A top Whitehall civil servant has sensationally quit in protest at government's failure to take the web seriously. Well, I'm sure someone's going to write it up like that, so it might as well be me.

It hasn't exactly been top secret, but 'Whitehall Webby' blogger Jeremy Gould has now formally announced his departure from the civil service. He's quitting his role as head of internet communication at the Ministry of Justice, to join us on the outside as a government-specialist consultant. The post on his blog is clearly from the heart; and even though it's no surprise to hear how he feels, I must admit I'm still a bit surprised to see it expressed so directly. And it's curious to see someone express precisely the same feelings I had when I went solo myself.

It's the right thing to do - for Jeremy himself, and on balance, believe it or not, for the greater good too. Good people need to be given good things to do; and particularly post-Web Rationalisation, individual departments rarely have enough substantial projects to make it worth keeping such good people permanently on the books. The System needs people like Jeremy (and myself) to circulate around, wherever and whenever we're needed. (A point made by Nick Booth earlier this week... although I'm wondering, don't us 'mercenaries' count too?) And in doing so, we see the ever-bigger picture, gain experience, and become better at what we do. I'm unquestionably better at this than when I started two years ago - and everyone benefits as a result.

But senior civil servants - and, frankly, Ministers - should heed his words. 'Web stuff is still not being taken seriously enough. There’s been a lot of talk over the last four years of how more senior strategic web roles are inevitable, but in that time its been just talk. So there was no next move for me. I've also found my extra-curricular activities being scrutinised and discouraged in a way I hadn’t expected after it being benignly ignored for the first year or so.'

Government is rarely short of good people, of which Mr Gould - a good friend, and a great colleague - was unquestionably one. However, it's never short of obstructive people who refuse to rock the boat, even when that boat is demonstrably about to capsize. Tom Steinberg talks about bringing computer-savvy people back into government; wouldn't it be a good start to stop losing the ones you've already got?

Welcome (back) to the real world, mate. It's much more fun out here. And you can actually get stuff done.

In November 2005, the Transformational Government strategy declared:

For customer information, self-service transactions and campaign support, services will converge on Directgov and Business Link as the primary on-line entry points; service-specific or stand-alone solutions will be phased out.

Er, so what is this site exactly? A stand-alone site with self-service transaction. And to make matters worse, there's no reference anywhere in the leaflet to the main Business Link URL. (Interestingly, several of the other partner organisations include their URLs in their logos, making them actually better represented than Business Link!)

A whois search reveals that the businessadviceday domain was actually registered more than a year after the Transformational strategy was published. There's really little hope for rationalisation when customer-facing activity like this, on behalf of one of the primary beneficiaries of rationalisation, is being handled through stand-alone websites.

PS: Today was actually the first time I've come across a whois which works for gov.uk - hosted by Janet. Well worth bookmarking, although its accuracy is limited. Most of the records for well-established domain names are listed as being 'created' in 2003, which clearly isn't right. But basic contact details are generally included, and should provide a starting point at least.

But the unknown in the equation is always: where to host it? You don't have to look too hard to find ridiculously cheap hosting deals in the marketplace: £30/year will buy you enough disk space, bandwidth and support/monitoring for most modest projects, often including automated installation of WordPress and other 'open source' software. But in government, in the midst of 'web rationalisation', it's inevitably a bit more complicated than that.

So here's my problem. At the moment I'm producing (on average) a new WordPress site every month - that's just me alone. And I've got a steady stream of people wanting to do others. These sites have to be hosted somewhere. The normal consultant thing to do would be to buy some cheap hosting in the marketplace, then apply a massive markup. Government ends up paying over the odds, and we end up with countless disparate WordPress installations. Nobody's happy, except greedy consultants.

But we can nip this in the bud. A central server somewhere, offered free of charge to any departments who want to run a WordPress project. It would only cost a few grand a year; put two sites on the same server, and you're probably already saving money. It's not as if we don't already have centralised hosting deals. And most importantly, you've 'rationalised' from day one. (Well, day two anyway.)

This would make my life easier as a supplier. It makes 'the centre's life easier, cos they know where everything is and can ensure it's properly maintained (security patches etc). It's a single migration strategy, if 'the central solution' ever provides equivalent functionality. In every respect, it works out cheaper overall. Everyone wins.

So here's my plea to the Powers That Be. Stop me before I proliferate again. Make me an offer I can't sensibly refuse. And save us all money and effort, now and later.

The best lesson I can draw from the day's proceedings is this: Just Do It. The day itself was proof. We all arrived with a common purpose, but no specific agenda. The framework was set in advance, and proceeded to fill itself. We all got stuck in, and it just worked.

You've got Steve Dale's example of just getting a Drupal installation into place, within a fortnight, to shock the client into a response. Or the MySociety approach of accepting 'The System' can't or won't deliver, and just getting on with it. Or my own WordPress-based crusade, I suppose. How to decide if Twitter or Seesmic has a role in government? - start using it, and let's see.

Since Saturday, I've heard of one person who's started a blog, and one person who's decided to get to grips with Facebook. Dave's (relatively simple) Pageflakes example has drawn some interest. I wonder how many had ever edited a wiki before signing up for the event? These are all baby steps, but they are the only way people will get the big picture. (Welcome aboard, guys.)

I firmly believe 'the shift' has happened, and government risks being left (even further) behind unless it exposes itself to the new world out there. COI's Transformation / Rationalisation isn't a bad thing in itself: the worst excesses needed to be reined in. But if we can agree what not to do, can we start agreeing what we can or should do?

Let's start small: a Directgov/COI blog, please. Then maybe a WordPress (MU?)-based blogging platform for Civil Service uses (like Microsoft did). A tie-up with Basecamp or London-based Huddle, to encourage lightweight project management methods. But the best idea of the day came (I think) from Graham from DIUS: a parallel version of Directgov in wiki form, allowing external experts to suggest improvements which might improve the 'real' version. Sheer genius. Let's just do it.