SoylentNews is people
SoylentNews
from the so-much-for-project-management dept.
A UK government report (board minutes from the Health & Social Care Information Centre) says that the National Health Service has £5 billion worth of Information Technology projects at high risk of failure:
The ratings are based on gateway reports assessing the risk of four IT projects this year. All are related as "red" or "amber/red" meaning successful delivery is either impossible or extremely unlikely. Those projects include the remaining electronic health records contracts with BT and CSC, due to end in 2015 and 2016.
According to the HSCIC report, the £2.3bn CSC Local Service Provider (LSP) programme has now been flagged as "red", up from "amber/red" when the Major Projects Authority last released its rating for September 2014. Both programmes were originally started in 2003/2004 and have had an extremely troubled history.
Other high-rated projects on the list included the £168m NHSmail2 programme, to provide secure email across the NHS, which has slipped from "amber" to "amber/red".
NHSmail2 is an upgrade to the NHS's Microsoft Exchange based email system. Computer Sciences Corporation (CSC) and BT Health London have managed IT services for different divisions of England (CSC manages the North, Midlands & Eastern cluster, BT manages the London cluster).
Previously: UK National Health Service Dumps Oracle For FOSS NoSQL
(Score: 5, Interesting) by ledow on Wednesday December 09 2015, @06:43AM
Context, from a Brit IT guy:
Every NHS IT project is a laughing stock and doomed to failure.
Consultants overspec, overprice, underdeliver but get their lawyers to write the contracts. They win in all possible ways because the people who hire them are too stupid to specify the system and contract to be fair to both parties.
NHS IT projects are legendary. They sent medical records to companies in India to digitise them so that your GP could get all your history on-screen. Ten years later, they admitted it was a total failure, that the people that were typing them up couldn't read them, that records were lost, or confused, or plain useless, and that mistakes like hypo/hyper were all over the place. It was abandoned. Last I saw, my medical records are still on the same pieces of paper as were put in there when I was born (yes, I have read my complete medical records, including notes on an eye injury I sustained at birth).
Everything is a disaster, and the NHS is the absolute pinnacle of calamity where IT is concerned. I'm still fighting with the local GP and pharmacy over my girlfriend's "electronic prescription" which keeps ending up going to the local superstore (4 miles away), when the pharmacy is opposite the doctors, which is on the same road as our house. And whenever we go to collect her prescription, the supermarket NEVER has the full prescription. Every six months, it happens again (after working as intended for a while, collecting from the local pharmacy), and I have the row with them again, and we go round the cycle again.
The NHS buy equipment on the worst kinds of procurement processes. My girlfriend works for the NHS, with millions of pounds of equipment under her control. She's had a £120,000 microscope out of action since it was new six months ago, and nothing's happened about it. It's just cheap shit foisted on her lab, then breaks, then never gets repaired, then in ten years it might get replaced. In the meantime, service and support do less than she can do on it herself. She has lab devices of that calibre running on Windows XP, not connected to the NHS backbone, which she has to move results over from manually, and has to be re-imaged once a year or so as the drives die - and because it's controlling millions of pounds worth of hardware, it's too expensive to replace and the manufacturer's don't care unless you're going to buy new.
If you buy something like a AA battery, it will cost DOZENS of times more than just walking down to the shops and picking up the most expensive brand. Because the guy who won the procurement contract for that is the guy that gives them 1% cheaper high-end stuff, so they get chosen as a preferred supplier and, after ten layers of abstraction, charge extortionate amounts for poor delivery of crap products that are extraordinarily over-priced... because they can. NHS procurement won't do anything about it.
If you see NHS and IT together, as a UK guy, you steer clear and walk away, or (if you're an evil bastard) you see an opportunity to get rich very quick and walk away scot-free not having delivered a damn thing.
Everything you see that works in the NHS is not down to the IT but despite it. My girlfriend is a highly-specialised professional, at the top of her game. She literally spent two years convincing the Trust to buy her a UPS for her incredibly expensive microscope (in the millions of pounds) that cuts out and loses all the captured samples every time the power blips (even with generator backup in the hospital!). Patients were literally told "Sorry, you have to re-do that incredibly painful bone marrow sample because we can't tell you whether you have cancer or not because the power went out and it fucked every result in the queue. Oh, and you'll have to wait another week to know anything from that too." The UPS cost £100. I saw the exact specs and model. The NHS paid THOUSANDS for it, and for someone to "fit it" (i.e. plug in two cables).
