Recently in project management Category

TfL spends £100m on temporary helpdesk

| No Comments
| More
Transport for London has spent £100m on a temporary contract that is still running after five and half years and will be clocking up more costs until 2014.

TfL gave a temporary IT contract to Computer Sciences Corporation in December 2006 after terminating an existing deal with Capgemini, claiming the stop-gap would be for just a year. But the temporary arrangement has been refreshed at least three times already.

TfL revealed the full cost of the temporary contract under coalition government orders to publish all contracts above £25m. TfL has like other departments not published its contract. But it did disclose a budget of £100m+ for CSC's "Marathon", the name of the temporary desktop computer contract.

A TfL spokeswoman said £100m was the total TfL had paid CSC under the temporary contract since 2006. The bill was inflated in January with another £21m for what the spokeswoman said was another temporary contract with CSC. This one would last until 2014.

She said TfL had extended its CSC contract again in 2011 while it conducted another IT review. Called "Project Horizon", it was completed in January. But TfL still needed more time to work out how it should organize its contracts. In 2007, TfL said it needed just a few months to work out how to organize its IT contracts. Then it stalled for more time until 2009 when it launched Project Horizon.

The spokeswoman said: "In August 2011, TfL had a contract providing a single point of contact for all IT incidents, the central helpdesk was provided by CSC. Following the completion of project Horizon in January 2012, the decision was made to extend the CSC contract to 2014 in order to provide time for us to complete the sourcing review and undertake and award any resultant procurement."

"Since the structural changes to TfL under Project Horizon, the information management team now operates for the whole organisation. In light of the new structure and emerging commercial and technological supply models we are in the process of reviewing the multi-sourcing arrangements to ensure they fully meet our business needs and provide the best service," she said.

CSC revealed January's temporary contract renewal in April, one week before its new CEO Mike Lawrie made his first presentation to Wall Street after joining the company in March. CSC had not had any good news in the UK since its 2009 cloud contract renewal with Royal Mail.

Liz Benison, CSC UK president, said in a press release: "We are delighted to be extending our relationship with TFL."

The temporary contract was set up by Phil Pavitt. Now chief information officer at HM Revenue & Customs, Pavitt is locked into a £multi-billion contract with Capgemini until 2017.

After Pavitt left TfL for HMRC, the transport body claimed to save money on its IT by bringing services back in-house from CSC. This is known only to have amounted to £6.7m over 2007 and 2009, according to TfL's board minutes.

Other budgeted TfL IT expenditure includes £100m+ with IBM for the congestion charge "enforcement" system, another £100m+ to Siemens for other congestion charge systems, £50-100m to NSL Limited for information systems and taxi licensing services, up to £25m to BT for an Automatic Vehicle Location system, up to £25m for SAP software licences and "maintenance", £25m for SAP managed services and application hosting and £25m to Computacenter for other hardware and software.

The limits of Agile: Emergn CEO on Universal Credit

| No Comments
| More
The Department of Work and Pensions set out in January to achieve the seemingly unachievable: develop a £2bn Universal Credit system in two years. An Agile software development methodology would make it possible, said the DWP. But its Agile credentials are questionable.

Computer Weekly was fortunate then to meet Alex Adamopoulos, CEO of Emergn, the firm of management consultants brought in to sprinkle Agile fairy dust over the DWP, and ask him about this miracle cure.

DWP is learning how to do Agile as it goes along, he tells Computer Weekly.

The learning process - the "transformation" in Agile-speak - will take as long as the project. Adamopoulos insists this is manageable.

"When you are doing transformation - when you are running a change programme at the same time as running a project - it's part of the process," he says.

Doing all this in time for Universal Credit's April and October 2013 deadlines is what Adamopoulos calls a "fair and aggressive time-line". He's confident the job can be done and Agile will do it.

Consider that in respect of the nitty-gritty job DWP is trying to do and it sounds like a high-stakes game of Scrapheap Challenge.

It has just 18 months left to integrate its vast estate of antique computer systems into a modern whole capable of handling £74bn budget and 6m people.

Scrapheap Challenge.pngNo Agile project has been as big, as ambitious and as public. But it's already looking hairy.

The department said in June the large systems integrators that dominate government IT were not up to the job, echoing the government's line on the causes of public computer bodges. It would rely on SMEs to do the Agile work on universal credit.

Empty promises

It then gave Universal Credit to the usual suspects. Accenture, BT, HP and IBM have been confirmed as lead suppliers to the programme.

DWP meanwhile refuses to say precisely which companies have been employed to do what work for what money under what contracts. It won't even confirm the names of those employed.

It will say only that it is using existing contracts to do the work. It has not updated public records with details of those contracts it has awarded.

A DWP spokeswoman said this week it would be improper to discuss those contracts it has signed because there are other contracts it is still negotiating. CW put it to her that this was nonsense. That was four days ago. It took seven days to get that far.

This is 18 months after the Prime Minister made a song and dance about transparency in contracting after the coalition was elected in 2010.

The Agile-in-government brigade also makes a fuss about contracts. Adamopolous makes a fuss about contracts too.

The government supply chain is "broken", he says. Suppliers became more interested in satisfying their contracts than their customers, he says.

Conventional IT contracts are said to stymie Agile projects. Emergn solved the problem by working with Gallen Alliance lawyer Susan Atkinson to design an Agile addendum the DWP tacked on the end of the contracts it has with Universal Credit suppliers. They remain unpublished, as do the old contracts we are told are rubbish. We will have to take them at their word that what has come is better than what went before.

Agile vs. Agile

Another problem with the government supply chain, says Adamopolous, is that these large integrators all have their own methodologies.

