The big idea around G-Cloud is a good one in that it takes a contemporary IT delivery model, designed for the 21st century era of the Internet, and uses it as the basis both to deliver new applications, as well as to modernise government IT. One of the problems with G-Cloud is that it has been presented as a very large, complex and expensive IT project at a time when such projects are seen as a criminal waste of money from the perspective of repaying the national deficit.
The big idea around G-Cloud is a good one in that it takes a contemporary IT delivery model, designed for the 21st century era of the Internet, and uses it as the basis both to deliver new applications, as well as to modernise government IT. One of the problems with G-Cloud is that it has been presented as a very large, complex and expensive IT project at a time when such projects are seen as a criminal waste of money from the perspective of repaying the national deficit. Another problem is that it focuses very specifically on getting the IT infrastructure right first (the pipes and tin) before moving on to implement the concept of the Government Application Store (G-AS). And yet use of cloud delivery by government is not solely dependent on G-Cloud, and actually the main financial business benefits are arguably entwined with the delivery of applications.
The time is nigh for a rethink as Francis Maude looks at ways to reduce the cost of government IT.
So – what is stopping the use of cloud?
It seems unlikely that G-Cloud will materialise in the way that John Suffolk articulated in the dying days of the Labour government, when the idea of investing in large public works to keep the economy going was one idea still on the chancellor's table. Now the situation in terms of G-Cloud is basically this: there is no money to invest and it seems unlikely that the suppliers will donate the infrastructure as a charitable act. On top of this to get buy-in for one large shared infrastructure project quite a number of long term outsourcing contracts would have to be unpicked. So, as far as G-Cloud goes we appear to be stymied. However, it can be argued that G-Cloud was always deliberately missing the point, which is that the government cloud project should have been looking first at business processes and outcomes, not at networks and data centres. Deliberately missing the point because in many ways it is easier to get consensus on infrastructure than on outcomes, especially when the decision-makers have the potential to change at every General Election.
But in terms of real value we should be looking at application areas and how they can be addressed via cloud delivery models. If G-AS becomes the focus for discussion then a lot of the real issues holding back government IT modernisation come to the fore. For a start there are genuine doubts amongst CIOs in government as to the merit of creating a Government App Store, analogous to the iPhone App Store. Given that many of the applications required by government are quite complex and/or specialised the iApp model might not prove the most successful in terms of delivering solutions that are fit for purpose. Actually, what would make more sense is to develop a managed service or platform-based brokerage to deliver business services to government.
This is an approach that K2 has been discussing for some time now: The basic premise of business service brokerage will be to provide pre-integrated SaaS solutions from a variety of ISVs writing to the APIs of, in this instance, the government’s platform, or to an open source platform managed by an IT service provider on behalf of the government. In this way it could become the platform for cloud delivery of shared provision of services.
Procurement policy and Information Assurance Rules are holding public sector IT back
The rudiments of this type of service are available to see in the way that the DWP service catalogue operates and might provide a departure point for G-AS. Alongside this type of initiative though, the government needs to reassess its procurement policies which currently are set up to maximise competition but which also act to minimise innovation by seeking always to create a level playing field for tenders and bids. This makes the purchasing of innovative solutions an extremely challenging task, and means that many government CIOs have reached a point where they are not even interested in having a look at the technically possible because they will never be allowed to buy it unless two or more vendors can deliver the same thing.
Looking more immediately at how cloud computing could assist with government efficiency drives, there are several publicly available cloud offerings that many government bodies could make use of to reduce costs - these include Google Apps, Basecamp project management software and MS BPOS. However, although use of these services could significantly reduce costs, government authorities and agencies are often prohibited from using them because of Government Information Assurance Rules about data protection. It is widely felt that these rules are holding IT departments back in terms of using modern delivery services, because the default option is that all citizen data needs to be extremely secure. Government CIOs are widely agreed that levels of risk are not being assessed correctly, and that the Government risk appetite is not in line with requirements for reduced spend. The accreditation system for data needs to be re-evaluated in the light of the size of the reduced spending on IT that the Government is proposing.
What to do?
The contract renegotiations that Francis Maude is undertaking under the auspices of the Efficiency and Reform Group need to take place alongside medium and long term planning which considers these issues. Asking suppliers to reduce prices and margins is not sustainable in the long term - a more effective approach is for the government to roll up its sleeves and take a long hard look at its procurement processes and Information Assurance Rules.
Comments
Post new comment