UK taxman waves through £168.8m Fujitsu contract because no one else can hold up 30-year-old infrastructure UK tax collector HMRC has awarded Fujitsu a £168.8m contract without competition to ensure critical applications keep running as projects to replace them are delayed and Brexit pressures mount. Under the deal, the Japan-based IT services firm is set to continue hosting for 13 HMRC applications on its Virtual Managed …
You are wrong. CHIEF was (and is) written in COBOL. It uses an extensive library to interface with the IDMS database and the transaction processor, and to simplify some parts of coding there is a light pre-processor to convert some bits of the code, but the code really is essentially COBOL.
Source: I worked on it.
So you have legacy code and infrastructure from Fujitsu with Capgemini, IBM (attempting) to manage AWS and Azure deployments to the cloud, while of course not trying to pressure HMRC to use IBM cloud?
Why not bring in Capita with Sopra in joint venture for CNI oversight and appoint Dido Harding at CEO to run the whole thing?
When your data gets shunted over to the mainframe in the obernight batch job, the calculations are still done in an excel spreadsheet.
I'd laugh if I didnt have to use the thing.
For bonus laughs, check out the list of exclusion cases on gov.uk. over a hundred ways that your tax cant be worked out on CESA, and instead needs to be done mmanually.
Thank God there are procedures and such to ensure that the Government gets the best deals possible, right ?
Oh, I'm sorry, the Government has decided to bypass the procedures ? How lovely. Just you try bypassing paying your taxes. You'll find the procedures are firmly in place and well respected, I'm sure.
As usual, one rule for them, one rule for us.
Well, that takes me back to my ICL days (before Fujitsu bought them by taking over STC) working on a VME mainframe. I even worked on the CHIEF system when it was run by BT some years ago now. And I remember BT losing the bid for continuing to run CHIEF in 2010. Fortunately (or possibly unfortunately) I was not TUPE'd across to Cap with the project.
They don't make 'em like that any more.
Funny, how old systems which actually work last for ever and cannot be replaced, whereas new procurements are riven with difficulties and over-runs.
I'm sure the difference between old and new systems is that the old systems had vastly less flexibility in development and, as a result, a lot less interference by "know nothings" with "grand ideas", so people took what they were given and liked it, usually because it did the job it was supposed to.
Maybe not in this case.Tax is one of the things that gets tinkered with on a regular basis.
It's likely to be a result of careful design a long time ago rather the frAgile development doing what's easy or absolutely necessary this sprint and leaving the difficult stuff like flexibility till later. And later. And really a lot later. Either that or the bits that threw in obstacles were hammered out of the way in the course of successive budgets - that's the Chancellor's budgets, the ones he reads to Parliament.
Funny and quite common actually. It comes from a very surprising factor: the main reason those systems have endured 30 years of being patched to death and used for everything is that ... they were initially not replacing any legacy systems.
When you have a reliable workhorse that more or less still works day in and day out (yes, with all the quirks and deficiencies and hand holding of a 30 year old system, but works) you can afford the luxury of delaying the replacement practically forever without no one having to endure the burn out of a complex replacement go live. How many XP boxes are still running today in both critical and non-critical roles? The workarounds (virtualization, network segregation, etc) required to run them with some peace of mind may seem costly, but short term are nothing compared to rolling out new ones.
Not "still entirely" but "also now" - I worked in VME Support in East Putney in 1987-88 - primarily in my sandwich year of my HND as a support technician but they taught me VME base and comms, so I did some second-line comms diagnostics too. I worked with John Mearns, who designed TME (on ME29s) and then TME* (which ran on VME) and I supported an ME29 and a Series 39 Dual (the wrong half however - but that was no obstacle for the brains in the office who got it fully working!). I then worked in Stevenage (STE04) 1989-91 in Office Systems Support (leaving shortly after ICL got completely swallowed by Fujitsu) next to some of the best IT brains in the industry. It's true though that much of the 3rd line was based elsewhere (Alsager/Manchester)
Having worked for, having been 'supported' by them, worked along side them and and having to rely on them as their customer for over the last 20 years, I always wonder how (OK I know, they undercut everyone on the bid process), Fujitsu still win UK Government work.
In fairness they do have so great individual people working for them, but they are often not in the position to make a decision they are more qualified to make then the person who does, I feel for them, and it's not helped that some of the people I've worked with who have been parachuted in, or worse brought in as contractors, to be a SME or otherwise tasked with delivering something, have never had any involvement with that product or technology before, ever, not even a passing interest. I've seen this in a Windows upgrade delivery where the contractors they brought in, due to a lack of staff, had never done desktop support let alone deployment before but had security clearance (guess how that went!), DBAs who have never touched any type of DB server prior to the engagement and network teams where they re-positioned application engineers and brought in contractors, both groups who had never touched any network before to make the new network team, and .
Sure many companies operate in the way I've just described, but I always find that Fujitsu are the worst at it and I actually left a job when Fujitsu took up a contract where there were taking over support of some of our IT.
And then the Horizon debacle, in that case that led to people being sent to prison, to the detriment of the customer (who wasn't blameless), again, not unique to them, but at some point (imprisoning people, maybe suicides?) that if you suspect there could possibly be something wrong with your system, or a capability that you know exists but tell a court under oath it doesn't,, sure an individual might make a personal decision, but several people? There's an issue.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
