Kablenet has picked up a story from a “senior downing street policy adviser” saying that the PM will talk, in the next couple of weeks, about the “Next Steps for Britain“. I’m hoping that some of the stuff that I talk about will make it into those steps, but I haven’t seen the speech. I’m theming my stuff on the “Next Stops for Britain”, we need to stop a few things in their tracks before we can realise our real potential. We need a kick in the pants in the UK on e-government and the best person to deliver that is the PM. From there we can run with it.

The first snow of the year had fallen during the night but by the time I arrived in the centre of Bucharest it was already a grubby mush, swept away by drivers who have the alarming habit of using all of the road rather than just their lanes. The monuments of the Communist regime are easily visible on the way from the airport – huge, austere constructions (the “free press” building that now houses more than 200 independent publishers; the natural history museum, the military museum, countless enormous houses – presumably repurposed now, no longer the homes of the favoured few). It has been a while since I felt the cold drip of melting snow down the back of my neck as the water cascaded from the balconies and cornices high above. The Parliamentary Palace, venue for the conference, is just enormous – the link will give you more data than you ever imagined you’d need. Nothing prepared me for how big it is, but you would certainly not call it beautiful. To think that Ceaucescu constructed it for his home, although he met his end before it was finished. Almost every square inch of floor, stair, banister, wall and column (think of an Egyptian hypostyle and you won’t be far off) is white marble. After the revolution, it was finished off with minimal furnishings, but vast numbers of ornate chandeliers, occasional carpets and beautiful, religious-inspired, murals. Inside, it looks amazing – amazingly awesome and amazingly empty. In a country where the average monthly net wage is a little over $100 it was both inspiring and shocking. A red bull at the airport costs as much as a gin and tonic that in turn costs as much as the taxi ride to the centre of town – about $3.

The conference got off to a late start and the speeches held to just a few minutes long. I noted as I opened mine that long ago I had ready the “one minute manager” but had not expected to give the “Seven minute guide to e-government”. My slides were not available so I adlibbed making just a few of the points that I had planned for my twenty minute talk. It turns out that seven minutes is about enough time to get across the essence of what to do next with e-government. Every few conferences I try and come up with a new theme, to stave off my own boredom and also to keep the regular attendees at these things thinking. The theme this time was “e-government is as much about what you stop doing as what you start doing”.

There were some good points well made by other speakers:

– The Italian Minister talked up the single electronic identity card; a government-issued extension of the existing ID card that will be the base for all e-government transactions. He also talked of reversing the previous trend of centralisation by giving power to the local municipalities, whilst maintaining control over standards in the middle. A risky path that is as likely to result in fragmentation as anything.

– The Russian delegate noted the intention to learn from other countries that were further ahead. He said that 70% of e-government services are not used by the citizens for whom they were developed – I have no idea of the source or accuracy of that statement, but it is easy to believe that some number of that order is correct. Like his colleague from the Ukraine, he worried that there were issues around provision of “last mile” bandwidth (certainly insufficient to deliver multi-media), semantic search capability, auto-translation of content and a lack of regional content.

– The Finnish delegate listed the various national databases (seemingly covering every possible piece of information needed) which are updated regularly (it’s mandatory) and used as the source for all necessary information. I’ve looked at these before and they are certainly a huge advantage in the e-government stakes. The most powerful quote came in this speech – “Don’t push new technology (it usually fails) but do be compatible with your customers”. Those working on digital signatures and certificates (and almost every delegate listed them as a priority), beware – if they don’t work in developed countries, they probably won’t work elsewhere.

– A last minute entry came from Canada where Mark McDowell, custodian of the website “Aboriginal Planet” noted how far Canada had progressed. Take-up figures were impressive – some 75% of tax returns will be filed online next year; 20%+ of Canadians use the Internet as their PRIMARY channel to Government and 40% say that they will over the next year. I noted that all of the website pictures he showed displayed a very standard “common look and feel”, as opposed to a “consistent look and feel”; perhaps they have gone too far with that and need to re-introduce, albeit carefully, a little more sub-branding. Online consultations will be introduced next year.

So, on the theme of “e-government is more about what you stop than what you start”. Here are my 7 principles, one a minute, for the next steps:

– Stop putting online what you already have offline. If the online service is not sufficiently different, no one will use it and you will waste time and money. Leafletsonline.gov.xx is not a strategy.

