Overnight he took charge of 12,000 systems and an $80bn budget, and was quickly presented with a stack of PDFs detailing $27bn worth of over-running projects. In the previous 10 years the number of US data centres had grown from 432 to 2,000, with, typically, less than 27% utilisation in computer power.
One of the first things Kundra did was to launch the IT Dashboard with pictures of each CIO and the status of their projects made available online. The IT Dashboard is a website enabling federal agencies, industry, the general public and other stakeholders to view details of federal information technology investments. The purpose of the Dashboard is to provide information on the effectiveness of government IT programs and to support decisions regarding the investment and management of resources. It is now being used by the Administration and Congress to make budget and policy decisions. Kundra believes that in government there is sometimes not enough sense of personal responsibility for failure and so this was a way to bestow some sense of personal control over IT projects.
Some back office systems that were not working were not bailed out – they were killed and replaced. A zero growth policy for data centres was introduced as was a Cloud First policy for all systems - Kundra did not mandate what type of Cloud delivery model was to be used, it could be either private or public, but it had to have the basic attributes of the delivery model.
Lesson101: CIOs must not become Dr No
Kundra has also sought to change the culture within government IT, because he was concerned that most CIOs were heavily focused on IT infrastructure (hence the huge growth in data centres) and were not focusing their anxiety about security in the right places. Kundra is eloquent about the cyber security arms race that all governments are caught up in, in order to combat crime and terrorism. But he does not think the answer to the cyber security arms race lies within most Government IT Departments, because there simply is not the skill level required to combat the threats being faced. He believes it is more logical to entrust security to specialists (including the US National Security Agency) and get on with delivering services to citizens. As he puts it: "CIOs must not become Dr No by raising security issues all the time."
One of Kundra's innovative changes to Government IT has been the externalisation of the data.gov API set to enable commercial providers to develop services for citizens. One of the greatest challenges facing government IT in Kundra's opinion is data privacy. He likens data to "digital oil" which requires safe storage. However, as long as the digital oil is safeguarded, then, he strongly believes, Darwinian pressure (in the form of the market) needs to be applied to delivery of enterprise IT. Data.gov was launched in May of 2009 and started with 47 data sets.
Today there are over 390,000 data sets on Data.gov on every aspect of government operations from health care to education to what's happening across state and local governments. This is part of Kundra's democratisation of data strategy and it is built on his belief that government IT professionals are simply too focused on IT infrastructure to deliver solutions that citizens want.
Citizens largely care very little about data centres, but they do care about the services made available to them. One of the success stories of this strategy was the development of a new student aid form which enables applicants to get straight through to the US Internal Revenue Service for information required to support the application – this integration eliminated the need for 70 questions on the form! The government is now putting up money for competitions for developers and the average American to come in and create applications that the government couldn't even imagine.
One thing he has not yet achieved (because it requires an Act of Congress) is to change government attitudes to the way IT is budgeted: currently the budget cycle takes up to two years to complete – that is the budget itself, not what it is spent on! So, as in the UK, work is afoot to get Government to move to agile project deployment methodologies to get a focus on execution rather than policy development.
The UK's ICT Strategy Implementation Plan has clearly been influenced by Kundra's approach in the US. One of the big differences over here, however, is that the public face of the Plan appears to be Francis Maude, rather than our Government CIO, Joe Harley. Hopefully, Harley will get more of a speaking role soon as the execution of the Plan gets under way.
---
Autor(en)/Author(s): Dr Katy Ring
Quelle/Source: Business Cloud9, 01.11.2011
Bitte besuchen Sie/Please visit:
