UK Government Department Still Runs VME Operating System Installed In 1974

Qedward writes: The UK government's Department for Work and Pensions is on the hunt for a new £135,000-a-year CTO, with part of their annual budget of £1 billion and responsibility for DWP's "digital transformation" to oversee the migration of the department's legacy systems which are still run on Fujitsu mainframes using the VME operating system installed in 1974.

Comments

Well, no one will hack in

[jfdavis668]

Hackers probably couldn't even find a manual for one.

Re:Well, no one will hack in

[Anonymous Coward]

If they can't, then they have no Google-Fu....

http://www.fujitsu.com/uk/services/application/application-development/vme/

Re:Well, no one will hack in

[Z00L00K]

superNOVA - that reminds me of Data General Nova.

Re:Well, no one will hack in

[93 Escort Wagon]

If it's anything like my old Chevy Nova, it'll light up the night sky!

Re:Well, no one will hack in

[sexconker]

Good news, everyone!

Re:Well, no one will hack in

[ogdenk]

If it's anything like my old Chevy Nova, it'll light up the night sky!

After it explodes, the "frame" twists and the engine falls out?

Modern Technology

[Galaga88]

How many modern systems can anybody imagine still working and apparently doing what we need them to 40 years from now?

Re:Modern Technology

[jbolden]

Give me what that system cost in 1974 inflation adjusted dollars and I'll be happy to flip out a modern system every year. Using cheap less durable components with redundancy is a better strategy. I live in a 1830s house so I get the advantages of good quality construction. But if I were building a house I'd use 2014 cheap materials.

Re:Modern Technology

[Anonymous Coward]

Give me what that system cost in 1974 inflation adjusted dollars and I'll be happy to flip out a modern system every year. Using cheap less durable components with redundancy is a better strategy. I live in a 1830s house so I get the advantages of good quality construction. But if I were building a house I'd use 2014 cheap materials.

And you are part of the problem plaguing society. I am always amazed how buildings constructed thousands, or even hundreds, of years ago are still standing although often in a state of disrepair due to neglect. A modern-day castle might survive a century whereas the castles throughout Europe remain or at least remnants of their existence survive to this day.

Re:Modern Technology

[guruevi]

Depends on what you declare as modern. I have a Sun UltraSPARC box from the mid-90's which is still used to cross-compile things. Some things just stick around especially in government and research but also in established businesses, things are kept alive for decades because there is no funding to replace it and for most projects the people that maintain it are cheaper than establishing a new project.

This is mainly due to the inbreeding and subsequent incompetence on behalf of the people in charge of finance and IT but also incompetence on behalf of the other managerial staff to recognize that those people in charge are incompetent.

To this day there are very few projects in large organizations like governments, schools and old businesses that include line items for security, maintenance and replacement. Most managerial types still think that solutions are a one-time cost and that, like machinery, it will run fine as long as a mechanic puts some oil and parts in it and when they need something faster or better, they can simply sell or reuse the old 'machine' and recoup 50-80% of the investment in the 'faster' machine.

Re:Modern Technology

[__aaclcg7560]

I had a Black MacBook (2006) that ran for eight years until the CPU fan went kablooey. The only reason I didn't take it down to the Apple Store to get it repaired was the obsolete 32-bit CPU. Newer updates for installed software are now 64-bit only.

Re:Modern Technology

[jbolden]

Absolutely true. Today's structures will not stand the test of time. Even the concrete we use is steel reinforced which means in 2 centuries without ongoing care, it will be dust. Though Jerusalem where limestone is dirt cheap and still an excellent building material is an interesting counter example to that global trend towards throw away buildings.

Re:Modern Technology

[TheRaven64]

I am always amazed how buildings constructed thousands, or even hundreds, of years ago are still standing although often in a state of disrepair due to neglect

That's what's known as survivor bias. The only examples you see of thousand-year-old buildings are the ones that didn't fall down. The ones that collapsed within a decade are long forgotten.

A modern-day castle might survive a century whereas the castles throughout Europe remain or at least remnants of their existence survive to this day

And yet, in the village where I grew up, and near countless other villages in Britain, there was a hill with a raised mound on top, which was the only remaining evidence of the castle that stood there 900 years ago.

Re:Modern Technology

[TheGratefulNet]

http://www.eevblog.com/forum/t...

nice old classic tek test gear. highly in demand by collectors and those who appreciate good old fashioned engineering and build quality. the last of the 'repairable' tek scopes, pretty much (and even this is borderline repairable, with many custom chips).

still, a few new caps, a new battery backed nvram module and you have another 20 or 30 yrs left on this scope.

search that same forum for other old test gear (power designs (brand) power supplies are also built like tanks and run forever. I have 4 of them at home in my lab and they date from the mid 50's to early 60's. still hold their precision and would cost $5k to $10k today if you could even buy them.

I have audio gear that I personally built in the 70's and 80's that still runs fine (hafler amps, etc).

today, its hard to find things built to last, but it USED to be the norm "before your mother was born", so to speak.

Re:Modern Technology

[Anonymous Coward]

"That's what's known as survivor bias. The only examples you see of thousand-year-old buildings are the ones that didn't fall down. The ones that collapsed within a decade are long forgotten."

No one claimed that everything built a thousand years was great. It's just that it's obvious we CAN build long-lasting things, even with older technology, so we're a bit upset when confronted with the modern throw-away attitude.

", there was a hill with a raised mound on top, which was the only remaining evidence of the castle that stood there 900 years ago."

Without knowing *why* the castle is gone, I have no idea what your point is.

Re:Modern Technology

[Anonymous Coward]

How do you expect to stay in business if your product never ever fails?

Tire manufacturers developed a tire that could last the entire life of a car. They took the design and intentionally added inferior compounds to ensure it degraded at a steady rate guaranteeing an eventual failure.

IBM had a line of monitors that suspiciously died shortly after the warranty expired. It was discovered there was a resistor that was weak and would last just as about as long as the warranty.

Kirby vacuum was presented by their supplier with a belt that didn't burn out. It was only $0.10 more than the existing belt. They declined.

Re:Modern Technology

[TWX]

We've seen screencaps of 14-year uptimes on Cisco 2500-series Routers before, so I'd bet that a lot of networking equipment, if high quality to begin with, could make it that long, assuming that it's still doing what the users need.

Re:Modern Technology

[gstoddart]

Give me what that system cost in 1974 inflation adjusted dollars and I'll be happy to flip out a modern system every year.

Sorry, I'm calling complete and utter bullshit.

I've worked on enough legacy systems to know they didn't start off with some astronomical budget. They built it based on a set of requirements, coded it in house, and then it gradually expanded over many years of service.

Mainframe applications aren't sexy or glamorous, they're built on relatively simple interfaces, and slowly expand in scope over time.

They keep running because eventually they're woven into fabric of every other business process you have until they become something you can't trivially get rid of ... because every other damned thing relies on it even if it isn't obvious to the user. You end up having to replace everything

My experience with migrating from legacy apps says you'd churn out a half asses solution, which isn't compatible with the existing stuff, and which can't be made so, and which would eventually be abandoned as untenable.

You'd produce some solution which might be good if it didn't depend on throwing away every other system which touched this.

The vast majority of people who claim they could produce a functional replacement for legacy software in a short period of time have never been involved in that kind of process.

If it was easy, they'd have replaced it by now.

The problem with looking for a "track record of transitioning a large enterprise from ageing mainframe technologies to next generation web, social, mobile cloud, Big Data and deep learning technologies" is that it's a set of requirements written by idiots who don't want to replace the system, they want something completely different which will involve re-tooling everything else that touches this existing system.

Put your money where your mouth is, apply for the damned job.

Re:Modern Technology

[Anonymous Coward]

And I'll show you a bunch of dead families.

Re:Modern Technology

[gstoddart]

That's what's known as survivor bias. The only examples you see of thousand-year-old buildings are the ones that didn't fall down

Except of course the overwhelming majority of stuff we make today would have no chance in hell of lasting 1000 years.

There are literally concrete structures created by the Romans we couldn't even begin to recreate.

This isn't just a survivor bias, it's stuff which was built incredibly well, and using techniques we can't reproduce. Even modern engineers admit that it's more than just luck that those things lasted.

Re:Modern Technology

[Anonymous Coward]

Probably, the castle was abandoned because a new castle was built in a better location, or it was simply no longer needed to defend the local area. The local villagers then stole the stones of the castle piece by piece, to build their own homes, until nothing of the original structure was left.

Re:Modern Technology

[kenh]

My experience with migrating from legacy apps says you'd churn out a half asses solution, which isn't compatible with the existing stuff, and which can't be made so, and which would eventually be abandoned as untenable.

This is something the federal government proves over and over again... The current tax code is a staggering collection of decades-old COBOL code, the air traffic control systems until very recently ran on vacuum tube computers, and the FBI has tried, and failed, repeatedly to transition off a collection of mainframe tools for agents onto something more modern and flexible, wasting billions of dollars in the process.

Re:Modern Technology

[__aaclcg7560]

I learned electronics on those on old Tek scopes in early 1990's. One day I slapped the scope in the side because it wasn't working right. My instructor came over and told me to never ever slap the scope under any circumstances. A moment later he slapped the scope and the problem went away. Go figure.

Re:Modern Technology

[mrbester]

Because, as a student, if you hit it and it breaks you did something dumb and reduce the number of units for the class to use. However, as an instructor, if he hits it and it breaks, it was due for replacement.

Re:Modern Technology

[csnydermvpsoft]

What basis do you have for the claim that we "couldn't even begin to recreate" those structures? There are certainly some ancient structures for which we haven't figured out how they were constructed with the technology available at the time, but nothing that we couldn't reproduce with today's technology.

The sticking point isn't technology—it's economics. A large portion of recent development has been around cost-effectiveness. This is why we're able to have so many more material possessions, even in the face of stagnant wages (for most classes). Of course, many (including myself) would argue that we've gone too far in this direction at the expense of durability, but that's an economic choice we've made. Look hard enough, and you can find any product that meets your durability specifications—if you're willing to pay the higher price.

That being said, I do agree with the sentiment that there is more than survivor bias at work. My house was built in 1916, and has an unusually open floor plan for its age. Lacking CAD, the builders accomplished this by massively overbuilding—the floor joists (with are already quite thick) rest on beams comprised of four 2x10's laminated together. Despite its age, this house feels more solid than just about any other wood-framed building I've been in. I have no doubt that if it were placed alongside a newly-constructed house and both left to nature, that the 99-year-old house would remain intact longer.

Re:Modern Technology

[Ed Avis]

I think you're missing the point. It is not about hardware durability. The original hardware installed in 1974 has long since been replaced (probably several times over). It is the software that costs money over the long term - hiring programmers to maintain it. And it is the software that is the reason the system hasn't been replaced with something else.

Re:Modern Technology

[oldmac31310]

Those hill forts were made from wood. Typical pre-Roman celtic structures. Wood decays, stone not so much.

Re:Modern Technology

[gstoddart]

Well, the problem happens when some technology evangelist or manager who doesn't know a damned thing about the existing system claims it's easy to migrate it to modern tools.

And neither the customer, nor the guy saying it's easy, has the barest clue about just how many other things depend on that system, and nobody can fully enumerate the functionality and corner cases.

And then you end up trying to shoe-horn a purpose built piece of software which has ran fine for decades into a modern paradigm, and realize you are failing utterly.

Because the modern tools usually simply can't accommodate all of the rules and logic in that system. They can't be cajoled into having enough flexibility, or simply can't do the same task.

People consistently underestimate just how well these systems do their job, and just how many little corner cases and integration points have been woven into them over the years. The platform is no longer elegant, or easy to explain, but it just keeps working. But dozens of other things rely on it, and if you change the underlying thing you rebuild everything else.

I've been on several projects trying to replace stuff built in the 60's and 70's -- and I wouldn't go near another one without very loudly saying how much risk is involved. Hell, even a system which has been around only since the 90s might be non-trivial to migrate away from -- precisely because in the 90s people were still building much more purpose-specific software.

It's a catch 22 ... they get increasingly difficult to maintain, but they sometimes are impossible to replace.

As I said, if it was easy to replace these systems, it would have been done already. Discovering just how difficult this can be has been the downfall of many a naive person who claims it's an easy thing to do.

Re: Modern Technology

[thetoadwarrior]

That's BS. There are plenty of homes missing now from that era and castles that went to shit too. Saffron Walden in the UK has a castle but you'd probably not notice since it's just a small pile of remains. Famous structures like Ely Cathedral still stand because it gets repaired on an annual basis. There are plenty of bits on it that are only a couple years old at most. They've even got scaffolding all up the side it right now. They've also been digging at the foundation. You underestimate how much effort goes into keeping old buildings going. I've lived in Victorian homes and new build and the maintenance costs are like night and day. My current home might not be around in 500 years but I don't care and why should it? It's not special and there will be far better ways to build homes in that time. Maybe by then all those old Victorian homes will finally be rid bad technology like lead pipes.

Re: Modern Technology

[thetoadwarrior]

Of course we can. We repair the things every year. The fact you don't appear to notice shows how good we are at it.

Re:Modern Technology

[itzly]

A novice was trying to fix a broken Lisp machine by turning the power off and on.

Knight, seeing what the student was doing, spoke sternly: "You cannot fix a machine by just power-cycling it with no understanding of what is going wrong."

Knight turned the machine off and on.

The machine worked.

http://en.wikipedia.org/wiki/Hacker_koan

Re:Modern Technology

[gstoddart]

Well, it's the way I've always heard Russian engineering described ... in the absence of finesse or the assumption of a skilled operator, you build the heck out of it, like you said. Take all of your tolerances and double them just to be sure. If you don't know your tolerances, build it as heavy as you can manage.

