GDS monopoly leaves UK.gov at risk of IT cock-ups, warns report The Government Digital Service's (GDS) current monopoly position on providing Whitehall IT places it at high risk of repeating the same costly IT disasters of the past. "With a monopoly position and a client base compelled to turn to GDS for advice, there is a risk that they could become an inefficient organisation removed …
Consultancy firm in 'government should pay more to consultancy firms' shocker...
Really, BDO are a disinterested party? I think not.
I'd rather have the inevitable scope-creep, overspend etc. limited to an organisation that is within government's power to reform. A permanent IT organisation is required for a permanent government.
As for 'who is going to maintain it', FFS, that's what IT organisations do. Who is going to maintain the Youtube API from 2012 that Google have just killed? Er, no one, that's who and that's increasingly private sector IT for you - pay us more money because we're about to pull the plug on something you're perfectly happy with.
Insource and be damned...
"With a monopoly position and a client base compelled to turn to GDS for advice, there is a risk that they could become an inefficient organisation removed from the efficiency drivers of the market," said a report by consultancy firm BDO.
Could...? Could...!?!
I know consultancies are supposed to be subtle about these things but come on! World + dog knows this already happened.
Posting AC as I work in IT support in a public body. We do buy stuff (the paperwork around that can be substantial) but a lot of work is creating bespoke applications as well. The needs of our staff are just niche for there to be commercially viable products out there. As long as the bespoke work is properly documented and isn't left to wither on the vine it should be okay.
As for things being late and over budget, it can be hard to get stakeholders engaged and when they realise the scope of what can be done there can be requests to add a lot of development at the end of the project. Our managers are usually goood about saying no or it will have to wait for the next development phase to be included. Not defending GDS there but there isn't enough information in the article to come to an informed opinion.
Copeland added: "I fear people are also being too dogmatic about using open source."
I have many issues with the Policy Exchange and it's lop sided ideas, I fear GDS is not being Dogmatic enough about open source, as the Tax Payers has paid for the code to be developed it should all be Open Sourced. The problem with closed sources is the supplier becomes the monopoly and that gives you a whole set of bigger problems (as an example do I need to say anything more than Microsoft)!
In one respect the Policy Exchange is correct GDS is becoming to monolithic and should be split up into three small organisations that are forced to compete, I'm a big fan of competition as this would then force each of the small organisations to keep improving.
Just asking.
Also, why is there an almost religious belief that .gov systems have to be "bespoke" ? It's attitudes like that which allowed Capita to bill Birmingham City Council £1.2 *million* a year to run a bespoke (and utterly crap) glorified CMS that is www.birmingham.gov.uk.
Because, like Russian homosexuals, there are no OSS projects that would have done the job.
That's just so 2013 for a mission statement, get up to date guys and start bullshitting about Government as a Service instead.
"GDS has become an in-house big consultancy firm itself, building everything bespoke, making it "special", unjoined up and delivering late."
In other words it's like the IT department at every big company I've ever worked for.
GDS build an excellent, user centric website which is rarely down, sets a model for other government sites worldwide, has revolutionised citizen engagement and which is highly regarded in terms of putting the citizen first and of rollout out agile ways of working. As far as I know GDS has saved the government a fortune which is more than can be said for how things used to be done. If GDS makes a mistake, it learns from it quickly - that's modern management thinking and the sort of thinking which has made Amazon, Facebook and Google into similar leaders in their fields.
GDS had a huge task on their hands, they have had successes and they have had failures. But they have had more successes than failures with less money than the way things used to be done. There is always a better way, there are always ways to learn and to improve. However chucking the baby out with the bathwater is hardly a positive approach.
Why shouldn't the government build everything in house? They run the country, they protect our national security, they pay benefits to millions of people every day, they protect the nation. They are our democracy and they are accountable to the people. Why then should we require a 3rd party subcontractor to run a website? What's the challenge that a 3rd party can do that a national government can't?
I notice that at no point do you refer to 'cost benefit' as being a determining factor in whether GDS should 'build not buy'. As a taxpayer I really think it should be.
What we now have is a rapidly growing empire that determines itself to be authoritative on all government IT matters. While GDS can admit to mistakes and move on, Gov departments will be left paying for the white elephants for years to come.
So GDS have authority without the inconvenience of responsibility.
Finally, I'm sure GDS would say that they've saved UK Gov. a fortune - their brave decision to standardise on black and white web pages has avoided months of civil servants and brand consultants agonising over a Pantone colour chart.
[snip screeds of self-congratulatory waffle]
I guess you work for GDS? Cos the outside view doesn't see anything like your view - instead it sees a self-absorbed body, obsessed with the latest buzzwords and delivering anything of little value.
