Oliver Bell’s weblog

The technology world is one of many buzzwords and phrases, one that you hear a lot at the moment is “Service Oriented Architecture” or “SOA” which sounds like a highly technical methodology for systems design, in reality SOA techniques can provide a very effective way of dealing with organizational complexity and divides as we work towards delivering cross departmental government services over the web.

During the final months of 1995 I found myself involved for the first time in an eGovernment project. The goal of the project was a simple one, we were tasked with providing a single “Smart Form” that would allow an individual to register as being self employed. Delivering this involved working with business processes in three government agencies, two involved in taxation and one involved in employment registration.

We looked at the challenge ahead as technologies and developers and decided that this would not be a complex system to design or build. After all, it was just a single form, digitising a single well defined process…

About two weeks into the systems analysis phase of the project the enormity of the challenge ahead started to become clearer. Simple data like a name or an address was not stored in the same way across the three agencies. The processes that we were concatenating into our single form were all being run with different service level agreements and delivering on each agreement was quite rightly something that was taken very seriously by the owners of each service. Finally it was also clear that the small amounts of addition overhead of work that would be need to run our single service just didn’t belong in any of the three agencies, and in turn didn’t have any manpower or budget in place to run it.

As a technology company the project soon began to look more like a nightmare than an opportunity, of course for the government we were working with these were exactly the types of lessons that they wanted to learn as they planned for a broader set of projects across the wider civil service.

It is interesting to look back on projects like that today and consider what has changed in the world of technology and how we might approach the project differently in today’s world. The simple answer is that these are exactly the types of challenges that information technology and large systems design principals now take in their stride.

The industry today talks a great deal about Service Oriented Architecture (SOA) which is a technical term for breaking systems into their constituent parts, and then publishing them so they can be used elsewhere while having minimal impact on the organization that provides the service.

In a Government context the advantages are clear. In a SOA world it is no longer critical that data needs to be managed in exactly the same way in every department, or that business processes need to be redesigned with the workings of the rest of the government in mind.

Today every government department can look at the services that it offers, be it licence issuance, tax collection or any other process, and then use simple web service technology to enable those services to be used by other departments or external commercial providers seamlessly.

The benefits are easy to see. Services are less complex and less costly to design and provide. With the right planning around national level architecture and data management, cross government services can be built without the need for costly and complex systems integration projects.

In some cases an agency who makes their particular line of business application available as a web service will find it being used in useful new ways by other government departments or by commercial organizations to provide services to citizens and businesses in ways that had not previously been thought about or funded.

My former team in Redmond put a lot of time and effort into looking at the right high level framework for this sort of environment would look like, the resulting work was called the “Connected Government Framework” or CGF for short. Today you will find the basic framework that the team delivered integrated tightly into many of the solutions and service offerings from Microsoft and our partners.

The lessons from this complex fifteen year journey help us deal with some of tougher issues that just about every government is facing today as they put their own plans in place for the delivery of online and electronic services. Not least of which are the still those same issues of shared service level agreements and data harmonization. Service Oriented Architectures and the technologies involved assist us in delivering complex business systems without the need to closely couple organizations or data in ways that may otherwise be less natural.

23/1/08 additional: Government Computer News this morning carries information on the release of Microsoft’s Citizen Service Platform, an announcement that was made at the Government Leader’s Forum in Berlin yesterday. Follow this link to read more;

Microsoft’s Citizen Service Platform incorporates the company’s work with local and regional governments over the past several years, and consists of templates designed to run in Microsoft operating environments for the most commonly deployed e-government services.

Microsoft will offer the initial set of online services to governments for customization and integration into their current environment later this year.

Over the last decade or more I have been involved in a large number of eGovernment projects, some at a national level, some regional and some local government projects.

At the Government Technology Summit in Phuket earlier today we were discussing what to look for in a project to help decide what sort of outcome there will be once it reaches maturity. I have tried to commit a few of those thoughts to this post, some are obvious statements in here (and for that I apologize!) and some that are maybe not so obvious.

As the reader I would ask you to note that at best this these comments are a guide, every project is different and every government is different - the outcome of any project can be affected by the passion of the individuals involved, the desire of citizens or businesses to support the new service or any number of other factors.

So, some things that might be worth while looking out for when you’re assessing chances of success;