The techniques used to build stuff out of stone had been learned over a very long period of human history, and was used to build stuff you expected to last forever.

Roman roads, or some Roman concrete structures have lasted far longer than any modern equivalent ... so much so that people are trying to figure out some of the process. Because the Romans made piers and bridges which still stand in salt water, and we really can't come close to that.

My guess, if humankind were wiped out tomorrow much of our engineering wouldn't last a hundred years, let a lone a thousand. And, in fairness, we build for different purposes and with different constraints.

I firmly believe that a tiny fraction of structures would last long enough to suggest a survivor bias ... in fact, I suspect long after the modern stuff had fallen apart some ancient stuff would still be standing.

Re: Modern Technology

[gstoddart]

Tell you what, build a concrete structure which will be left in salt water for the next two thousand years and see if it lasts.

The Romans could do it, we still can't.

Say what you will, but there are actually ancient construction techniques we really cannot duplicate.

They may have had what we consider to be primitive tools and materials, but there are things which were built that modern engineers are trying to understand how it was done -- precisely because they know we can't duplicate the feat.

It's only in the last year or two that people are starting to understand some of the mechanics of Roman concrete.

But it far exceeds anything we can make these days.

Re:Modern Technology

[jeremyp]

You could probably flip out a modern system every year for less than the annual support and maintenance contract that the DWP is probably buying.

Re:Modern Technology

[DerekLyons]

today, its hard to find things built to last, but it USED to be the norm "before your mother was born", so to speak.

Well, no. While it's true that the race to the bottom has increased over the last few decades - things pretty much have always been "built to sell", and if they lasted that was a bonus rather than a design feature.

Re:Modern Technology

[jeremyp]

using techniques we can't reproduce.

Guédelon Castle

Re:Modern Technology

[jeremyp]

Without knowing *why* the castle is gone, I have no idea what your point is.

Usually, if it was a stone castle, the building materials were robbed out to make new dwellings. The reason that people could do that is because the owners abandoned them as being shit places to live.

With buildings, as with other man made items, technology moves on. Generally speaking, a house built now will be more comfortable, easier to heat and more suited to modern life styles than a house built 50 years ago. Who cares if the old ones fall down? If you are going to knock it down and replace it with something better, money spent making it last a millennium is wasted money.

Re:Modern Technology

[Lab Rat Jason]

Queue obligatory Monty Python reference:

When I first came here, this was all swamp. Everyone said I was daft to build a castle on a swamp, but I built in all the same, just to show them. It sank into the swamp. So I built a second one. That sank into the swamp. So I built a third. That burned down, fell over, then sank into the swamp. But the fourth one stayed up. And that's what you're going to get, Lad, the strongest castle in all of England.

Re: Modern Technology

[thetoadwarrior]

Like anything else some things fail straight away and some go on forever. If there was something genuinely better about their concrete then all their stuff would have lasted. But everything from paintings to buildings that we want to keep rely on regular maintenance. Yes the Romans have some concrete in water that thanks to materials like volcanic ash last longer but of all the things they built most of it doesn't exist now or requires maintenance. Compared to all things built around the globe a could structures in Italy are insignificant and can't be used to prove all old building methods were superior.

Re:Modern Technology

[mspohr]

You only see the old buildings which were built well because the cheaply built ones have all fallen down. It doesn't mean that there weren't cheaply built buildings in the past.

Re:Modern Technology

[sjames]

The problem is that the economics don't really line up for the consumer. All we hear is about how spending an extra $1 on the most common point of failure in a $500 appliance would double the price. That's some really "special" math!.

People buy cheap because the expensive products turn out to be the same as the cheap product except for branding too often.

There is a such thing as overbuilding, but for the most part we are far from it today.

Re:Modern Technology

[sexconker]

You could probably flip out a modern system every year for less than the annual support and maintenance contract that the DWP is probably buying.

And you'll still have to pay the annual support and maintenance costs for the software and environment.
Hardware and associated cost isn't the issue.

If by "system" you mean hardware and software, then you're a moron. Not only could you not "flip out" a new "system" every year, you sure as fuck wouldn't want to. You wouldn't even be able to get a complete fiscal year done on one system before you toss it out and jerry rig everything onto the new system.

Re:Modern Technology

[Anonymous Coward]

Using cheap less durable components with redundancy is a better strategy.

it's not a better strategy, it's a peculiarly Chinese strategy. That's why most of their stuff is junk - it's on purpose.

Are you factoring in the costs of downtime and migration every year? No, you're not.

