Traditionally, the public sector, in common with most large organisations, has relied on commercial off the shelf (COTS) software to run ICT systems and processes. In most instances, this comes from global commercial enterprises such as Microsoft and Oracle. This COTS software uses proprietary code and cannot easily be reused across the public sector – reducing value for money, flexibility and agility. Importantly, this also impacts our opportunity to reduce risks to service delivery.
In 2004, Government formally articulated the policy that it would seek to use Open Source wherever it gave the best value for money in delivering public services. However, there were many barriers to widespread adoption of Open Source. The software and wider IT markets were immature and did not have competitive products that were easy to include in enterprise business solutions. Suppliers of COTS software were often opaque in their dealings with Government regarding supply chain, terms and conditions and a refusal to treat Government as a single entity. This made like for like comparisons with Open Source extremely difficult. In addition, the Government IT profession had limited skills and a risk-averse culture that limited uptake of Open Source and did not challenge suppliers about technology solutions.
In recent years, many public sector organisations have since demonstrated that Open Source can be best for the taxpayer – in web hosting services, NHS infrastructure and as components in critical systems such as directgov and Electronic Vehicle Licensing. The software and wider ICT market have also developed and made Open Source products more competitive and easier to include in enterprise business solutions. Government has addressed some of the internal barriers to Open Source – through the development of the IT Profession to re-establish skills and cultures, the establishment of the CIO Council (leading to more openness and exchange of information) and crucially, the agreement to the Cross Government Enterprise Architecture framework. The techniques and cultures of Open Source have been adopted in other parts of Government business, for example in the London Borough of Camden development of their web content management tool and the work of the Cabinet Office Digital Engagement team.
The Open Source, Open Standards and Reuse Strategy was published in February 2009. It states that Government will actively and fairly consider open source solutions alongside proprietary ones in making procurement decisions. In addition, Government will, wherever possible, avoid becoming locked in to proprietary software. In particular it will take exit, re-bid and rebuild costs into account in procurement decisions and will require those proposing proprietary software to specify how exit would be achieved. The strategy includes an action plan that is a positive programme to ensure an effective level playing field between open source and COTS software. It also includes actions which will ensure Government will use open standards in its procurement specifications and will require solutions to comply with open standards. Government will continue to use only open standards for documentation such as ODF, PDF and OOXML. The G-Cloud will host the G-AS which will hold existing open source code and solutions for reuse across the public sector.
In order to achieve the key outcomes desired by Government, the CIO council have commissioned OGC and Cabinet Office to ensure implementation of the action plan. Using the governance structure in Appendix XX, the Open Source, Open Standards, and Reuse working group will deliver clear and open guidance for ensuring that open source and proprietary products are considered equally and systematically for value for money. By 2011, public bodies will store and share records of their approval and use of Open Source software on the G-Cloud. The Government Applications Stores will hold Open Source solutions that are available for reuse in the public sector and by 2015 public bodies will review existing solutions available before going to market for new solutions.
Sounds like they are making the right noises but will it become a reality? Open source is the bedrock of many fine companies showing that it is the right tool for the job on many levels (e.g. Google, Facebook, Stock exchanges, and many government organisations in continental Europe). uk.gov is lagging behind in its pandering to microsoft and other proprietary products, exacerbated by poor outsourcing companies such as EDS (see the NHS disaster)
Recently the whitehouse went over to the free open-source Drupal for its website/CMS, on the other hand, Peter Mandelson claims £1m a year upkeep for his site: http://dizzythinks.net/2008/11/mandelsons-departmental-website-aint.html
As a public sector worker who uses and relies upon open-source heavily, I know that it is a far more flexible, secure and reliable tool than what the corporate sales guys are pushing (and get a large cut from). Lock-in to proprietary products and protocols (including but not limited to Microsoft products) is a sure fire money pit as you ride the train of repeat upgrades and downtime while fixing security issues.
Ah yes, open source.
Government does not use open source because:
1. They have a hell of a lot of certification standards. If the software is not certified, it cannot be used.
2. There is no desire to demand open-source software because they do not understand what they are going to do with it. It may require customisation for example.
3. They do not choose open source software because they do not understand the solutions it can resolve.
Government IT customer choose solutions, they do not list a set of requirement and let the supplier choose the solution.
If a supplier chooses open source, they are normally precluded – why?
Because the vendors have already been selling like mad to the Government IT customers to sell their brand direct into the customer base.
We are lacking a structured approach to implementing Open Source in this country. Policy is one thing but reducing the risk of Open Source implementation and assessing and selecting the right projects to deliver at an Enterprise level takes an approach that contains baselining, assessement of solutions, a generic implementation plan and the identification of implementation and support capacity. This level of activity can fall outside of the capability of single Local Govt institutions….and if we are serious about open source we need to apply structure to a variety of products making it easy for any publicly funded institutions to pick up an Open Source product, fully assessed and prepared for Enterprise level implementation.
