90 per cent of the UK's NHS is STILL relying on Windows XP

Move On Folks, Nothing Of Surprise To See Here

As someone who used to work in a permanent role in NHS IT until made redundant, I can tell people reading this that the trusts in question simply don't have the resources to do anything about it. IT support in the NHS is a daily reactionary process, fire-fighting, if they manage even that. In most trusts it is regarded as a necessary evil, an unwelcome financial overhead, in terms of hardware, software and staff support. If they are still running XP it means the PCs in question are probably incapable of running any other version of Windows and don't have the staff resources or the financial means to outsource upgrading as a project anyway. In the county I worked in, only one Trust out of the four had a rolling replacement programme. Of the other three, only a number of months ago I learnt that one of the teaching hospitals was still running XP, even though they had outsourced IT support to the IT services arm of a really world-famous IT company. Apparently the expectation was that support would be cheaper and better than a continued in-house operation, but if they actually saved any money none of it was re-invested. This was a hospital that once applied for Foundation Hospital status.

When cheap disposable PCs / tablets can be had for little more than £100 it would make sense to let local trusts buy their own PCs strictly for internet access while all internal systems run on separate hardware which never touches the internet so doesn't have to be continually upgraded.

Decent People

How about instead of trying to pay naff money for naff tech people, they pay decent money for decent tech people... that way they don't have to keep binning billion pound projects, because they've not even come close to accomplishing the task at hand...

A database which is accessible and easy, ... pay experts e.g. facebook, pay experts on what makes things easier e.g. the doctors they already have and not "experts" which haven't worked a day in a hospital e.g. health care ministers.

You have GOT to be kidding

Is this report authentic? Oh sweet pogo chocolate Jesus, WTF?

I'll try to put this into context

I've just completed a crash migration of around 500 XP machines on a hospital site (they were the remaining few from around 4000). Of that 500 around 60 could not be upgraded for various reasons of compatibility. Almost every machine had a different issue, some examples of what were found as incompatible:

Switchboard software

Security camera controlling software

Power / light management

Temperature control sensor suite in catering, and in pharmacy

Engineering stock control

Medicine-specific labeling software

Security badge printing

X-ray viewing equipment

Various medical scanning / imaging gear

Car park barrier management (and payments)

Numerous automated lab machines

Print server for ancient dot matrix printers (payslips)

Many bespoke databases written in the year dot using Access 97/ Delphi / Borland / VB5/6........

Ancient commercial software packages that are too expensive to replace - or for which there isn't a direct replacement (e.g. blood glucose analysis)

Hardware for which Win7 drivers don't exist (e.g. a couple of expensive Samsung high capacity scanners)

Call monitoring

Emergency pager message sending

Out of date cashier/till systems for which there's no replacement budget

and the list could go on and on............

Yes, some of these are easy to fix given a budget. But there's no budget. And the key point is only around 60 machines out of around 4000 are affected. I believe the basic premise of the story is wrong: most NHS trusts are well on the way to replacement, but are stuck with a hard core of machines which are an expensive PITA to do anything about

And virtualizing XP does not always work, especially when XP need to communicate with certain hardware - and most hypervisors are not OK with that.

The fault is the operating system

It's one job is to run applications. That's it. If every 3-5 years the version of os is changed and in the process breaks all the apps that ran on it before, what is the point of it.

So the point must be to make money for the os vendor. They have no interest whatsoever in building an os ecosystem that keeps apps working.

Like others have said - no-one gives a shit about the os. They want it to get the hell out of the way so they can use the apps they want to use with as minimal change as possible on how to start them.

Sainsburys supermarkets still have XP in their stores as well. Sure there are lots of copies still in use due to legacy software that will never get replaced.

Blame MS for writing a working OS with XP

They produce an operating system which runs programs.

A program I ran on my XP machine produces

Unsupported 16-Bit Application

then too much blurb for me to type in.

The language was extrememly popular in the 90s for database applications, I know of at least one large hospital system written in it (I knew the head programmer).

DOS 6.22

MUDOS fine

Real/32 fine*

WFW fine*

95 fine*

98 fine*

NT4 with some issues*

2000 fine*

XP fine*

Vista 32 with some major issues*

W7 32 with showstopping issues (NETBIOS)

W7 64 no hope

* running client server as well!

W7 32 was killed off due to killing off IP support for the program which provides client server access.

We do factory software.

So what did we do in the real world?

Vista - upgrade to XP, or run in a low resolution mode

W7 - tough, we tried, at first use your XP PCs, later test our new Windows software.

Huge DOS system, it took over 5 years to get into a viable WIN32 version. MS does not understand the software industry, you buy software to do a job, not because it is written in X.