Re:Modern Technology

[sjames]

It certainly can be done. The problem is that it requires highly skilled developers committed to a multi-year, multi-phase project. It tends not to be done because management is rarely willing to commit to such a project and isn't willing to pay to have a full analysis and scope put together. No sane developer is going to be willing to do such a complex and detailed analysis for free up front knowing how likely it is that the paying work either won't get done or will be farmed out to code monkeys working from the free analysis they just did.

Re: Modern Technology

[Blaskowicz]

What I recently learned about is how asian temples at least in Korea are teared down and rebuilt every 20 years to 40 years (I don't remember and I don't know if it varies). Wooden parts fit together (no nails, bolts etc.)
There is not the western obsession with preserving extremely old buildings and parts of building in their original state for as long as possible, but instead cyclical continuity like the fable of the hammer whose head and handle were replaced mutiple times each.

Re:Modern Technology

[B33rNinj4]

Very few, sadly. They aren't designed to be long-running workhorses.

Re: Modern Technology

[cnkurzke]

The only "permanent structures" we will leave to our next millenium ancestors will be:

1. trash dumps
2. nuclear powerplants (will those be the equivalents of castles and cathedrals of the 31st century? Oh, look at those complex curved titanium barrels they were able to produce with their ancient primitive tools!!)

Consider this, When we talk about nuclear storage facilities, we intend to build structures which will last tens of thousands of years. Good luck.

Re: Modern Technology

[Anonymous Coward]

The Romans had what I'd call Engineer-Builders. Most engineering done today happens in an office and we have generally accepted and documented the "best" way of doing things, which tends to be cheapest and most efficient. In Rome the construction techniques were closely guarded secrets, never really published and there wasn't such a uniform approach to building, hence by constantly redefining what they were doing they have seen many permutations we have not. Their engineers had real world experience with aggregates and different techniques of forming concrete we simply do not have and will not have because no one is going to experiment with it. They had a lot of ingenuity and experiment often, what remains and was kept are the projects that succeeded, without seeing the failures we don't have a good path to follow when attempting to understand how some of this was built. I disagree that we cannot replicate what they have done today, I think largely we can but not using the methods and techniques that they had.

Re:Modern Technology

[jbolden]

I haven't seen that. I can see how an additional 300 $1 parts might double the price though. This started as a discussion about Apple, which is a good example. The extra $.20 per screw, $5 for how the glue is applied, $25 extra for this part... turns into a few hundred extra in cost.

Re: Modern Technology

[Anonymous Coward]

My house is from the 1890's, the problems that I've had with it are from the modifications done in the 1970's. The modifications were the right thing to do like wider windows and wider doors. The old lintils were stone but the new ones were reinforced concrete, the steel in them had rusted and had cracked the concrete. The replacements I have put in place have galvanised steel so should be good for long enough but the bricks won't last forever (I imagine an ongoing extremely slow chemical change) but still probably more than a lifetime left in them so fine for my needs.

Re:Modern Technology

[jabuzz]

The castles where mostly abandoned because they became obsolete. A canon will punch a hole very easily in most castle walls. Secondly the majority of the castles in the United Kingdom are close to the border between Scotland and England. Even prior to the union of the crowns in 1603 the border disputes had faded away and you no longer needed a castle.

To be honest unless it was deliberately pulled down, almost if not all the castles in the United Kingdom still survive if only as ruins. The Victorians pulled lots down unfortunately.

The churches have survived better because they tended not to fall out of use. Though a good few where rebuilt bigger.

Growing up I could look out my bedroom window at a castle over 900 years old and a church over 1000 years old. Within walking distance was a Roman wall that was close to 2000 years old.

Re:Modern Technology

[jbolden]

I and the GP were talking about the cost of the mainframe not the application layer at all.

As for legacy conversions click on my link I've done dozens of them and quite successfully. Absolutely capturing business rules is a big deal. And frankly most mainframe applications were labor inefficient in their construction. That doesn't mean $100m worth of programming can be replaced for $1m but it can be replaced for $10m.

Re:Modern Technology

[jbolden]

I think the GP was talking about the hardware. If not then on every Unix box there are routines still running from 1974. For example VI/VIM is an extension of ED (ED -> EX -> VI -> VIM) which is 1971.

Re:Modern Technology

[King_TJ]

Yeah, ANYTHING is possible given enough talent, dedication and funding.

In reality though, even these multi-year, multi-phase implementations tend to go way over budget and fail to yield everything promised.