• It is never about the technology. I have seen many eGovernment projects have started out as an effort to embrace technology in one form or another without any clear goal to redesign the process that the technology is supporting or a clear policy mandate that supports the end result. Very often these types of project fail. The technology in itself rarely does anything useful if it does not have a really clear business objective to follow. Look for the policy and business agenda.
• It is always about the leadership. Like any process reengineering project, a great eGovernment project has to have a we defined and empowered leader, and that leader generally has to be somebody who is accountable for the business change that the project will bring. Looking back on the many eGovernment projects that I have been involved with success is often driven in association with a committed minister or mayor. The policy mandate has to support the change and the technology that will be necessary for any project to succeed. Look for the leader and listen to what they have to say.
• Policy and Technology in perfect harmony. Technologists and policymakers rarely “connect”. They usually talk two different languages and therefore often their objectives do not align in the way that the project needs. In any project there has to be a component that will ensure that technologists and policy makers talk – before policy is made. Look for the communication plans that bring policy and technology people together in a fruitful manner.
• Knowing the target audience. Knowing exactly who your target users are can be hard for government entities, they generally have to deliver services that reach everybody regardless of disposition. Unfortunately this does not make this point any less important that it would be for any other project. You have to be able to ensure that your end result reaches the constituents that it is designed to assist, and that they can access it on devices that they choose to use. Very often eGovernment projects target the web as a primary delivery channel, in many project it may be better to consider a kiosk, the mobile phones that are in most peoples pockets, digital television services or one of the many other devices that exist in society today. Look for clearly defined and articulate demographics for the target user base along with an analysis of how they want to consume the service.
• Where is the user? All too often technology seems to be the starting point rather than the user (think through some of the current debates that we’re currently having in Asia). Good technology design needs to place the user at the center or it will most likely fail. As with most large projects, focus groups and other ways of gathering feedback from your target user base will be a guiding light as project plans are executed. Look for a clear articulation of the benefit that the user will get from the project once it is complete. Think about how useful your target group of users will find your final project deliverable.
• It isn’t about the Web. eGovernment is rarely about building Web sites or using technology purely for administrative and operational purposes. It’s more often about fundamentally rethinking the way public service delivery happens. For example, would a policymaker commence a new hospital building program if they understood that pervasive healthcare technologies mean that in the near future patients will be able to self-medicate and control many conditions in the comfort of their own home? Look for the pervasive impact that the Internet enabled project is making on policy development and service delivery.
• Measuring success. As with any endeavor, knowing what success looks like makes it much easier to define goals, targets and plans that will get you there. I have seen many eGovernment projects start off as entertaining technology experiments with no specific end goal in mind. Again, these projects rarely seem to get very far! Look for a clear definition of success, make sure it is clear how you will recognize when you get there.
• Embrace the legacy, be ready for tomorrow. Government data centers can at best be described as complex. Over the years business processes are designed and built on whatever the best of breed (or sometimes just the lowest cost) technology of the day is. Any new project needs to be able to be make use of existing components of the data center while providing support for devices and delivery channels that don’t exist yet. In past years when government technology was often designed in a bespoke manner for a single purpose this would have been a near impossible objective, today we can select technologies that are designed to deliver applications over the Internet to huge numbers of people. For government this provides both a clear view into which technologies and standards will be useful today, along with confidence that following the market will provide an evolution path into the future. Look for the plans that adopt technology that is in the market today, then clearly analyses market trends to provide a roadmap around what comes next.
• You’re going to need your people, bring them with you. Very often eGovernment projects involve consolidating service lines to provide seamless services to business and citizens. Naturally this is threatening to some who are measured on delivery of their own service or efficient running of their department. Think about how staff are measured and rewarded, make sure that there is a clear safety net for their careers in a new framework that provides more efficiency, collaboration between departments, and simpler services for citizens and businesses. Look for a framework that supports the people involved and ensures their involvement and commitment to the project.

So, there you have it. This isn’t anywhere near being an exhaustive list, as I said at the start of this post every project is different so the most you can derive from this text is a little helpful guidance. I don’t for a moment believe that I’m an authority on this topic, every project I participate in teaches me something else that eventually needs to get added to this list.

I add that I frequently feel humbled as I watch the complex machinery of government solve process and technology problems on a near unimaginable scale.

If you have additional points that should make it into this list I would love to hear them.

I’m sat here with a huge pile of paperwork preparing for the Government Technology Summit in Phuket later this month. Microsoft is sponsoring an award for technology leadership and the paperwork represents the amazing array of submissions put forwards by governments in Asia and elsewhere in the world. 

Fifteen years ago I was involved in some very early experiments in the United Kingdom to look at smart forms and identity management, it has been astounding to watch technology and policy evolve to deliver the projects that we see today.

Looking at the submissions in front of me I would describe many of them as Phase 3 projects. I’ll explain…

I have watched many eGovernment projects evolve over the years. When it comes to thinking about what comes next for each of these projects I tend to think about their evolution in three phases.

Overall this is probably about as simple as models get, but it does give some clues around how to build a road-map for the delivery of an  online Government service, along with some introspective guidance that relates to the services that you are delivering today.

Phase 1, Presence. Most governments are past this phase these days, but many began here. Early eGovernment projects had the goal of providing basic information about a department of a service through the medium of a web page. Very often these pages would provide a form you could download and print, maybe an address for your local office, or other guidelines around how to work with the department.

Phase 2, Transactions. Today there are a large number of completed and evolving projects in this category. These phase 2 projects very often take an existing business process that has been working for years and set about digitizing it in one way or another, frequently a web form is put online and citizens or business are invited to come and fill it out in this new environment.

In themselves these projects are useful, they make government more accessible and they decrease the amount of paperwork involved in getting something done.

Phase 3, Seamless Services. Ultimately every customer I talk to today about their eGovernment objectives has some element of Seamless Service delivery in their plans. Technologists might describe this as the Web 2.0 goal for government services. To deliver these types of service government has to be ready to think about the business processes that exist today and be ready to redesign them around a model that will deliver a new level of flexibility and openness.

In society today citizens and businesses have access to a wide range of devices, upon which they manage a wide array of data. In the world of Seamless Services those devices play an integral role in engaging with government as part of their everyday purpose, often they will complete government transactions transparently as part of a commercial transaction or another everyday activity. An example might be applying for an electronic visa in your destination country as you purchase the ticket to travel there from an online travel agent.