So long as you have decent antivirus software and don't use them to access the internet I don't see the problem

Given that legacy web apps that require IE compatibility are a large part of this problem - and one that also exists in other big industries (engineering, finance) - it surprises me that no one has developed a browser based on open-source code that emulates IE but runs on whichever OS is convenient.

I think there is a subtle difference between 90% of NHS Trusts being reliant on a small number of xp machines to provide services which they have yet to migrate. Versus 90% of NHS Reliant on XP.

As someone who works in the NHS in IT I can say measures are always put in place where a decision is made to retain an XP device.

For example if a security door management system was designed to run on XP and the company wants £40,000 to update the software to run on a Windows 7/10 or Server, or a new system is going to cost £100,000. Measures can be put in place at a much-reduced cost. Things like network segregation be it physical or virtual. Hard/Soft firewalls etc.

As a public body we have to show value for money, and in some articles you praise the boxes which have been running in the corner for 20 years that no one has touched, yet kind of condemn the NHS if they even attempt to do something similar.

I would argue I know of no NHS trust which has XP as its desktop of choice for the majority of its users. Most desktops and other end user devices have already migrated to Windows 7 or beyond.

Win XP still has its uses

Don't diss Win XP as that is the latest windows I can run on my ancient Thinkpad with only 256 meg of memory but also critically has a now nearly nonexistent 9 pin serial port. Of course that bad boy never goes on a network and is only used for testing instrumentation through serial (need it to be mobile).

None of this surprises me. We have customers using old versions of an application we used to sell which ran in what used to be called Tandem NonStop hardware. Those same customers often keep those machines going by buying spares through eBay and the like. We are talking large high street businesses running relatively mission critical functions that have the ability to impact retail commerce throughout the UK if they stop working.

I sympathise with the NHS.

If you've got a platform that works stay with it. The continual changing of platforms is a major inconvenience or worse. For all that overall things get 'better'. In quotes because you need to consider by what measure.

I still use XP. And win10 and win7. Because I have hardware that literally needs XP. And I have other hardware (and software) that needs win7 or 10.

I get more speed out of XP on my Asus A8nSLI than I do from Xubuntu. I run an HP3150 on XP where there's no drivers for anything else.

Note there's no drivers because they've chosen not to make drivers. Not because it is some sort of impossibility.

So here at home I can see the planned obsolescence and I can see the pragmatic difficulties.

Now the NHS has massive investment in critical software. I'm in Australia, I've seen the expensive debacles you can have with massive software installations especially when new. We had an enormous fiasco when the govt commissioned online Unemployment/Job Seeking software.

And I used to be a programmer.

Porting software to new OS's can easily become a nightmare. It is not a trivial exercise. Simply changing software to fit new printers can be a nightmare. Simply incorporating new reports can be a nightmare. And these nightmares cost money and time and inconvenience thousands or millions of people.

Simply because software and hardware is old doesn't make it not fit for purpose. Move a lever by hand to achieve a task. Then put in a servo motor you switch on to move it. Then remote control that servo motor. Then put in a command centre that remotely controls many such motors. Then put in a command system that oversees everything.

That's modernising. That's technological development. That's how it goes.

At the bottom that lever is being moved, is all.

We need to ask if we need all the rest of the 'development' and in the calculation we need to factor in the costs - including human cost in all aspects.

We're not in a dreamland here. Look around, enquire, and see how many banks and such are employing software written in COBOL, still to this day. How many scientific institutes are using software written in FORTRAN. And so on.

XP is not 'wide open to hackers'. That's typical media beat-up hysteria.

NEW system are 'wide open to hackers' simply because they are new, untested for the most part. Hasn't history shown that very clearly? How many security updates in the first year of win10?

All systems are 'wide open to hackers' because that's a phrase without adequate definition. A system apparently impervious to attack today, lauded and applauded, looks ridiculous tomorrow after a backdoor or an achilles heel is found.

All systems are like cities on a plain surrounded by besieging armies, the hackers.

All systems benefit by being isolated. Firewalls. Partitioning. Circuit breakers. Parts of a system that are not required to have access to other parts are ideally cut off.

All systems benefit by RAID and similar philosophies.

Tremendously sophisticated and complicated software/hardware systems with online criticality benefit from long periods of stasis wherein they are studied and improved and protected.

Demand for constant change and thoughtless unnecessary 'improvement' brings uncertainty, unanticipated issues, complication and danger.

I am totally on the side of the NHS and its managers and particularly its IT people.

It is better that it remain firewalled behind a well understood XP system, doing what it does and doing it well with securities constantly strengthened while work goes on to develop a different and parallel system on some other platform, software and hardware, that can work in parallel once created and prove its validity by running without fault and impervious to hacker attack for six months or more.

That's the way to go.

It is a question of building another system.

Not a question of attacking this one or the people that have built it, run it, protect it.