If it involves NHS and IT, just laugh and walk away. Honestly. It's not even news any more. This has been the case for 20-30 years now.
(Score: 3, Interesting) by frojack on Wednesday December 09 2015, @06:58AM
Consultants overspec, overprice, underdeliver but get their lawyers to write the contracts.
Heh, another Brit thinking its all special in the UK.
Big government just about ALWAYS gets it wrong. All around the world.
Remember Obama having to call in Google, and Oracle, and every body else to bail out Obamacare?
The bigger the contract the higher propensity to fail. The rule of 5s: Any project involving more than 5 companies, or 5 million dollars, or 5 years, has a one in 500 chance of being completed AT ALL, let along on time, one in 5000 chance of being on budget.
No, you are mistaken. I've always had this sig.
(Score: 3, Insightful) by edIII on Wednesday December 09 2015, @08:08AM
No kidding. He may as well have been describing the Pentagon instead of the NHS. $10,000 hammers, $50.00 band aids, $1,000 toilet seats, as the sayings go. If you want to get really rich in the U.S on the taxpayers tits, go into the military industrial complex. Their budget isn't anywhere near 5bn pounds. We wish. More like 2,0000 times that in the last 10 years alone.
Comparable fuckups are quite easily found around here :)
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2, Insightful) by Anonymous Coward on Wednesday December 09 2015, @09:27AM
The bigger the contract the higher propensity to fail. The rule of 5s: Any project involving more than 5 companies, or 5 million dollars, or 5 years, has a one in 500 chance of being completed AT ALL, let along on time, one in 5000 chance of being on budget.
Wasn't like that before e.g. Manhattan Project, Apollo program. And plenty of other successful military aircraft were built too (Harrier, Lightning, Panavia Tornado, F14, F15, F16, A10, Apache, Lynx). Same goes for many other stuff like ships and submarines.
Perhaps the parasite load is too high nowadays: http://despair.com/products/consulting [despair.com]
The solution is getting rid of the parasites and the people who keep adding them in.
(Score: 3, Interesting) by frojack on Wednesday December 09 2015, @05:29PM
And plenty of other successful military aircraft were built too (Harrier, Lightning, Panavia Tornado, F14, F15, F16, A10, Apache, Lynx). Same goes for many other stuff like ships and submarines.
No.
You name a bunch of successful projects, to suggest that large scale failed IT projects have always been with us.
Yet IT has not always been with us.
And IT systems seem particularly prone [computerworld.com] to monumental spending over runs yet in the end being scrapped totally, because they could never be completed.
http://calleam.com/WTPF/?page_id=1445 [calleam.com]
No, you are mistaken. I've always had this sig.
(Score: 2) by DeathMonkey on Wednesday December 09 2015, @07:11PM
Remember Obama having to call in Google, and Oracle, and every body else to bail out Obamacare?
No, I don't remember Oracle bailing out the system built by Oracle. [latimes.com]
(Score: 2) by frojack on Wednesday December 09 2015, @07:32PM
Try not to confuse Oregon's hapless non-existent system with the Federal exchange that crashed, burned, and smoked for weeks back in 2013.
http://www.businessinsider.com/obamacare-website-healthcare-gov-google-oracle-red-hat-2013-10 [businessinsider.com]
No, you are mistaken. I've always had this sig.
(Score: 2) by DeathMonkey on Wednesday December 09 2015, @07:36PM
Built by CGI Federal, another private firm. [washingtonpost.com]
(Score: 0) by Anonymous Coward on Wednesday December 09 2015, @08:30AM
Nb: they are trying to do the same with procurement in research councils as well. E.g. they told folks that they have to use Air Products for supply of gases rather than BOC and prices for everything doubled, quality went down. Fortunately the scientists know enough about what they are doing to ignore management.
(Score: 3, Funny) by c0lo on Wednesday December 09 2015, @10:33AM
Your opinion, sir. For us, the contractors, is a good YoY revenue. At least, while it lasts.
(but I see you know this already)
What you may not be aware is the fact that after we screw up a project beyond endurance, the govt allow us tendering for the next major project. So, for instance, NHS busted? No problem, maybe the immigration or perhaps education need a gargantuan IT project and it will still be us at it. Because:
---
* finishing it is not important! Running it? The longer the better!