"They come to you prepared to sell their solution. When you come to another supplier, they have a different approach. There's no continuity," he says.

The idea is that suppliers use their different methodologies as a competitive instrument, much as software suppliers use proprietary standards to stake a market as their own.

Adamopolous accuses systems integrators of "bad behaviour". They use their time in a customer account as a "land-grab". They're too busy competing with one another when they are supposed to be serving their customer's interests.

This is not the case with Universal Credit, he says. Emergn solved the problem of competing methodologies by imposing its own trademarked methodology on Universal Credit suppliers instead. The Major Projects Authority, a Cabinet Office and Treasury auditing team, commended this, as though it was any better.

Universal Credit Starting Gate Review.pngBut the DWP didn't really know Agile. The MPA said it had been trained to do the "interpretation" of Agile favoured by Accenture, its lead developer.

This was a  "slightly less 'lean' version" of Agile than something or other it did not specify in the report (which we publish exclusively here).

Agile is a 10-year old systems development methodology. It's been combined with similarly adaptable business process disciplines to produce an amorphous methodology for tending an amorphous reality in the wider business world.

How could either the DWP or the MPA declare confidence in it or the Universal Credit project when they can't even pin a tail on it?

Agile-like

Emergn's own version of Agile, by Adamopolous's admission, isn't really Agile.

"It's a combination of lean, Agile, design thinking and systems thinking," he says. "We've taken products from each to develop our own way of working."

This is what Emergn has applied to DWP and its suppliers. And that's important because, he says: "Agile might not be the answer for the problem they are trying to solve."

The Institute for Government gave a similar view at the Agile Business 2011 Conference in London last week. It draughted its own Agile principles after extensive interviews with proponents of lean and systems thinking.

IfG researcher Jerrett Myers held a talk about progress made since the IfG published the influential System Error report he co-authored in March, and which advocated Agile.

The only progress of note came three weeks later when the Cabinet Office adopted the IfG's Agile recommendations in its ICT Strategy, telling all government departments they should get at least one Agile project on the go, and establishing DWP as Agile lead for government.

The biggest lesson had been the identification of a significant challenge: that two-thirds of all government IT is outsourced, under inadequate contracts (and to the same suppliers).

"The message from all of this is that Agile can be very difficult to run in a large organisation and it can take some time," Myers said last week.

The MPA had already concluded in reference to Universal Credit, Agile was "unproven at this scale and within government".

Computer Weekly reported these findings. Adamopolous is very unhappy about it.

"I don't think its unproven," he says. "I think it is proven. It depends how you define scale."

Unproven

But he can't cite any examples of Agile projects as large or as complex as Universal Credit, especially in UK government. He resists citing his own large customers, which include the BA project used repeatedly as the example for Universal Credit.

The Agile community has a lot to prove. It seems sensitive, perhaps nervous about its ability to wrest control of large, complex projects from the hands of ignorant bureaucrats and placing it in the hands of engineers and users in the real world. It must be the way we build 21st Century infrastructure projects: complex, adaptable, devolved, modular. We're almost there. Don't spoil it now.

A lot's riding on Universal Credit. Adamopolous insists the £2bn project is Agile.

"Its very Agile, from the vision statement, to the stakeholder buy-in, from the acceptance from the managers and the teams, and from the suppliers' perspective.

"It is an Agile transformation programme. Universal Credit is as Agile as they come," he says.

That appears to mean it will by its Agile nature adapt to its antagile parameters, which are many: the need, as Adamopolous puts it, to spread the Agile way "downstream" from those core managers and teams who have been "transformed" thus far; its reliance on HMRC's conventionally-run Real-Time Information project; the need to squeeze mammoth requirements into a mouse-run time-scale; to accommodate contracts the DWP already has with those oligopolistic suppliers who were supposed to have become disfavoured in UK.gov.

Collusion

It means these large suppliers, who have been accused casually of anti-competitive collusion, are now being officially encouraged to collaborate.

Adamopolous emphasises the importance of collaboration for this and all Agile projects. He is proud of the extent to which suppliers are collaborating together over Universal Credit. The Agile way involves teamwork.

"We have a collaborative supply base on the ground," he says. "That's a unique selling point. Because you usually have a competitive environment. When you're in an account and you have suppliers on the ground, they compete for more business.

"But what's nice about this particular arrangement is the collaboration of the suppliers to do what's right for the customer. That's a huge point because it's not what you would traditionally expect."

Adamopolous is uncomfortable being asked about the line between collaboration and collusion. He doesn't believe allowing Universal Credit's large suppliers to be matey might cause any problems. The interview becomes tetchy.

We meet the day Steve Jobs dies. Fortune magazine columnist Philip Elmer-DeWitt had told the Radio 4 Today programme how Jobs had stopped living in a hippy commune when one night he slept under the kitchen table and "watched everybody sneak in and steal food from the refrigerator".

Collaboration

Adamopolous says its right to be cynical but customers are demanding their suppliers collaborate more.

"The collaboration isn't cosy," he says. "It's not like we are all hanging out together and acting like one big company.

"There is a spirit of doing what's right for the customer. That is an Agile principle. It's people over process...

Alex Adamopolous - CEO - Emergn.png"...And transparency," he says genuinely, though after a prompt from his public relations assistant.

"The main goal now is to work for the common good," he adds.

But there appears to be no more transparency in Universal Credit than there was under any public or private sector IT project that went before.

Adamopolous refuses to say how much Emergn is getting paid for its work on Universal Credit or to discuss his contract. Emergn's a private company, he says. Private companies don't have to discuss what they do, he says. He won't even say what Emergn's turnover is or how many employees it has.