– Stop replicating within your own government what has already been done. Separate ministries implementing separate systems for key functions such as authentication, payments etc is a waste. If people have to do the same thing too many times to get similar results, they won’t use the services.

– Stop building websites. The world probably does not need anymore and governments certainly do not. If you have one, that is enough. If you have more than one, start a plan to get back to one. Think how much easier it will be to focus on delivering content that is relevant to the citizen if you don’t have to piece it together from hundreds or even thousands of websites. If your people can’t find the service, they won’t use it.

– Stop and think before you say you want to include citizens in the consultation process. If 10 million people suddenly give you their views online, how will you manage? How will you let them know their views are valuable and appreciated? If the feedback is not valued, people will stop giving it.

– Stop designing services from the government view of the world and start designing them from the citizen’s perspective. If people have to think too hard about the structure of government, they will give up, use the ‘phone or go to an office for information. And while you’re at it, design all your websites consistently. That doesn’t mean make them the same, but do make them “consistently different”. That means consistent navigation, consistent styles, consistent places for related links but it also means consistency of tone and voice. It doesn’t mean make them all blue with grey text and yellow borders or anything remotely like that.

– Stop putting online tax first in the list of transactions to go online. Start putting online benefits, pension management … anything that delivers direct value to the citizen. Tax can come later – once you have given them money online they are more likely to let you take it from them online.

– Stop worrying about whether the target is valid and start delivering against it. E-government is primarily about benefit to the citizen. What follows from that is transformation of government, competitive advantage for the economy and probably a better environment all round (less money spent in one place means more spent somewhere else).

And that was that … back to the UK!

I closed the speech I gave today with this slide:

The audience were mostly private sector people – suppliers and vendors – trying to engage with the various offices of government in Northern Ireland. I tried to convey the impression of big opportunities available now to create services that would make a difference to citizens, leveraging infrastructure (such as the Gateway) that was already available. Things looked positive, hopefully we’ll get some interesting responses.

Interactive Bureau, a web design agency, has published a survey of government websites. In 200 pages they’ve skimmed over about a 1000 sites and reviewed in detail 20 key ones. The first thing that struck me as strange is that they don’t seem to have commented that 1,000 sites is about 990 too many. How can the citizen find their way around that many sites without being totally confused? Apart from that, the conclusions raised seem appropriate – sites have information that is difficult to navigate, use complicated language (a whole new meaning of govtalk) and, (for me) most importantly, they note that design is inconsistent. There’s some talk about a task force to address the problem which is too small and being swamped – I’m not sure who they are, but it might be OeE I guess. The Register picked up on the story and re-iterated the point that the Number 10 website comes in for some of the worst criticism.

One story picking up on it came to the conclusion, somewhat oddly, that the entire e-government programme should be abandoned. Rescuing themselves a little later in the piece, they nearly correctly concluded: “It is clear that the Government’s current Guidelines, despite their eminently practical approach to Web site provision, are not being adhered to. Therefore, it may be that the only way to do this is to impose rigorous standards on Webmasters, and to monitor their implementation, based on a Government-wide high level strategy”. ‘Nearly’ because I think it will take a little more than imposing those standards.

I’ve re-worked the slide that I usually use of the Gateway. I’ll be using this in a couple of conferences that I have this week, one in Northern Ireland and one in Romania. Belfast I’ve been to before, but Bucharest is going to be a whole new place for me. I’m looking forward to it.

Anyway, the new slide tries to convey the need for two types of access to government back end systems, (1) for thin clients and (2) for fat clients (aka Accountancy or payroll systems etc). I don’t see anyone else in the world addressing these two needs with any degree of success on a cross-government basis (honourable achievers include the Australian Tax Office and the US IRS – but neither is cross-government). To do that, the theory is that one great piece of central infrastructure (the Gateway) is better than many departmental systems. On top of the Gateway you can initially layer rules engines and a variety of web services (say a payments engine, a secure mail server, a government-citizen text messaging service, an appointment reservation/booking systems etc) – but ultimately the Gateway-type system grows to include those too.

Here’s the picture:

A few days ago I rambled about the relative maturities of any given country’s e-government attempts measured by the number of .gov websites there. I’ve thought this through a bit more and think it might be a fair reflection.

