This story provoked a rather predictable outburst on how Government spends far too much money, and doesn’t get value for money*. It’s a perspective that’s expressed frequently — most recently for the police.uk’s £300k price-tag.

Often Government does pay too much. We all wonder why BusinessLink costs £36m a year. But this isn’t one of those cases. Reading Room charged £585 for the ICO’s favicon, and most people seem to be objecting to that figure on the basis that it would have taken them 5 minutes to do it, and therefore, that Reading Room must have charged £585 for 5 minutes work.

But, as anyone familiar with the delivery of actual government projects will know, absolutely nothing ever just takes 5 minutes. Not when you have the client, two government organisations and two or three other contractors to deal with. Not when you’re working within a delivery framework that requires multiple levels of approval before anything goes live. Not when government has deliberately taken a “belt & braces” approach to contracting — ironically usually done to aid accountability, but which invariably creates procedural bloat and extra cost.

There’s a huge difference between making a favicon and making one for a big organisation, with lots of people who need to give input and approval. There’s a huge difference between making an informational website that you think is good, and making one for an entire country, maintained by an entire government. There’s a huge difference between your website and your Government’s website: because in any activity, time and cost increase with the number of people who are involved.

The real question here isn’t why this specific favicon cost £585. That’s pretty clear: Reading Room charge £600/day (which is competitive), spent a few minutes making the favicon, and the best part of 7 hours making sure everyone was happy with it. Which, I strongly suspect, is exactly what they were asked to do by their client.

We absolutely need to work on making government more agile and getting better value for money. And it is starting to happen, as dxw’s success over the past three years illustrates. But blind invective won’t accomplish that.

Suppliers need to help government understand the true (small!) nature of the risks that online technologies present. And we need to show government how it’s possible to do things quickly and cheaply on the web.

And I think that practical help and understanding will get us there much more quickly than premature outbursts of uninformed anger.

Edited to add: @threedaymonk has an excellent writeup on this too.

*I have to admit to partaking in that in a small and light-hearted way.

One of the drives in my RAID5 array seems to have comitted hara kiri. Just as I was making my morning tea, its death screams beeped forth from the hall — not very honorable — and on reboot, the RAID BIOS screen throws up a horrid red “Broken array” message.

Since the whole server is a bit shagged anyway, I think I’m going to decomission it and move to mirrored 1.5Tb drives. This will give a bit of extra capacity as well as using less power than my current 4-drives-plus-raid-card configuration. I can bump Ubuntu up from 7.10 to the 9.04 beta as well.

So — with drives ordered and an external USB caddy on the way for backing everything up — I shall find out about setting up drive mirroring in software under Ubuntu. Wish me luck!