The common good is as amorphous a concept as Agile. Thus transparency in Universal Credit appears to extend only as far as it is convenient for those handling its £2bn budget. It lacks even as much transparency as would be necessary to encourage competition. It falls well short of the ideal punted by the coalition government and the proponents of Agile with all their matey thumbs-up talk of collaboration.

It is indeed unclear how, without transparency, collaboration over £2bn of public money could become anything other than collusion.

The Agile way is commendable on the face of it. Its emphasis on devolved decision-making and team work are intuitively better than disempowered working methods designed for manufacturing production lines of an earlier age.

Industry is learning that complex tools like software can solve complex problems in more intimate harmony with the outside world only if they are operated by workers empowered to think for themselves.

Unemployed.pngBut as ever with systems built to support high-stakes political gambits like the coalition government's benefits reform programme, the fear of reputational failure is greater than the risk of project failure.

Politics too has yet to get Agile. Power is not as adaptable as it would have its tools be. The concomitant lack of transparency has been characteristic of past IT bodges. There are yet only vague reasons why it won't happen again.

How FiRe Control burned £494m hole in public finances

| No Comments
| More
MPs published yet another account of bodge and splurge in government IT today. It was a familiar story. But one that for once had lessons for the coalition government.

The topic of today's autopsy was the stinking carcass left from the last government's IT-led reform of Fire and Rescue Services, called FiRe Control. Begun in 2004 and cancelled in 2010, it burned through half a billion pounds of public money.

The project was tabled in a frenzy over the "War on Terror". The emergency services had to be got ship-shape, and quick. So the Department for Communities and Local Government called in military contractor EADS. They then proceeded without proper planning to waste six years and £494m getting nowhere.

The idea was the encourage localism among the 46 Fire Services. DCLG did this by forcing them to ditch their existing control rooms and replace them with nine, pan-local centres. It then tried to force a new computer system on them. It said they had to change their working processes to fit the new system, which it aimed to take 'off-the-shelf' from EADS to save money and time. So much for localism.

If this sounds similar to plans being pursued by under the coalition's banner of localism that's because it is. And that one's under DCLG's remit too.

The Fire Services were of course excluded from all this important work. Their input would be required only when it came to paying the bill for post-implementation costs.

In all the rush, DCLG and Treasury approved the budget before they had determined whether the project was feasible. They sped past warnings from Gateway reviewers. They rushed on without properly planning what they were doing.

They also lacked the staff to do the job. So did EADS. So EADS subcontracted the work. DCLG notched up a bill of £69m for flash-Larry consultants. It paid £42m alone to PA Consulting. And the project was still so poorly managed the Public Accounts Committee said today it was among the worst it has ever seen.

The contract was so troublesome it took three years to draft. When it was agreed in 2007, the control centres were three months from completion. It was still "weak" and "poorly designed".

EADS went off on its own trajectory that had nothing to do with the contract anyway. The document had no means of keeping them in check. DCLG got through five senior responsible owners and four Project Directors - the people who are supposed to keep things under control.

DCLG kept insisting the project should continue. Local fire services meanwhile opposed the plans. They didn't like the reforms being imposed on them. So much for localism.

The department has put up another £84.8m to clear up the mess. The Fire Service has nine empty, regional control centres without an IT system to run them and staff still follow the same working processes they did in 2004.

The PAC said people should be held accountable. But it sounded like bluster. It didn't name the nine civil servants responsible, nor the lead consultants, nor the lawyers who drafted the atrocious contract, nor the companies to which EADS subcontracted the actual work. It didn't even name EADS or its Cassidian IT subsidiary. Nor did it actually account for the lost £494m.

The PAC had no shortage of colourful adjectives to describe just how poor this New Labour IT project was. It's recommendations were familiar: consult users, manage things properly, hold management accountable. We've heard it all before.

Computacenter row threatens blockhead end for open source in Bristol

| 1 Comment
| More
IT giant Computacenter has raised the prospect of legal action against a small open source supplier for complaining to Parliament about its "Microsoft bias".

The veiled threat followed a letter sent to MPs on the Public Administration Select Committee (PASC), in which Mark Taylor, chief executive officer of Sirius, described a project involving Computacenter he said exemplified the the government's declamation of large IT suppliers: that they dominated the market and crowded out competiton.

Computacenter stated on learning of the letter it was seeking "advice" because it was "potentially libellous". It now says making such a statement did not in itself constitute a threat.

That depends on your perspective. Mark Taylor, chief executive of Sirius, which with a £2.3m turnover is about one-thousandth Computacenter's size, was scared stiff when he heard about this. Yet if Computacenter did sue him for being a whistle-blower, it may prove his point.

The Parliamentary committee to which Taylor sent his letter is exploring allegations that companies like Computacenter abuse their market power by punishing small suppliers when they don't keep in line.

The big IT suppliers are so powerful, it is said, they control the vast majority of public sector IT, which is done in their commercial interest. That may count as quid pro quo from supplier's perspective. But when that supplier is part of an oligopoly servicing a monopoly, it is a little unfortunate for anything trying to breathe outside that comfortable little world that has become known as government IT.

This must be most upsetting from Computacenter's point of view too. It has a certain Blue-chip reputation it must present to its customers. Anyone trying to daub the façade with graffiti must be firmly stamped on.

Computacenter has declined to say what offence it has taken with Taylor's letter. It is seriously perturbed, however, at being identified as one of the boo-hissstems integrators that are said to treat SMEs so poorly.

The multitude of winds that have whipped up the current bout of government IT reviews and inquiries (in the name of SMEs, agile, procurement reform, open software, open standards and open markets) threaten an existential crisis for the large suppliers.