The graph above shows what I think the relationship is. The e-government target is announced – everyone rushes to build websites and, because they are hard, few transactions get added. Websites grow exponentially, transactions arithmetically at best. At some point, someone realises this is dumb and more effort is put into transactions. But only when the effort on websites is actively countered and sites are shutdown is enough effort put into transactions to really make this happen. As the website count gets lower, better and better information is available with each transaction … and e-government success results, i.e. high take-up.

The trick, obviously, is to recognise early that this is going on and take steps to reduce the website count – that’s the bit where good central infrastructure, consistent look and feel, well-researched customer feedback, focused content audits/rationalistion, content tagging (metadata and taxonomy) and RSS-feeds come in. Without those things, I predict that it’s all going to be a bust. Of course, a consultant is an expert who will know tomorrow why the things he predicted yesterday didn’t happen today (borrowed from Laurence J. Peter who, googlism tells me, is the one that first told of the “peter principle”).

Autonomous government agencies often have their own Web design, which can confuse users. I think “does confuse users” would be more accurate. I trailed this same thought in my Computing piece a few weeks ago.

Darrell West has published what I think is his 3rd report on global e-government status, and the full report is here. The most startling conclusion is that, STILL, few websites offer transactional services. More than last year, but still not many … “12 percent of government websites offered services that are fully executable online, up from 8 percent in 2001”. That might mean that 88% of sites are not much use or that they haven’t got round to transactions yet. We can sort that out by killing off the redundant sites of course and by making the sites that remain fully integrated with transactions.

What would be an interesting number is how many sites each country has, perhaps measured per head of population or per million heads (or maybe per million people online) – I would imagine that the number would be lower for those countries both at the top and the bottom of the table (the top countries would have figured out that too many sites was bad and the lower countries are unlikely to be able to invest as much – or, if they haven’t figured it out yet, they’re going to have to soon). Plotting that on a matrix and showing the trend from last year would tell me if I’m right that countries start the e-government process with one site per department, quickly move to dozens (or even hundreds) and then go back the other way when they figure out how to consolidate to bigger, more capable portals. That leads, ultimately, to a single highly interactive, deeply personalised portal that can view across every government system. And that probably tells me that it’s a step or three beyond the web browser. We’re talking something pretty clever to get to that and if we have to figure out how to render pages that it shows in different standards and on different screens and whatnot then it will be too hard to do. So, what’s the next UI?

I was pointed at this article to help me understand open source … not a bad article at all, but it still talks mostly about operating systems. I’m past that – I completely understand why I would run an open source o/s, but I’m stuck on how to build a scalable, performance, high capability system. By the by, I liked the site that this piece came from (Newsforge) and have added it to my regular links on the left.

Not everything that needs to get done needs to be measured. A report on performance based e-government in the USA notes that “Excellent cross-agency coordination is seen in the priority e-Government initiatives, but stove-piped systems and processes remain an obstacle to an integrated e-Government” and “The Federal Enterprise Architecture (FEA) is recognized as the necessary, but missing scaffolding for all agency e-Government initiatives”. I think everyone in government anywhere in the world will identify with those points. But what are people doing about it in government? The silo-based initiatives have been there forever and continue to proliferate. Only if we put strong controls in to prevent decisions being made on IT projects by silo can we hope to address the issue … and couple that with an architecture that allows modules to be developed throughout government by whoever has the expertise and need, but that can be plugged in and made available to all.

The US Feds have announced that they plan to reduce the range of software that they buy. This is a fundamental and necessary step ahead of re-engineering backend systems – you want a stable, known baseline that you can upgrade from. The fewer systems in the baseline, the easier it is to consolidate. You can also start to amalgamate at a hosting level, reducing security and network spends (just how many firewalls does any government need?). Given that the US have a budget of more than $50 billion for IT I would imagine that they’d be looking to make some serious saves here. A small hump in costs for the replacement in year one and two and then dramatic reductions – with the suppliers held by the short and curlies to deliver. This latter point is pretty crucial – a failure in a system that is just one of many just affects that department, but a failure in one that supports 5,10,15 or 20 departments or agencies is a whole different world. I wonder if the suppliers are truly ready to deliver fully mission critical software to support this level of critical infrastructure.