Facebooktwittergoogle_plusredditpinterestlinkedinmail

There is no doubt that the Public Sector “Cloud First” and “Digital Transformation” strategies are paying off in terms of the government, local authorities and health sectors delivering better and more cost effective services to citizens and it is only fair to point out that the success in those areas has been driven by Government Digital Services (GDS).

But where next with Digital?

We are seeing the emergence of “Platform” style services such as GOV.UK Verify and GOV.UK Notify. How far can GDS take these components of Government, can they achieve Government as a Platform? How long will they take? Does the vendor and supply eco-system have a bigger role to play in the development and delivery of these services.

One thing is clear, the future of Digital Transformation will see a shift from monolithic, custom developed applications to service applications stitched together with APIs and common authentication solutions, ultimately this is an approach that is not dissimilar to the Service Oriented Architectures (SOA) enterprises have sought for many years it has just been updated with the new world of cloud computing, network ubiquity and development frameworks that deliver a maturity that was hard to attain. This is all great news for the development of new applications but what about the legacy, the applications built over the last 30 years at a cost of millions of pounds, what can we do with them? (more on that later).

New applications driving the digital economy.

With more development frameworks available, cloud capabilities such as Platform as a Service (PaaS) and now the rise of Containers, developers have an unprecedented toolkit of technology on which to build a new generation of applications. They can build applications faster, cheaper, of higher quality and easier to deploy and support. This is the engine behind Digital Transformation coupled with a ubiquity of devices that allow us to access systems bringing customers closer to the services they want to consume and service providers closer to the customers they want to engage with. The ability to engage is ever increasing allowing service delivery costs to reduce and service quality to improve, so what is holding us back from doing more?

There are very few areas in Government that could be considered new or “green field” so the problem is that Government has a huge amount of legacy which is compounded across local government and health where there are multiple software solutions for the different functions of service delivery making the ability to integrate common services such as GOV.UK Verify that much harder.

Whilst using all of the new capabilities available to build new software applications to replace the old is admirable it carries with it both a cost a risk that may not necessarily stack-up and because both are often unknown then they get placed in the “too hard to do” pile.

Moving beyond Infrastructure as a Service (IaaS)

The waves of technology adoption have always fascinated me, the pace at which the IT sector moves means that it constantly reinvents itself as new technologies come along and go through their maturity and adoption cycles when we look at the technology behind Digital Transformation we see those same cycles occurring. Looking back to the early 2000s virtualisation made it’s entrance (in the x86 world at least). We saw test and development shift to virtualisation in the early years and as the technology matured it was soon apparent that there were significant advantages to moving production workloads across also. Moving to cloud we have seen a similar pattern with many enterprises shifting test & dev virtual machines on to cloud Infrastructure as a Service (IaaS). The challenge for the enterprise is that this is all infrastructure and whilst it can often be consumed cheaper, is more flexible and faster to deploy to, it is still infrastructure and offers little value in terms of improving the delivery of service to the customer, only improvements to the applications can actually deliver that value.

The grey area now is containers being focused more on the efficient delivery of the application and affording a portability between different environments that has been hard to attain in the past. Technically this is a further extension of the infrastructure so a containerised legacy application is still a legacy application just delivered in a new way and is likely to take little advantage of the environment in which it is deployed.

The use of PaaS is where I am starting to see more significant activity for enterprises with significant investment custom developed applications and large teams of developers. They are finding that in order to achieve their Digital Transformation goals they need to develop new software or new customer facing applications, they need to do it quickly, they need to deliver consistency and high quality which traditionally has been hard to do all at once so they also adopt an Agile delivery approach in attempt to achieve it and they underpin the process with a PaaS technology approach. Enterprises adopting this approach are seeing so much benefit in relation to the delivery of new applications they are now asking the question – “why can’t we move our legacy applications to this new development lifecycle model?”

Transforming the legacy

Moving legacy applications to these new environments is a daunting task, many legacy applications are poorly documented and the developers may have left but there is still significant value in the business logic locked within those applications, talking to enterprises about their Digital Transformation programmes their priority is unlocking the integration points for the application to allow them to build new applications that will deliver a new user experience whilst retaining that proven business logic backend and system of record. Once integration is understood then move to PaaS, Agile and DevOps delivery becomes the next goal and where the target environment is a Public PaaS then securing the application at a code level is also a priority.

The more engagement we have with customers the more we start to see a shift around the next stage of application transformation from a monolithic code base to something that is more aligned with micro services and then on towards being cloud native. The aim of this shift is breaking down those larger applications into smaller services to promote the re-use of business logic and allowing individual business logic components to be supported on a component basis making future change faster.

The work we are doing at Fedr8 enables enterprises with significant numbers of custom developed application to understand, quantify and de-risk the Digital Transformation journey by turning the unknown unknowns of legacy applications into known knowns with quantifiable metrics that allow for informed decisions to be made, in a significantly shorter time frames, for less cost and with more certainty.

Facebooktwittergoogle_plusredditpinterestlinkedinmail