April 28, 2010 by Jim Morton
After a couple of years of persuasion, intimidation, bribery and tearful pleading we are now able to show how the strategic vision and concepts that we have been pushing are now becoming a reality and are staring to provide benefits.
The decision to use existing and new projects to help build a strategic architecture has been endorsed by our management team and we are piloting the use of a design authority as a method of aligning our project work with the task of defining and building an architecture based on key concepts such as open standards, re-usable services, infrastructure as a service and business process management.
Along side this more formal work the open data project has shown that embracing new approaches can provide new benefits to the authority quickly and cheaply. Within three months, a two-man project team have designed, developed and deployed an application to provide WCC’s new open data initiative with a repository for data as well as a public facing interface to provide access to the information in question. Received wisdom is that such a project, resourced and carried out using more traditional methods would have taken up to nine months and incurred a far more considerable cost.
So how have our strategic concepts helped this project to deliver quickly and cheaply?
- Service oriented infrastructure: The conscious decision to abstract traditionally embedded physical computing elements within our reference architecture opened up two key routes for the project to take. Firstly in terms of raw storage we could take advantage of Amazons EC2 cloud storage product, which will scale with demand and has a lead time for production deployment measured in hours rather than weeks. Secondly the developers chose to use the Ruby on Rails platform-as-a-service provider Heroku as the development and production environment for the open data application. This means we were up and running in hours, have incredibly simple development to production deployment processes, zero licensing costs and relatively inexpensive hosting costs (significantly cheaper than procuring building and supporting a physical and logical server environment).
- Commitment to open standards and practices: Open standards are clearly a must for an open data project, but the non-proprietary nature of the development means that we will be able to provide it as an open source project for other public organisations to make use of. This could contribute to the growing community supporting/promoting open data around the public sector as well as hopefully saving other people the time and trouble of starting such a project from scratch.
- A focus on re-usable web services: The open data project was able to benefit from the previous work done to develop re-usable web services as a basis for our iphone application. Using single sources of the truth to drive multiple channels or applications will demonstrably improve the efficiency and accuracy of our ICT architecture. As an early example the same web service from http://ws.warwickshire.gov.uk/libraries is driving the iphone application, is available via our open data application at http://opendata.warwickshire.gov.uk/datasets/libraries and is already being used by an external developer on a mapping project at: http://www.meanboyfriend.com/overdue_ideas/2010/04/putting-warwickshire-libraries-on-the-map/. What we now want to see are the services made available by the open data project being re-used as widely as possible by both internal systems and external web sites and applications.
The challenge is to translate this way of working to a broader and more business critical set of projects. We have started the work to define strategic fit and take a new approach with a diverse number of projects such as HR-Electronic Records, Disabled Facilities Grants, Children and Young People’s Plan and our Network Comms Database.
As these projects progress we will be publishing our findings and experiences here.