Which is precisely why a torrent of EDS, Capita, BDO (didn't that use to be Stoy Hayward?) and other Tony Blair-inspired 'contractors' are now lining up to bleat at their loss of income. GDS over the last 3 years is estimated to have saved billions in development costs - look at how quickly https://www.gov.uk was put out and how little it cost - estimated at just under £95m in total, compared with average externals quotes of > £2 Billion.
It's fair to say that the developers and delivery teams in GDS are doing a good job of delivering what they are told to deliver.
I think you are missing the point here that GDS are also given power to influence and straight out mandate solutions to departments. With little experience or understanding of IT at the scale of Government, control functions cause delay after delay to projects and procurements in these big departments - and the miniscule savings of a million here or there that GDS generates by building a whizzy website are utterly swamped by the cost of pay and rations in an extended project.
As others have said, GDS is the very definition of a self absorbed ivory tower dwelling team, utterly unaware of the impact of the decisions and statements that it makes in isolation
I have worked in the Scientific Civil Service (a long time ago); then as a science/technology specialist and advisor in a very large public organization; then as the senior technical bod in a high tech private company; and then, until I retired, as the MD and major shareholder of a tech/consulting/software/science company. I have a radical idea. Why not go back to the old days when the UK science/tech civil service used to be "a job for life".
They used to oversee all of this stuff. For a large project you would have a (very) few (very) short term technical consultants in to teach/mentor the full time civil servants. The government types knew that, short of gross moral turpitude, they had a job for life - So they tended to make decisions based on their long term careers inside the service and (I can say this with a straight face) for the general good.
This tended to avoid the responsible people choosing whatever technology was new and likely to gain them employment outside the service. I have personal knowledge of a large public service contract in the late 1980s that was awarded because it used the new, shiny, coming-thing that was Sun kit. The project was implemented, and then the staff left for private enterprise. Something similar happened with Java in the mid to late 1990s...
It seems to be a mind-set issue - Way back then if you asked a public servant what they did they would tell you that they were employed in the public service. They did not say that they were a network engineer. This stuff worked really well until the commercial and political fiddling and cronyism that came in later. To encourage good people to stay, salaries were at least comparable with those outside (not now)."Special Merit" promotion grades were available to skilled people - If you were good at your job you were promoted, but you still did similar technical work. A "normal promotion" generally meant that you stopped doing what you were good at, and became an administrator (which many techie types are very bad at). Obviously there were the typical project creep/budget and implementation problems that you would expect, but there was little perception of practices that tended toward subornation.
Anyone who says that the old fossils that might result in the decision making process would mean that progress will stall, probably does not understand the basic problem. It is not all about RoR and Javascript and whatever is new and cool - What works, and is still probably needed, is mostly large *NIX systems, z/OS and an infrastructure to handle generic web traffic. So more open source, HTTP, and COBOL please.
I am expecting a lot of down votes from people who do very well with the current, broken, regime...
Not sure which way to vote to be honest. I disagree with you on some of the details (RoR is hardly new, and if you're building a web based system COBOL isn't going to be much help even if you can even find enough people to do the work), but the general gist of it you do make a fair point. If people have stable jobs they know they'll be in long term and are able to make decisions on how things are done, generally they'll try and do things in a way that won't screw them over in the long run. Unfortunately short term savings rule the roost now, both in government and industry, and it's rare that people have enough influence to push through a decision which is the right one rather than the cheapest or fastest one
You are both right and wrong. There is a mindset difference but at the same time, we do actually need to be seeing modern technologies in use in government. People don't engage with systems that are clunky and complex.
Websites need to be smooth and easy to use. Javascript is good for that. You're doing pretty much the exact same thing as the newbies in the jobs - you're discounting the new technologies as unnecessary when old technologies could do it but failing to take into consideration the real-world use cases and requirements of the systems.
If you look around the world at the large data crunching websites, you almost certainly won't find much COBOL in use anywhere! Facebook handles a billion users and it certainly doesn't use Unix and z/OS.
The technologies needed for a project should be chosen to fulfil the requirements of the project, simple as that.
@localzuk
I gave you an upvote.
I agree that people don't engage with systems that are clunky and complex. A simple easy to use interface, if well designed, shields the user from the data store which can be as complex as required. COBOL is still used in a lot of systems, for instance I believe that it still lurks within DVLA systems. A lot of legacy banking type applications use old z/OS and/or COBOL but the punter sees a web-based front end.
Facebook, I think, uses their own customized Linux across much of their systems Being old, I consider Linux to be a *NIX, although I admit that I might be biased towards systems that have a BSD legacy.
You may believe that I might be discounting new technologies, but I have developed stuff using them until very recently. One thing that I have noticed is that putting system logic in the interface almost always causes many problems, particularly with maintainability - I have seen several JavaScripted applications that do this.