I've seen it happen, first-hand, when a company I worked for decided to implement a new ERP system and phase out a number of other applications and processes. They DID shell out the money to get the analysis done properly, but the problem really came in with ability for the new software to perform as intended. Bugs were found during the roll-out, but the team doing the implementation had too many layers internally to get those bugs corrected in a reasonable and timely manner. (EG. They'd have guys flown out and paid by the hour to give employee training on the new software and how to do something with it. During that process, they might run into a glitch -- but as trainers, they weren't really capable of fixing it or even having a direct contact with someone who could. They'd just gloss it over and move on, promising to "make a note of it".) If they even remembered to pass the note along about the bug, chances are it went to some team responsible for collecting the reports from other people on staff with the company. So now the bug was perhaps poorly or incompletely documented in some kind of bug tracking database. When will the actual developers get to addressing that bug? Hard to say. As likely as anything, they might claim it wasn't reproducible with their environment, or ask for more information. But since our people didn't have access to the database directly, we'd never be able to directly reply with that additional info or help finding the bug.

There were times my boss fixed problems in their system himself, because he knew more than the average person about the back end database it used. When that happened, I remember him telling their people about the fix and it was pretty much ignored, as in "Well, that's great then! Our team should eventually figure it out too and roll it into a version upgrade...."

For millions of dollars spent, I don't think you should have to work late nights finding your own fixes for their defective code ... but that's how it went down.

Re:Modern Technology

[jbolden]

I actually people on the IRS system. It isn't staggeringly complex it is moderately complex. They also have a lot of dysfunctional management, lack of proper skills, unclear budgeting all the way up, changing objectives, and unnecessary requirements all of which drive the costs sky high.

Re:Modern Technology

[disambiguated]

Even given the short(er) lifespan of modern buildings, most of the buildings we tear down are torn down while they are still structurally sound and useful. They just aren't useful enough. We tear them down because they are not the building we want on that site any more. For example, consider single family houses in the middle of a crowded city. They don't have modern energy efficiency, safety or ease of maintenance, and they don't house enough people to justify the forgoing the alternative uses of the land they are on. A few rich people will live in them anyway, an we keep a few around as historical sites, but the rest we tear down to build something we want more.

Computers are even worse. If we wanted to, we could build a smart phone that would last a century but nobody wants that. It won't even be 5 years before we won't want the phone we have, because there will be better ones by then. Centuries-old buildings were built in a time when technology and society itself were relatively static. With today's pace of technological and societal change, it doesn't make sense to build much of anything to last that long because what we will want in a few hundred years or even a few decades is unknowable. Forget about thousands of years. Anything we build today is not going to be what we want anymore long before that. I suppose that some day, when the pace of change stabilizes, we will start building more permanent things again. In the mean time, there's a balance to be struck between construction/manufacturing costs, and what we expect the thing's useful lifespan to be.

Re:Modern Technology

[sjames]

You should disassemble failed products more often and ask the questions. BTW, screws don't cost anything like $0.20 in the quantities that Apple would buy unless you use that very "special" math.

You may have lost the thread, yours is the first mention of Apple in this thread and TFA is about a UK government mainframe from the '70s.

But going with apple, considering how touchy they are about humidity, a dab of silicone putty behind the sockets would have been a good call for a cost of nearly nothing.

Re:Modern Technology

[sjames]

That's down to the quality (or lack) of the contractor. It sounds like the typical crap you get from one of the big vendors trying to shoehorn an expensive "off the rack" product into a very custom environment. It probably would have worked better to directly hire the developers.

Re:Modern Technology

[clawsoon]

That's a great point. The evolution of life has worked the same way. There are some proteins which are interacted with and depended on by so many other proteins that changing them would be catastrophic; I happened to be reading about tubulin and actin today:

The likely explanation is that the structure of the entire surface of an actin filament or microtubule is constrained because so many other proteins must be able to interact with these two ubiquitous and abundant cell components. A mutation in actin that could result in a desirable change in its interaction with one other protein might cause undesirable changes in its interactions with a number of other proteins that bind at or near the same site. Genetic and biochemical studies in the yeast Saccharomyces cerevisiae have demonstrated that actin interacts directly with dozens of other proteins, and indirectly with even more (Figure 16-15). Over time, evolving organisms have found it more profitable to leave actin and tubulin alone, and alter their binding partners instead.

The more complex a system becomes, the more it gets life-like constraints like this.

Re: Modern Technology

[disambiguated]

If there was something genuinely better about their concrete...

A quick search shows there really was something better about their concrete:

Ancient Roman Concrete Is About to Revolutionize Modern Architecture

Discovery of 'Lost Recipe' for Ancient Concrete Provides Foundation for Future Cities

The Riddle of Ancient Roman Concrete

Re:Modern Technology

[Anonymous Coward]

So let's get rid of survivors bias. Out of 180+ cathedrals built in France over th last millenium, only 1 suffered partial collapsed (Beauvais) (also 4 were destroyed by war, but this is not the problem here).

Re: Modern Technology

[Anonymous Coward]

Using a performance criteria which was only met accidentally by Roman concrete and still has no bearing to practical uses of concrete is stupid.