Microsoft bias?

Wintel laptop.jpegTake for example the question of whether Computacenter has a Microsoft bias. It's like asking if the Pope is a Catholic.

Computacenter built its business selling "Wintel" computers and infrastructure. When it made 30 employees millionaires overnight on the occasion of its 1998 flotation (link: sic), it was thanks to the Wintel computing boom. It has branched out a little over the years: it now does services too: though mostly on farms of Wintel computers for big corporations. And while it has its customers' interests at heart, it is essentially like a great, huffing bull with Microsoft branded on its balls.

Computacenter's boom years were the late 90s when the corporate mantra was economies of scale. Customers got the economies of its scale by buying thousands of Microsoft computers. Now the market has gone sour and the typical contract has become so large that only Microsoft suppliers can satisfy them because only they are large enough.

Sirius discovered this after writing Bristol City Council's ICT strategy this time last year. Bristol was down with its open source angle and adopted its recommendations. Sirius then couldn't pitch for the work because it wasn't on the official procurement lists. It went in with Computacenter but was shown the door, said Taylor's letter to MPs, when he protested over the Microsoft bias CC had put in Bristol's proof-of-concept open source pilot.

It would be a terrible irony if the project failed because Bristol had no choice but to ask a Microsoft reseller to demonstrate how an alternative to a proprietary infrastructure might be feasible. It would also be shameful for Bristol, a City that that has made the Zeitgeist its identity: multicultural, ecological, collaborative, egalitarian, open. That's what people from Bristol say about their own City. So if not Bristol then where?

If Computacenter's pilot finds against open source, more than seven years of work at Bristol would have been for nothing. Bristol's entire infrastructure would be based on infantilising proprietary technology and its vision of being a Digital City regenerated by a small army of creatives up to their arms in collaborative computer code would be somewhat obscured.

To understand how awful that would be, imagine Bristol's vision being one of municipal authority as bountiful Big Society fount, its computer systems built open and spread like nourishing tributaries throughout the City. Or imagine, conversely, great multinational corporations sat atop the globe like gluttonous octopuses, their proprietary software systems sucking the life and inspiration from the computing generation.

You might detect a little bias there. It's merely one point of view, though one Computacenter is unlikely to plaster over mail outs funded by its next injection of Microsoft marketing development money.

Corporate image

That doesn't mean Computacenter can't get with the programme. But Microsoft and its reseller Computacenter, the model of the noughties corporation, represent the antithesis of the Bristol zeitgeist: monocultural, rapacious, tight-lipped, dog-eat-dog, proprietary.

Granted, Computacenter looks fairly enlightened when viewed from within the retarded world of the City of London. In its last financial results, Computacenter chairman Greg Lock proudly declaimed how the corporation had achieved enlightenment: it had adopted the UK Corpote Governance Code, "not simply because we must do so, but rather because it is the right thing to do."

And bravo. Don't worry that the Code's key principles read like the listing for the soundtrack of Thatcher's Britain: "Leadership, Effectiveness, Accountability, Remuneration, Relations with Shareholders".

If a corporation is an organism and the board of directors its brains, the Code is comparable to one of the first great periods of human enlightenment, when prehistoric man first started organising into paternalistic clans governed by power tempered with honour.

It is progress a little in advance of that earlier enlightenment when pre-prehistoric man learned to adulate the heavens. Ug. Heavens: glorious. Me: glorious. You: cower, worm, before my competitive mastery. Though its morality is essentially tribal.

This may explain why Taylor this week accepted a post on an SME board being established by the Cabinet Office to help them tackle the problems they are too scared to air in public. There might be safety under the shadow of reforming minister Francis Maude.

Since Taylor was one of the key players behind both the coalition government's and Bristol's ICT strategies, you would imagine him wily enough to deal with a lumbering corporation like Computacenter on his own. That is assuming Computacenter's Microsoft bias has indeed scuppered Bristol's open source pilot. Both Bristol and Computacenter say the game's not over. Taylor says Computacenter elbowed him out and submitted assessments that had been fixed to favour Microsoft.

After slogging away for 30 years by the rule of Mammon, Computacenter may have just lumbered innocently into the midst of a political thorn bush. It might now look up and see how much is riding on Bristol's pilot. It's failure will mark the failure not just of Bristol's ICT strategy but of Maude's and Taylors.

Computacenter would do well to back up and consider while licking its wounds how the idea of software freedom has taken hold in mainstream politics. If it does anything else it will end up looking like the ignorant box shifter it has long tried so hard not to be. Why not give Bristol what it wants?

Universal Credit possible if politicians don't interfere, says IT chief

| No Comments
| More
Benefits sign.pngDWP can manage the massive reorganisation of computer systems demanded by Universal Credit as long as politicians don't move the goalposts and over-complicate matters for departmental techies, said a senior departmental techie.

The department has meanwhile concluded after an audit of the shanty town of benefits systems on which it has to build Universal Credit, some of which are genuine antiques, that it must scrap some of them because it would be impossible to adapt them in the time given.

The project, which involves merging six different benefits systems into one in five years, has also thrust DWP into a close partnership with HMRC that will involve building a unified system from components consolidated across their two computing infrastructures.

Steve Riley, IT director at Job Centre Plus, told Computer Weekly the two departments were already working to consolidate their systems into core components to be incorporated into a unified Universal Credit system, under the eye of programme director Terry Moran, former chief executive of pensions.

But an ongoing review of DWP systems was determining whether the strategy would indeed deliver UC and another two major policy reforms the coalition government had requested be implemented simultaneously: replacing disability living allowance with an independent living payment and introducing a single tier pension.

