in trouble. 

—–

Tags: Software Development //