So why not form a national group whose sole aim is to shape Open Source products for widespread use across govt, instead of ‘considering’ Open Source alternatives, as in Holland, each govt institution would need to state exactly why they were not adopting the appropriate ‘national solution. Each national soluton having already had its viability assessed and proven. Not a new Quango but a co-ordinated ‘club’ of Local Govt IT Managers briefed to deliver savings through collective rather than individual activity?
Well, for the amount they are spending on IT (£100b in 12 years?!) one would think they could pay people to write their programs (possibly utilising existing open source code).
If you want a fine example of how to ignore Open Source and waste money by being tied into an existing supplier for a major slice of IT services – who clearly have an inability to deliver in new developments then look no further than the link below. It does require the extra condition of having incumbent ICT management with a poor knowledge of what is achievable using Open Source – and that is maybe where we have to start – by ensuring that key decision makers in Govt IT have adequate knowledge and skills to make these decisions?
http://paulcanning.blogspot.com/2009/09/lessons-from-great-2009-birmingham-city.html
Much as I hate to admit, I think the government Open Source policy announced 12 months ago is pretty good. The issue is that it was announced 12 months ago and absolutely nothing has happened since. In fact the Microsoft MOU with OGC was re-signed at almost the same time.
The policy announced last year had in fact been kicking around for a long time and it was only rushed out after George Osbourne wrote about OSS in The Times. As usual with gesture politics no thought was given to implementation of the policy and that’s why there has been no progress.
There are many suggested ways to get OSS software more widely used in government – from mandating it to incentivising the SIs to use it more but the simple truth is that it requires some will at the top level whatever mechanism you choose.
3/10 – John tries hard but finds some of the concepts difficult to understand.
COTS? What a load of rubbish!
The COTS tag is used by suppliers to dishonestly drive up costs by delivering heavily customised software without the rights to the source required to allow modifications to fix bugs or add new features. So, once the initial contract is signed, the supplier has the government over a barrel – and the public gets shafted.
It’s well known that most of the money made in this type of contract is made from maintenance. This is where COTS = COSTS. Lack of source rights means that competition over service is not possible and competitors (big and small) are locked out of the market.
Open source has many potential benefits. One that is often overlooked is that it allows entrepreneurs and small businesses currently locked out by rules on supplier size to compete on merit for parts of a contract.
But the root cause is the failure of the Government to ensure that the public secures rights to access the source, to fix bugs and to add features to the information systems it buys. So, though moving completely to free and open source software (FOSS) would solve this problem, this isn’t not necessary.
Just ensure that the public has the rights it deserves for it’s vast expenditure. And if the current generation of suppliers want to walk away – let them go. New entrepreneurs untainted by failure will step up to replace them.
The second key is to provide a mechanism to allow public spirited advocates to speak on behalf of alternative innovative information system solutions during the procurement process. Increased use of FOSS should not just be about repeating the failures of the part at reduced cost. The purchasing process is dominated by those with vested interests in ensuring an expensive but inappropriate solution are chosen. High risk, expensive revolutionary approaches are chosen in the knowledge that the public will continue to pay until they work, without even considering whether low cost, low risk evolutionary approaches might deliver more at less cost.
It is particularly ironic that the public in the UK includes some of the world’s forecast computer scientists, world famous information architects and top coders but the state fears to ask their expert advice. The cost and impact of these systems are vast. The process needs to be more like a public inquiry and less like back room horse trading.
Going in the right direction
COTS were all Lockheed licensees and then often implemented on Oracle and Microsoft n.b. now it’s coined as DoI and DoE
Open source is definately the way word, and the UK gov should be encouraging open standards and development rather than using proprietary systems
Open Source. Absolutely.
Spend the money on actually making systems work for the benefit of the general public, instead of letting it go into the financial black hole of expensive proprietory software and the accompanying software management consultancies.
Why this doesn’t happen is a mystery to me – though as there is significant money involved I can make a guess.
(Schools should use and teach about open source software too.)
Open source is not a panacea, though in some instances it can be very useful. If you are going to use open source then some policy for deployment is required, where the skills are going to reside, indemnification in the case of commerical loss and so on. Also useful would be a commitment to use vanilla commercial software. Much additional cost typically comes from customising existing applications, often for little actual benefit.
Also, ‘by 2015 public bodies will review existing solutions available before going to market for new solutions’ is quite appalling in that they don’t do this already, and it appears that accelerating this very unambitious date could bring significant savings.
Open source on the desktop is a disaster. Anyone got any idea why all those big corporates don’t convert their desktops to Linux? Then why should government do the same?
People here who tout OpenOffice have clearly never used the spreadsheet it comes with, which is more than 100x slower than Excel.
And Base is even worse, ugly, slow and crashes.
“Open source on the desktop is a disaster. Anyone got any idea why all those big corporates don’t convert their desktops to Linux? Then why should government do the same?”
IBM do, and they also sell what they use too. Open Source it may be, but it is certainly not free and certainly not cheap.
Open Source is great, but you still need expertise. Any one can look at the source code, but just having it and being able to look at it does not mean that you will be able to fix it on your own.
I can open the bonnet of my car and look at the engine, but when it breaks I’m going to go to the mechanics – the government will be the same.