"Our part for the politicians is that if they keep the benefits simple, we can do this," said Riley.

"One of the projects I worked on was pension credit, which was supposed to have been a simplification of pensions. It ended up being more complicated than what we had. So there is a partnership with politicians that they keep it nice and simple as well," he said.

Scrapping VME

Riley described the systems strategy at a recent Inside Government conference, where he said the DWP had concluded that it must scrap some of its oldest computer systems to get the job done.

"What we are building with Universal Credit, we are hoping to re-use a lot of what we've already got. [But] Our big systems are really difficult to change. The testing of them is three or four months alone

Old VME system compressed.png"We've got a large number of outdated, inflexible IT systems - VME systems," he said. "Changes take about 18 months in the lifecycle of a VME application."

"We can't manage it with those VME systems. We will have to replace those with systems that are componentised."

DWP hoped it could extend the consolidation and reuse programme beyond UC, so that it could build its other major reform projects using the same systems components. The matter was being reviewed to see if reuse would allow the systems to be delivered simultaneously.

It was certain, however, that UC would reuse core components consolidated across numerous existing systems.


It was a vast project, but DWP hoped it would be made simpler by delivering it in smaller chunks, in the agile fashion promoted in the Cabinet Office ICT Strategy.

Reform

DWP had bought into a rules engine called OPA it hoped would cut months from the time it would take to systemize the rules for UC.

It had also identified the core activities that would be consolidated from all its existing benefits systems: things like collecting evidence, calculating payments, making payments, maintaining accounts.

Citizens were meanwhile expecting things to be done in an online. Job Centres were no longer like miserable betting shops. But the VME systems were holding back DWP's rejuvenation.

It took 26 weeks to train DWP staffers to use the systems. There were 11 different systems employers used to put information online. It aimed to handle 80 per cent of claims online.

It's new policy was self-service and digital by default as long as it didn't exclude people with accessibility issues. It had cut its use of paper 50 per cent and aimed to automate 75 per cent of its processes. But about 30 per cent of people who relied on the DWP were digitally excluded in one form or another.

HMRC orders supply chain to heel

| No Comments
| More
HMRC building.png
HM Revenue and Customs has ordered ICT suppliers to step in line as it tries to cut costs from one of the UK's largest computing infrastructures and fix the perceived failures of one of the government's largest outsourcing contracts.

The initiative has seen HMRC take steps to address a problem Cabinet Office has sworn to tackle as part of its ICT Strategy: the exclusion of SMEs from government business and the tight control the UK's public sector ICT oligopoly has over government contracts.

In close collaboration with Cabinet Office, HMRC has told Capgemini to include SMEs in project meetings in Whitehall, indicating the extent to which the 240 suppliers to whom Capgemini subcontracts work its £8.5bn Aspire contract have been excluded to date.

HMRC has also ordered suppliers to open their kimonos so it can see for itself if it's getting value for money from them. The main criticism against large contracts has been the power they give outsourcing firms to inflate prices and suppress innovation.

Mark Hall, deputy CIO at HMRC, told government IT managers at a recent Inside Government conference, its changes were part of a programme to better manage its "commercial model" with suppliers.

"We have started a new process called Tripartite," he said, "Which is where we were always challenged, around getting everybody to the table.

"So now when we have conversations, the prime contractor, ourselves and the SME delivering the goods are in the same conversation.

"We are working around cost savings and working very closely and collaboratively with the cabinet office. This is not just about what we can negotiate, its about what UK government can negotiate," he said.

Suppliers told to open up

HMRC aimed to cut £900m, 25 per cent of its cost base. The other major reform programme involved identifying "value chains". That meant understanding where its expenditure was delivering value, from the citizen "right through to the to the final piece of technology in the supplier".

"This is quite challenging for supply chains because it demands transparency," said Hall.

"You need to be able to open up and look at the whole supply chain. This is not dissimilar to what motor manufacturers or logistics companies do. It's the same as Lean, but where Lean works bottom up, we are doing it top down," he said.

HMRC tap.png
HMRC was in addition scrapping unwanted computer systems to cut the cost of processing £435bn revenues and £39bn of payments it transacts with 38m citizens and businesses every year.

"It's about moving from a complicated legacy estate due to multiple inheritances, to what would be a simplified IT estate based on just 13 core platforms," he said.

HMRC hoped the consolidation would be self-financing. Savings taken from decommissioned IT systems would be pumped back into the rationalisation programme.

The ICT Strategy had nevertheless not changed HMRC's "direction of travel" since Cabinet Office launched it in March, said Hall.

"Its made us think slightly differently," he said.

It was not dissimilar to the last strategy, published under the Labour government in January 2010. But he did concede it was less about centrally provisioning an enterprise architecture and more about collaboration and the Agile methodology.

Agility problems

HMRC felt Agile was a challenge, though, when it had to rely so much on external suppliers.

"The new approach is far more agile, far more collaborative," said Hall.

"The question is then how do you [create] a delivery approach that allows that to happen, because agility's great but you do all need to work in the same way," he said.

It's in-house team of 200 developers were "increasingly" being turned to Agile. Their future reform was assured and was being examined.

Marion Sinclair, information systems strategist at Kensington & Chelsea Borough Council, asked Hall an Agile question, though she didn't mention the A-word.

Was there any way to align IT with business processes in a way that allowed projects to keep up with the speed of change and deliver efficiencies sooner? Hall said HMRC was trying to do that with Lean process re-engineering. It took power from IT people and put it in the hands of process people.