If we wanted to design a concrete for millennia range salt water resistance we could do that better than the Romans. We'll just stick to concretes which are easily pipeable, are used with (tensioned) rebar and have strength the Romans couldn't even dream of ... because that has more practical applications.

Re:Modern Technology

[Guy Harris]

the air traffic control systems until very recently ran on vacuum tube computers,

Presumably you're using hyperbole to say "they're old"; back in the late 1960s, a system based on modified System/360's was installed, and those were at least two generations of IBM computer away from vacuum tubes. Those have been replaced at least a couple of times since then.

Re:Modern Technology

[Kittenman]

Well, the problem happens when some technology evangelist or manager who doesn't know a damned thing about the existing system claims it's easy to migrate it to modern tools. .....

Sorry, disagree. Disclaimer: I work on legacy systems, and they've provided me with $$$ for many a year. But for 'legacy' I prefer 'industrial-strength, production' that happens to run on a mainframe (or Enterprise Server - whatever).

But I digress... these old systems are massively tailored (I'm on one now...) and have been doing the job for decades. But saying that they're not easy to convert or replace is obvious (if you mean it's difficult) or wrong (if you mean it will never happen). Modern IT tools and techniques (pick your favourite ones and insert in these brackets) can solve complex business problems. Might take a while to work out why the original did that thing it does on March 14th if it's a Tuesday... but that's tailoring.

If it's a problem that can be solved by IT, then IT can solve it.

Now another question ... should the existing system be migrated or totally replaced? I've got four or five old clunkers here that make me weep, and I'm leaning towards replacing the functionality with nice, modern easy-to-understand stuff. Keep the business process, replace the code.

BTW, where do I apply for this GBP135,000 job?

Re:Modern Technology

[Cyberax]

We can build structures that can last for millenia. That's easy. For instance, https://en.wikipedia.org/wiki/... utterly dwarfs pyramids and it's expected to last longer.

It's just that with rare exceptions people do not WANT such structures. A typical house could probably be rebuilt much cheaper in 50-70 years and with better safety. For example, the house where I live right now was built 5 years ago - it certainly doesn't look as 'solid' as a renovated Victorian house where I'd been living earlier. However, the new house is an order of magnitude better insulated so it requires much less heating and cooling. Its sound insulation allows me to run on a treadmill listening to music while my girlfriend is sleeping downstairs. The electrical wiring allows me to power a coffee maker, electric kettle and a dishwasher without burning down the whole block and so on.

But I totally expect that in 100 years it'd replaced with a house made from paper-thing carbon-nanotube reinforced smart self-forming concrete with built-in automated furniture extruders. While people will be telling each other that nobody can make anything lasting more than a week, unlike people in 2010 who could build stuff lasting for _decades_.

Re:Modern Technology

[whoever57]

I had a Black MacBook (2006) that ran for eight years until the CPU fan went kablooey

In my house we have, not one, but two working Thinkpad T43s. One of which is used daily (in fact it is rarely turned off).

Re:Modern Technology

[Anonymous Coward]

Except the people who you talk to will be the people who tell you HOW they Believe the business is run. Talk to the poor sods like me who pee that sort of old legacy code running then we will tell you a different story of the process are really down, Except our stuff isn't sexy because it isn't run on the previously mentioned execs desktops but over night and all they see is the GUI front in powerpoint etc that runs of that data.

Re:Modern Technology

[Anonymous Coward]

I got a car does just that.

Re:Modern Technology

[ogdenk]

Because the Romans made piers and bridges which still stand in salt water, and we really can't come close to that.

Actually, that's not true. We've figured out how to make Roman Concrete but it has it's own limitations and is less safe to work with in some ways and much harder to make in vast quantities cheap. Our construction processes and material shipping methods would have to change considerably in order to use it in a widespread fashion.

Re: Modern Technology

[thetoadwarrior]

And likewise if you do an image search for roman concrete you can see it still wears away. It's more flexible and more resistant to salt but it still wears away and in fact is apparently weaker. It's a trade-off and all those nice buildings in Italy still require maintenance.

Just look at the colosseum, you can clearly see newer touch-up work. http://upload.wikimedia.org/wi...

Re: Modern Technology

[thetoadwarrior]

Well that and Mt. Rushmore among other far more modern structures and if the old roman, victorian, or whatever stuff is still standing then it will be thanks us. In fact in another 2000 years arguably there may be no original pieces left on some if not all structures. But who am I to break the circle jerk of how things have gone to shit and everything was better in the old days.

Re:Modern Technology

[Anonymous Coward]

When a politician makes a suggestion like that, I like to counter by talking about something they do, more or less, understand: laws.