**
and "Nobody was fired for buying from IBM/MS/Oracle"
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by TheRaven on Wednesday December 09 2015, @11:29AM
after we screw up a project beyond endurance, the govt allow us tendering for the next major project
I think it's slightly improved recently, but it used to be even worse: If you have 'experience' with government contracting then this was considered a major benefit and your bid would be heavily weighted as a result. If you had taken 100 government contracts before, then you were very likely to get the next one, even if you had failed to deliver on any of them. This is why companies like EDS kept getting contracts, in spite of never managing to complete any of them successfully.
sudo mod me up
(Score: 0) by Anonymous Coward on Wednesday December 09 2015, @11:32AM
(Score: 2) by Thexalon on Wednesday December 09 2015, @03:59PM
With all that said, the NHS gets better results than the US health care hodgepodge, and at much lower cost. If you think it's bad in the UK, just check out what's going on on the other side of the pond.
Also, without the context of the total number of software projects on the table, the £5 billion number may or may not be all that significant. If, for example, there's £5 billion worth of failure, and £15 billion worth of success, then that's actually a pretty good track record, and the fact that they're keeping close tabs on that £5 billion is a good sign. If, on the other hand, there's £5 billion worth of failure and £1 billion worth of success, then there's a serious problem.
You seem to be working under the impression that corruption is the exception rather than the norm.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 1) by purple_cobra on Wednesday December 09 2015, @09:50PM
Disclaimer: I gave up proper (i.e. paid) IT work a number of years ago following time out of the industry due to illness. A number of years later I got a job turning awful doctor-speak (and the ones who speak English as a first language are often the worst) into something a normal person can understand.
But yeah, I can confirm this, at least the bits I'm privy to from my position in the food chain and from the few people I still keep in contact with still doing that sort of job. A friend of a friend was working on the NHSMail project and what I heard was, quite frankly, terrifying in terms of what is being wasted.
Of note:
Our radiology system runs on Java. Very bad, very slow, probably unmaintained and tied to an ancient JRE Java.
All the applications that read patient data supposedly read it from the same place, yet can give results when accessing the same record.
Patient records are scanned into a central record system. This, in theory, should be great but as all the text is scanned as images, you can't search for anything within them. That would be a pointless exercise anyway as a) like all NHS software, it uses a proprietary window library and b) the indexing is broken in some fundamental way. Also, starting up the application wipes your current paste buffer because there's a trivial bug in the software that causes text pasted into patient records to delete itself when saved (I suspect it's a CR/LF issue triggered when copy/pasting from a Word document, but see my caveat above) and rather than fix that, it was easier to zap the paste buffer because after all, no-one will just re-copy/re-paste it!
I could go on, but having to deal with this utter shit every day is painful enough without describing it in gory detail when I've escaped for the day.
But yes, everything is a) shit, b) expensive, and c) it will only get worse as we've had successive governments (New Labour, Tory-Lib coalition, Tory) who have a massive hard-on for privatising every damn thing rather than trying to fix what we have. Funniest part of that? Surgery in the private sector carries a 30 day warranty. If anything goes wrong after that, you get to pay again. You get more of a guarantee on a £5 kettle, FFS. The NHS investigates and cures a fair few private healthcare failures, yet it's private healthcare lobbyists who are trying to get rid of the NHS. New head of NHS England is....a former private healthcare wonk.
The bits that work do that pretty well, but the gradual switch from a national, unified health service to a bunch of competing mini-health services (that is, in essence, what foundation trusts are) has screwed the NHS. The place I work for will be in special measures inside of 2 years thanks to a PFI contract (i.e. a giant mega-mortgage) and the board's wonderful idea of saving money by not providing services. Our shiny new MBAs (hired despite the recruitment freeze) are only accelerating the process.
FWIW, the classic case study for IT/procurement failures while I was in university (pushing 20 years ago now!) was...the London Ambulance Service. And NHSMail2 will, no doubt, be a classic case study for the class of tomorrow. NHS procurement is a disaster which, alongside the internal market and the reams of bullying, hectoring managers, is going to kill the NHS once and for all. IT may be a disaster, but it's one piece of a bigger picture, one that I'm not sure is solvable.
Never mind, eh? At least it isn't Corbyn and his band of Trots!