HMRC was meanwhile still wringing savings out of its £8.5bn contract with HMRC. It had renegotiated the contract in 2006, 2007 and 2009. It would realise £125m of savings from the latest negotiations this financial year. They would equate to £1.2bn of savings over its life.

Agile will fail GovIT, says corporate lawyer

| 58 Comments
| More
In a guest blog, corporate IT lawyer Alistair Maughan argues that Agile development is an evangelical fad ill-suited to government IT.
 
Maughan_Alistair High Res.jpgThe Government ICT strategy had some good ideas. Agile project management isn't one of them.

The Cabinet Office expects Agile will reduce the risk of ICT project failure. It's a nice idea in theory. But it won't work in government IT. It won't work in the real world.

Two of the most cautionary examples of failed ICT projects in recent years demonstrated the drawbacks of Agile.

The court battles of BSkyB v EDS and De Beers v Atos Origin showed that when Agile projects go wrong, they can go spectacularly wrong.

The Agile methodology is meant to deliver IT projects flexibly, in iterations. It's meant to involve customers more directly and adapt quickly to their changing needs. This means the final system only emerges gradually. It means customers don't pay a fixed price for a complete project. They pay for a commitment of resources.

But the lack of clearly defined project roles and requirements is a problem for Agile.

Agile evangelists argue fiercely that the conventional waterfall development methodology is unrealistic. They say the sheer scope of work required by its pre-set deliverables often leaves it unable to fulfill expectations. They set themselves up to fail, say the evangelists, when they should be working collaboratively for success.

I'm prepared to accept on trust that, if all goes well, Agile may reduce the risk of project failure. But Agile simply won't work in the real world of government ICT. We need a Richard Dawkins to bust the myth of the Agile gospel.

Sceptical

There are four clear reasons why Agile won't work in government ICT. The most obvious is that government customers want to know up-front how much a system will cost. That's not so unusual, is it?

Under Agile projects, you pay a given amount of money for a set amount of effort. But you can't guarantee a specified outcome for a specific price.

This won't work in government. Departmental budgets are managed very tightly, and they must be approved. Agile implies that charges for time & materials should be open ended. Government departments won't accept that.

Government is also legally required to follow open procurement rules.

That means comparing different bidders on a like-for-like basis, and deciding on best value for money. Agile can't give you a clear specification of outputs up-front. Nor can it give a definitive up-font price.

So how are government bodies supposed to make Agile comply with the legal requirement that public procurements are fair and open?

Unprotected

As if that isn't problem enough, Agile offers insufficient means of remedy if things go wrong.

This is a particularly sensitive issue for government, where departments suffer public opprobrium if their project isn't a resounding success. The press, the National Audit Office, and the Public Accounts committee (PAC) will give government a kicking if they can't make suppliers pay for the damage they caused.

Agile makes it hard to apportion blame because the customer is intimately involved in the work. Since Agile contracts lack clear contractual delivery obligations or remedies, how do you enforce properly? How do you recover loss or damage if there's a problem?

I wouldn't want to be the first Permanent Secretary to admit before the PAC that his or her department has no real right of legal recovery from a failure.

Poor fit

Agile is fourthly not suited to public sector management structures.

Decision-making is centralised in government. Civil service structures ensure every important decision flows up to senior levels. The Cabinet Office has under the current government taken even greater power over ICT projects. But Agile decision-making (over requirements, for example) flows down. This is key, so small devolved teams can react quickly and adjust to new scenarios.

It is inevitable that Agile decisions will go through management hierarchies in central government. This will be like kryptonite to Agile projects.

Agile projects rely on decisions based on mutual trust. They are therefore well suited to in-house projects. But the faith they ask customers to have in service providers makes them ill-suited for external developments.

You can have an ICT project with a watertight contract, clear deliverables, openly and legally procured, with a fixed price and appropriate remedies if you don't get what you want. Or you can have an Agile project. You can't have both.

I do appreciate that as a lawyer specialising in large ICT projects, you may think, "Well, he would say that, wouldn't he?". But my job is the help create successful projects.

I've seen too many projects flounder for a lack of trust between customer and supplier to think the answer to government's ICT problems is the Agile credo of, "Let's trust each other some more".

Partner and head of Technology Transactions at international law firm Morrison Foerster, Alistair Maughan has advised on large public and private IT contracts including HM Revenue & Custom's controversial 10-year £8.5bn deal with Capgemini. Follow him on twitter @ICToutsourcelaw

Police data hub raises doubts over open source policy

| No Comments
| More
A landmark software deal that exemplifies key elements of the government's public sector reform programme may have exposed shortcomings in open source policy and plans for an IT-enabled Big Society.

The deal involves the National Police Improvement Agency (NPIA) selling its Code List Management System (CLMS), a core component of the Police National Database, to Liberata, a private sector ICT supplier.

Public and private police in the Big Society - crop2.pngWhile a Cabinet Office endorsement may turn CLMS into one of the major components of the IT-enabled Big Society, the Liberata deal shows what the government's Localism Bill may mean for local government ICT and the Cabinet Office ICT strategy. This forecast isn't good.

The CLMS data hub deal is itself the hub of this conflation.

It involved NPIA giving its software to Liberata. The ICT supplier promised in return to offer CLMS services free of charge to the public sector. It could otherwise do what it liked and NPIA would get a share of profits.

Yet for all its apparent daring, the deal suffered a terrible lack of ambition. NPIA had raised the issue of open source with Liberata, acknowledging the government's preference for open source software.

NPIA was down with it, but didn't write it into the contract. Liberata's commercial interests were intractably old-world. It would give the public sector the service for free, but not the software code.


