By Tony Collins
A part of the European Commission is investigating a decision by the Met Office to appoint an IBM executive as CIO while he worked at the same time for IBM, the organisation’s main IT supplier.
The investigation was prompted by concerns of campaigner Dave Orr who wrote to the EC about the Met Office’s appointment of an IBM secondee David Young as CIO for two years between 2010 and December 2012.
Now Michel Barnier, the EC Commissioner responsible for internal market and services, says in a letter to Orr’s MEP Sir Graham Watson that the EC’s Directorate-General for Internal Market and Financial Services has been carrying out “an in-depth analysis” of the facts presented by Orr.
As part of this, the EC has written to the UK government seeking clarification on a number of points.
Some of Orr’s concerns arise from the Met Office’s responses – and non-responses – to his freedom of information requests. One of his concerns is of a potentially cosy relationship between the Met Office as a publicly-funded organisation and its principal IT supplier IBM; and he has wanted to know why the job of Met Office CIO was not openly advertised in a competitive recruitment process and whether its appointment of an IBM secondee had the potential for a possible conflict of interest.
Orr said that the secondment had the potential to confer a unique and significant intelligence and relationship advantage for IBM that other supercomputer suppliers could not hope to match. “In my view, that is anti-competitive and may in spirit at least, fail the EU procurement rules,” said Orr.
Barnier said that the existence of a conflict of interest would “depend on a number of factors such as the precise role and responsibilities the position entails, in particular whether it includes formulating and preparing technical specifications or tender documents for future IT contracts that the Met Office may put out to tender”.
It is also relevant, said Barnier, whether the terms and conditions of the secondment “impose any obligations or restrictions on the head of the department to prevent conflicts of interest, both during the secondment and afterwards”. He also wanted to know if internal rules were in place to prevent conflicts of interest in the course of tendering procedures.
The Met Office and ministers said that Young was not involved in procurement decisions relating to existing supercomputer facilities. Norman Lamb, then minister at the Department for Business Innovation and Skills, said last year:
“Any potential conflicts of interest regarding David Young’s appointment were fully considered prior to his appointment and his terms of engagement specifically cover these …
“David Young had no involvement in the procurement process for existing supercomputing facilities, either for IBM or the Met Office, and he will have completed his secondment and left the Met Office prior to the selection of replacement supercomputer facilities.”
A wise decision?
The decision to second an IBM employee to run the 300-strong IT department, which is based at the Met Office’s supercomputer site in Exeter, raises questions that may go beyond the potential for a conflict of interest.
As Young was unable to be involved in some buying decisions and was unable to attend the technology strategy board to avoid any potential for a conflict of interest, did the Met Office restrict itself unnecessarily in hiring a CIO who faced these constraints?
Did the Met Office waste money – and a precious two years – hiring a lifeguard whose terms of employment required him to wear handcuffs?
The secondment of Young came at a difficult time for the Met Office – and some of the main difficulties it faced in 2010 are largely the same today.
Responses to Orr’s FOI requests and a report by the House of Commons’ Science and Technology Committee highlight some of the Met Office’s challenges:
– A need for modernised software that will take advantage of next-generation supercomputers.
– A need for a replacement supercomputer that has twice the power of the existing one which operates close to one petaflop (one thousand million million floating point operations per second).
– Funding a new supercomputer (with optimised software) at a time of cut-backs in government spending.
A Met Office Executive Board paper said that its executives have had “soft” negotiations with various suppliers about next generation supercomputer technology. They spoke to Bull, Cray, Microsoft, NEC and SGI.
“Vendor presentations indicate that performance increases will come from increasing the number of processors and/or adding co processors designed to process arrays of data efficiently, rather than increasing the speed of individual processors,” said the Met Office paper.
The Met Office says that “significant optimisation work will be needed [on the code] and, if this is not completed around 2014, a delay in the launch of the procurement may be unavoidable.” It has been seeking software engineers with experience of Fortran (which was originally developed by IBM) or C, Unix or Linux and Perl.
A House of Commons report in 2012 emphasised the need for new technology at the Met Office. The report of the Science and Technology Committee “Met Office Science” said in February 2012:
“It is of great concern to us that these scientific advances in weather forecasting and the associated public benefits (particularly in regard to severe weather warnings) are ready and waiting but are being held back by insufficient supercomputing capacity. We consider that a step-change in supercomputing capacity is required in the UK.”
MPs acknowledged that “affordability is an issue.”
The Met Office declined to answer Orr’s FOI requests about the cost to the taxpayer of employing Young.
Since Young’s secondment ended in December 2012 the Met Office has hired one of its own employees as CIO. Charles Ewen has worked for the Met Office since 2008. He works with science teams to operate the Met Office’s high performance computing facilities. He is responsible for the development and implementation of the Met Office’s ICT Strategy and for the internal technical teams within the Technology Information Services Directorate.
The Met Office hired Young for the best of reasons: after a succession of internal management changes it wanted a highly professional, stabilising CIO. But did it need a CIO from IBM, its principal IT supplier?
That the Met Office was sheepish about the appointment of an IBM secondee was, perhaps, revealed by its website which, in giving a profile of Young, did not mention – at first – that he was seconded from IBM. After Dave Orr’s FOI requests the Met Office corrected its website omission, making clear that Young was on secondment from IBM.
The Met Office has been in existence nearly 16o years. It was founded by Vice-Admiral Robert FitzRoy in 1854 as the Meteorological Department of the Board of Trade. It is highly regarded internationally. A testament to the quality of its computer models – which are used for daily forecasts – is that its “Unified Model” is licensed in Norway, Australia, South Korea, South Africa, India, New Zealand and the US Air Force.
Scientists say that a three-day forecast today is as accurate as a one-day forecast was 20 years ago. But in the UK the Met Office gets a bad press – not always unjustifiably. There is a perception that the accuracy of forecasting is not improving. Sometimes it seems poor.
The algorithms that form the basis of weather and climate models place huge demands on supercomputing architectures. The models produce exceptionally large volumes of data. Although the Met Office had a new IBM supercomputer in 2008 it soon needed more powerful hardware and modernised software.
So was it a good idea, with all the challenges the Met Office faced in 2010 – including the need to persuade the government of the need to fund new supercomputer facilities – to appoint a CIO for two years who, because he was an IBM secondee, had understandable restrictions on his freedom to do his job, restrictions the Met Office has been reluctant to reveal, despite Dave Orr’s FOI requests?
Hole in the head
The Met Office may regard an EC inquiry into its appointment of an IBM secondee as the last thing it needs now. But accountability should not be left to the occasional scrutiny by a Commons committee – or to Dave Orr’s FOI requests.
Tony – Thank you for writing up this story after a 2-year campaign.
The Met Office needs to guard its weather forecasting reputation and climate change is way too important a topic, to be in any way, influenced by a single supplier of key supercomputer technology – whether you subscribe to human activity impacts on global warming or not.
SCIENCE FIRST, COMMERCIALS SECOND!
There is another aspect about this that has a European dimension: In aerospace, Europe has a strategy that a competitive European aerospace industry is key to European prosperity and Europe does not simply rely on US technologies. Thus we have EADS with Airbus competing directly with Boeing.
I would say that supercomputer technology is as key to Europe’s future prosperity as aerospace, as it has many modelling applications, alongside weather and climate change.
Is simply buying IBM kit (that in turn is heavily US Department of Defence cross-subsidised) a good industrial strategic option for Europe or does Europe need its own Airbus equivalent in supercomputer technology?