Searching For Patterns in Digital Modernization
Most of the patterns we use (knowingly or not) to guide digital modernization work in government come from the world of software development. We are drawn to these patterns because they are widely used and well understood in the software world, and they enable us to think about complex problems in ways that are easier to understand. We also typically think of digital modernization work as primarily work that revolves around software and technology.
In practice, these patterns do not always work well (or as well as we think they should) because a big chunk of the hard work of doing digital modernization in government is organizational, legal, and bureaucratic. Software development patterns don’t really help with these things all that much. The change we seek is simply beyond them.
The Strangler Fig Pattern is a software development pattern that is well known to people in the world of civic tech, and has inspired government digital service teams for many years. It is one of the most widely known pattern for digital service teams in government because it offers one way to migrate from an old legacy system to a new software solution — a very common challenge in government.*
One of the drawbacks of the Strangler Fig Pattern that I have seen is that using it successfully can require lots of time. Finding the seams in legacy application that you can break off and build out on a new technical stack is often very complex (and contentious). Efforts to use this approach sometimes stall after the migration to a new system has begun, but before it is finished. This sometimes has the effect of leaving organizations in the unenviable position of having multiple systems in place underpinning a service and never fully completing the transition away from the older one.**
Another very common pattern used in government digital work is the Facade Pattern, an approach that hides the complexity of a particular service or function behind a friendlier front-end.*** Some common examples of the Facade Pattern in action can be seen when digital teams:
- Create a digitized version of a form that on submission gets converted into a PDF version of a paper form that then goes through the same (or similar) review and approval process as paper forms.
- Use RPA tools and applications to automate existing business process steps — like in procurement processes — speeding them up and (potentially) reducing the errors resulting from human data entry but keeping the same basic steps of the process in place.
- Leverage LLMs to simplify or summarize complex or disparate information without changing the structure, quality, or location of that source data.
All of these uses of the Facade Pattern can result in a better experience for the people inside government or those that depend on government services, and all are worthy of investing time and energy into under the right circumstances. But none of these examples change the fundamental issues that can make service delivery more challenging. They just make changes at the epidermal level of government organizations. Sometimes this is enough, but where do we turn for patterns when we aspire to more fundamental change?
The book, Platformland, by Richard Pope, came across my radar as I started thinking about the inadequacy of commonly used patterns for digital modernization. It’s a book full of really good ideas about what public digital services can be, and it’s organized around a set of patterns that can be used to build what he refers to as the “next generation of public services.” It’s full of really useful ideas and observations that will assist many of the people working in public sector organizations today to lay the groundwork for better digital services.
For those working to modernize the digital infrastructure of government, we need new patterns that can help us as we pursue true digital modernization. This new book offers a compelling and useful set of new patterns we can start to use.
But the work we face is lengthy, complex, and daunting. We need to continue to refine our approaches and fill our toolboxes with more strategies and approaches to improve the quality of government digital services.
What addition will you make to our collective digital modernization toolbox?
* For a more comprehensive picture of software patterns specifically focused on legacy system migration, check out the book Monolith to Microservices, by Sam Newman.
** Some more recent work by the team that pioneered the Strangler Fig pattern for modernizing large legacy systems are now looking at Generative AI as a way to potentially speed up the process.
*** If you think I’m misapplying the facade pattern to these examples, I’d love to have you share your thoughts in a comment.