The decision may have immense consequences, both for the profit the pair will make, and in the opportunity cost the public sector will take.

The significance lies in what CLMS does and what it may become. It was developed as the heart of a circulatory system between different criminal databases held by the UK's 53 police forces and agencies.

Thus linked, the databases have been combined into a central, Police National Database (PND) to be launched in June. CLMS helped make their data capable of being shared and combined by making sure they all used the same data taxonomies: the different values deemed valid for different fields in a database.

Murders

The significance of this feat of data engineering was not lost on the Cabinet Office. The 2004 Bichard Inquiry into the intelligence failures that prevented police thwarting the murderer of Soham school girls Holly Wells and Jessica Chapman had blamed poor communication between police databases. It's taken this long to integrate police databases and realise the operation's implications.

Andy Waters, a systems architect who managed the CLMS commercialisation at NPIA, said the agency convinced the Cabinet Office cross-government data sub-committee (called X-Gov Information Domain) the system would benefit the wider public sector.

60 to 70 per cent of government data is held in code lists. Incompatibility is rife. CLMS cut development of the Police National Database by six months to one year and lopped 20 per cent from its cost by ironing out the differences, between say, codes different databases use for gender. It thus already does for the police what the Cabinet Office wants to do for the Big Society: make it interoperable.

So the CIO Council awarded CLMS "exemplar" status. Waters said Cabinet Office wants to roll it out across government as part of its open standards push.

Internet visualisation - crop.png
"We are engaging with Cabinet Office now with a view to more widespread adoption of CLMS," he said. "Our premise is to provide a single one-stop shop for all government data standards. That's the reason why Liberata are making it free to government, to encourage them to adopt it.

"We are in discussions as to what changes [Cabinet Office] require to enable them to mandate its use across government," he said.

Big Society data hub

Cabinet Office wanted CLMS capable of handling post codes, sort codes and other more complex data structures common in the wider public and private sectors.

While Liberata develops the upgrade, NPIA is doing a pre-sales routine with government departments. When CLMS ticks all the boxes, the CIO Council will give it "Champion" status. It would become the de facto Code List Management System for the Big Society.

CLMS would enable a significant part of Cabinet Office's open data and open standards policies. The software makes it possible for anyone to see and use the code lists that populate the PND. The 15,000 crimes that populate the PND's, offences field for example (yes, 15,000 offences), are publicly available; as are the vehicle make/model lists.

Such transparency would fuel public and private innovation. CLMS would then provide mechanisms for homogenising and interfacing between different code lists managed by different public and private bodies, fuelling more innovation. That's how an IT-enabled Big Society is supposed to work.

Even so, the NPIA/Liberata model does not bode well for the Big Society reforms.

The government's Localism Bill intends to give local authorities a general power of competence, which will allow them to operate commercially. Swingeing budget cuts have forced Socitm to advise them not to retain software engineering teams, forcing them to rely entirely on private development for public IT innovations. The combination will cause local authorities to commercialize more public IT systems.

So what?

NPIA effectively gave Liberata a commercial hold over one of the major circulatory systems of the IT-enabled Big Society; a blank cheque to commercialise access to open government standards and data.

CLMS is attractive to Liberata because the private sector will flock to it: to have such close interoperability with government systems will become a commercial necessity. The standards set in the public sector by CLMS may thus become standard throughout private and public Britain.

Liberata.pngWorking from NPIA's conservative estimates, this could net Liberata £62m. NPIA is counting on it. Liberata paid no money for CLMS. NPIA will instead get up to 8 per cent of revenues, which it reckons may amount to £5m over 5 years.

Though NPIA is a quango, its Liberata deal exemplifies what localism will mean for ICT more generally when councils get a general power of competence. Crucially, this demonstrates the impotency of the government's open source policy when faced with the prospect of short term commercial gains.

Open source impotence

Waters said NPIA would prefer the CLMS software to be open source - i.e. for the software code to be freely available, not merely for the service to be free of charge to the public sector.

But NPIA left the decision to Liberata. Since the government favoured open source, it assumed Liberata would. It did not deem it necessary to make open source a contractual obligation.

Open source impotence.pngOpen source seemed like a no-brainer. If the CLMS source code were open, anyone in the public sector could contribute enhancements or add-ons. The more people add things on, said Waters, the more valuable the service becomes. Liberata saw this too.

"Liberata have stated it's their intent to go the open source route," Waters told Computer Weekly. "[But] I can't speak on their behalf. It's their decision. The nature of the concession contract is we give Liberata the commercial freedom to develop the service as they see fit."

But open source is poorly understood by industry. It is therefore being overlooked as an empowering model for government's Big Society and Localism schemes.

This is because the incumbent industry works to the end of profit by whatever means, while open source is on a completely different trajectory: it's the bottom-up model that is meant to define this government's political term.

Industry ignorance

It is not surprising therefore that the £100m Liberata, which was bought by equity investors in January, appears like other software suppliers to have considered the government order that public sector software should be open "where appropriate" and decided without hesitation that its not appropriate.

David Mitton, CLMS business development manager at Liberata, told Computer Weekly he couldn't discuss the open source question because it was still developing its commercial strategy.

"Whether it's open source or not, I'm still working through those details. I don't have a yes or a no. I've not even discussed it," said Mitton.

But Liberata has already done the deal. If it wasn't designed on an open source business model, Liberata is even less likely to discover the model to be appropriate after considering its potential routes to market. What is more apparent in the Liberata example is industry's contempt for open source and the Cabinet Office policy that calls for it to be put first.