The problem with the Civil Service and the politicians that "drive" them is that there is very little domain knowledge. Many people working there are "well educated" and by no means stupid, but there is a functional ignorance in anything remotely technical or scientific. Worse, many decision makers do not have the background or skill in the domains they are administrating to come to a view that will stand scrutiny by outside, informed, opinion. So they use consultants, whom have seen them coming...
The Civil Service then really doesn't help itself by rotating out the people that the contractors have trained up. The contractors have to start all over again with new ones every couple or three years. This appears to be an out of date hangup from colonial days about officials "going native".
To give the Government credit, they have recognised that big ticket projects are likely to equal gigantic (and expensive) failure - so they tried something else. The result has had some modest success (for at least 2 orders of magnitude less cost). It is by no means perfect, and some of the bad habits of old are creeping in, but by banging on at them in the technical press we can hold them to account because, at the very least, they actually understand the questions and the proffered answers.
To try and bespoke build all digital systems via GDS is absurd - just as buying them all would be.
Each case should be looked at individually, is a bespoke system necessary? Or is there an off the shelf product that can do the job for less cost?
Open source should be a necessity, regardless of the system, so that the government doesn't get locked in to any supplier contracts and to allow for rapid changes to be made should they be needed.
The complaints made by Copeland in this article scream "we have to outsource everything!" to me though.
And where can you purchase a system for e.g. paying road tax - yes one that integrates with VOSA's list of cars, MOT certificates and people verification systems? No, me neither.
Policy Exchange is just blathering its usual right-wing dogma "out-sourcing is better than a government monopoly", without taking into account the consensus that GDS do great work compared to the private sector. (See also BBC, NHS, etc)
If PE had a leg to stand on they would use simple metrics to establish quality and value for money. And compare with Cap Gemini, Fujitsu, IBM and Capita IT contracts. I bet there is no contest.
And they might also consider getting with 2015 and have a mobile-friendly website... No, seriously!
A much better focus would be why 400 local councils who all do the SAME thing, who do a heck of a lot of out sourcing, can still deliver such bad IT.
> And where can you purchase a system for e.g. paying road tax -
> yes one that integrates with VOSA's list of cars, MOT certificates
> and people verification systems? No, me neither.
You can easily get systems that have multiple integration points and interfaces that can communicate with other systems. Do those services support SOAP or some other form of data exchange format? That stuff isn't really difficult to configure into flexible integration points. You don't have to build custom applications all the time to handle that sort of thing.
Precisely. GDS's main outputs are characterised by a facile emphasis on appearance, a distrust of any technology that has been in use for more than ten minutes, and total lack of QA. Sorry, but taxpayers expect government sites to be correct.
I'll keep this to a single example: the "Check if you're a British citizen" page. Section 2 of the 1981 Act, which confers British citizenship on anyone whose father was both a citizen and born here, is completely ignored. Getting the right answer involves digging out the actual legislation (FTR, at http://www.legislation.gov.uk/ukpga/1981/61/section/2). But that requires Google, of course. So why isn't there a link from https://www.gov.uk/check-british-citizen to legislation.gov.uk ? Is it because it might frighten the horses or is it possibly because the person who designed the page didn't have the slightest idea what he (or she) was doing ? Answers on a postcard, as they say.
Here's your postcard:
Your example is wrong.
The first question asks if you're born on or after 1 Jan 1983, which is the commencement date for the 1981 Act. If you select 'born after 1983', then 'I was not born in the UK or a qualifying territory', then you are asked if either of your parents is a British citizen 'not by descent'. This is the clause you're looking for.
I presume that this would be the same BDO as in,
"Our PPP/PFI team understands the needs of procurers, providers and funders. They successfully balance these needs to achieve results and bring this experience to bear on our wider public sector work."
Bit strange how BDO's entire ethos appears to be "you should be using our provider partner Capita".
The government's requirements are generally unique, there is only one government per country as opposed to thousands of companies...
Besides that, "building inhouse" is not a bad thing as it ensures the platform is wholly owned and controlled by the government, and not beholden to a third party.
As for the extent to which they build things themselves, it's not like they're building everything from scratch - they will take a collection of existing technologies, integrate them together and apply whatever unique customisations are required for the task at hand.
If they had gone to one of the traditional outsourcing companies they would still have ended up with a bespoke system, but one which they don't own or control and are beholden to the supplier for, plus it would probably build from much more expensive base components and still have very significant customisations on top.
Agile, innovative yet standard: quite a tall order. Would these be the standards that do not allow for Welsh, let alone anything foreign (Estonian or Korean names)?
You only need to look at the dates on the documents in
https://www.gov.uk/government/collections/identity-assurance-enabling-trusted-transactions
to see that the 'constant improvements from user experience' cannot have been fed into them.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