"I would say its just guesswork and conjecture at the moment," said Mitton of the open source question. "I'm looking through the government ICT strategy paper, and open source - I've not even seen it in there.
Mechagodzilla.png

"It's clearly an important part of all software solutions going forward, and it's on my agenda to deal with it this week," he said.

But did Liberata know whether there might be a potential advantage in making the CLMS software open source, and what that advantage might be?

"I'm very busy," said Mitton. "I'm not prepared to debate that with you. I've got no more to add."

The deal looks lucrative for both Liberata and NPIA, especially if it gives Liberata a monopoly over public sector code lists. But it misses the point of the Cabinet Office IT reforms, which are ultimately a realisation that IT suppliers had got used to making too much money because they had too much power.

The bottom-up model, in being more open and collaborative, would naturally mean more modest gains all round. All round.

Government to end ICT "oligopoly"

| No Comments
| More
The government has promised to bring down the ICT oligopoly as part of a strategy that may have seismic consequences for the public and private sectors.

The 18 large ICT suppliers that have controlled 80 per cent of Britain's much maligned public sector IT will have the rug pulled from beneath them if the reforms promised in today's Cabinet Office IT Strategy work as envisaged.

Big Ben.pngPolicy makers will be freed from the lead shoes put on them by the lumbering, multi-billion IT contracts that have tied them to the big suppliers and hold political initiative back, the policy claims.  

Open standards will be "imposed" on technology in the public sector, creating a country-wide computing platform that will subvert the misshapen procurement regime by allowing young, innovative firms to merely plug-in to UK.gov.

Such an ecosystem may become a feat of civil engineering to define the early 21st Century: a public work to make monoliths like the National Programme for IT seem like the work of 19th Century engineers.

The reforms will bring far-reaching changes to every corner of the public sector. They will require the civil service to agree common working practices in order for the computer system processes built upon them to be interoperable.

"The government will put an end to the oligopoly of large suppliers that monopolise its ICT provision," declared the strategy.

But government would "move away from large ICT projects that are slow to implement or pose a greater risk of failure" only "where possible". The government is already locked into IT contracts, such as the £8.5bn Aspire deal HMRC has with Capgemini, that preceded the last Parliamentary term and may yet outlive the present government.

The government would nevertheless adopt a "presumption against" IT contracts above £100m, while HMRC got a special mention for a plug and play website into which IT SMEs have been plugging tax data apps.

Open source

The strategy also carried forward a raft of reforms begun under the last government. The coalition would re-use its software systems, instead of buying them anew for different departments.

It would build an "asset register" and a government app store. Senior officials would be made to take more interest in and responsibility for their IT systems so fewer of them turned into hash.

But just as the last government in 2009 promised "a level playing field for open source", and a year later promised the same again. It has now been promised again. Open source would get a level playing field and would be used "where possible".

Both the government may have to step up their campaigning efforts if it wants its next ICT strategy, in 2014, to reflect any serious support it may be have for open source.

The programme of reform would, appropriately enough for one proposing lean and agile development methodologies, only last 24 months.

Any longer would be too long to leave the large ICT suppliers unsupervised.

We gave up, says GovIT SME

| No Comments
| More
It's so hard dealing with the large suppliers who control UK government IT that its not even worth bothering to knock on their door.

So said a public sector SME that has built bespoke software for the last 20 years. Systems integrators are supposed to manage government's IT supply chain, bringing SMEs into the loop and keeping costs down. But SMEs say they push costs up and pick the best work for themselves.

Steve Elliott - SFW.png
Steve Elliott, director and co-founder of SFW, told Computer Weekly he had given up knocking on the doors of systems integrators because they never answered.

Elliott had approached most of those few SIs that control the majority of public sector IT. The £4.5m company had never found any success with such approaches at all.

"We don't engage much with SIs now. Because we haven't been successful in the past we prefer to continue to engage directly with our clients," said Elliott, whose business gets 80 per cent of its business from the public sector.

SMEs are usually scared of speaking out, for fear of being punished commercially. Elliott insisted he wouldn't name names.

But he claimed to have stories where SIs had closed him out of clients or made money for old rope.

One involved an SI approving the work SFW was doing for a client. It involved developing a browser add-on so the client could display some proprietary file formats online.

Quote my quote

"The SI had to see how much it would cost to evaluate and test it, so it was like a quote for a quote. They wanted to do a feasibility study to see how long the quote would take," said Elliott.

"When you compare it to what we were charging and the time-scales we were charging it was a different order of magnitude really. They were taking weeks to do something that would normally take us a few days.

"The department was frustrated. Its not that they were over-charging. They were charging as per the contract," Elliott said.

Without specific exmples there is no way of verifying Elliott's story. SMEs often make such unfounded reports. All SIs get branded alike. None specifically get the blame.

System error

Nevertheless, SIs blunted the competitive edge of those few SMEs they did employ by lumping a five to 10 per cent fee on top of their costs, said Elliott, whose clients include ACAS, DEFRA, and the Veterinary Laboratory Agency.

SMEs are most resentful of this fee, even though they acknowledge SIs are charging for procurement services rendered and risks shouldered.

Elliott said the trouble for SMEs started around 2005 when the Office of Government Commerce (now part of the Cabinet Office) began aggregating its IT business into such large clumps SMEs were automatically excluded. OGC Buying Solutions simultaneously cut the numbers of suppliers on its Framework Contracts in response to changes in EU procurement law. 

SFW had spent a few years on the S-CAT framework in the days of the e-Government push and had seen growth of 20 per cent a year. It was kicked off in 2005 and for the last 3 years growth has been flat. The market influence looks strong here, but Elliott thinks a "significant" difference is made when SIs control access to clients.

Subscribe to blog feed

Archives

-- Advertisement --