How would they feel, I ask, if someone offered to rewrite the entire legal code of the country from scratch? Subject only to the proviso that the moment the new system goes live, every pre-existing law, contract and precedent becomes obsolete and inadmissible in any future proceeding. Oh, and the change is irreversible.

Re:Modern Technology

[K. S. Kyosuke]

Yeah, and try to install new utilities and services in ancient buildings. As monuments, they're nice. But at the time of residential buildings changing at breakneck speed (compared to the centuries before), building something that will last the next thousand years for actual use seems a little bit meaningless.

Re:Modern Technology

[Anonymous Coward]

Ancient cement/concrete.
Cement used today is different that tat used by romans.
Our cement lasts for 100-200years.
The one which was used by Romans lasts a lot longer.
And noone knows how they did it.

Google it and you will know more about it.

And while I understand both sides of this discussion I want to point out simple argument.
Structures created in past were more durable so parents could pass it to their children so they don have to build it and spend a lot of effort for it.

Note that with apropriate maintenance (which is not expensive and consists mostly of avoiding water damage) wooden cottages are quite comfortable and can last for more than 200years.

So in past when you grow up you could just get a house for free. And now especially in developed countries everyone needs to build their home and sometimes do it a couple of times...

old != bad

[AndroSyn]

My money is on this VME system being around for another 20 years while the mess of Java and Oracle(you know they're going to use Oracle). It'll be overpriced, late and won't actually work.

Just because something is old, doesn't mean it needs replaced. In short, why not just upgrade the mainframe?

Re:old != bad

[AndroSyn]

That should be while the mess of Java and Oracle is implemented...

Re:old != bad

[Shinobi]

Nono, like other big IT projects in the UK, it will be using "the very latest in Agile know-how", and cost 3 times as much as any clusterfuck that involves Oracle, take 50% longer, and spread 300% more blame on "old fossiles"....

Disclaimer: Had to interface with a EU project under UK IT auspices last year.... Painful....

Re:old != bad

[Anonymous Coward]

Because eventually, with stale software systems, maintenance costs blow up and making any changes ends up being nearly impossible.

Re:old != bad

[AndroSyn]

And so begins the IT cycle of reinventing wheel after wheel and learning nothing in the process.

Re:old != bad

[ranton]

Just because something is old, doesn't mean it needs replaced. In short, why not just upgrade the mainframe?

I have no idea how common VME developers are, but when dealing with legacy systems you do have to worry about being able to find qualified people to work on your software. Not only are the skills rare, but most people are going to be wary about pigeon-holing their career by focusing on such a obscure system. You will either have to rely on sub-par employees or pay well over market rates.

Hiring expensive employees / consultants may still be desirable over a risky migration, but the expense (either in salary or in low quality employees) shouldn't be ignored.

Re:old != bad

[Anonymous Coward]

Just because something is old, doesn't mean it needs replaced. In short, why not just upgrade the mainframe?

Because that won't pay the salary of a bunch of brogramming hipsters?

Re:old != bad

[Anonymous Coward]

Arguing for the modernization of a 40 year old system is not reinventing the wheel.

And learning nothing? Software development has changed dramatically since then.

Re:old != bad

[Wintermute__]

Changed? Sure.

Improved? Well, the jury is out.

Re:old != bad

[Anonymous Coward]

s/Software development has/Management fads have/

FTFY

Re:old != bad

[Anonymous Coward]

Fucking clueloss anonymous weasel. Ever heard of Y2K or 2038? Those errors are in the CODE, and they're time dependent. Code with 32-bit, epoch based dates will not work 100 years from now (or 25).

Re:old != bad

[Anonymous Coward]

My money is on this VME system being around for another 20 years while the mess of Java and Oracle(you know they're going to use Oracle). It'll be overpriced, late and won't actually work.

Or someone will come up with the brilliant idea to use a "standard system", and undoubtedly they'll choose SAP.
We all know what follows...

Re:old != bad

[Anonymous Coward]

Oh yeah. Because new, established, and verified technology and methodologies over the last 40 years are all just fads. Got it.

Re: old != bad

[Anonymous Coward]

He could very well be an 18 year old high school dropout. That means he's a professional Ruby and/or JavaScript "engineer". That, in turn, means he has no idea what a "bit" is, he was a toddler during Y2K, doesn't know about UNIX time, and hasn't worked on anything more complex than a simple blog system.

Re:old != bad

[Anonymous Coward]

Fucking clueloss anonymous weasel. Ever heard of Y2K or 2038? Those errors are in the CODE, and they're time dependent. Code with 32-bit, epoch based dates will not work 100 years from now (or 25).

And Y2K was little more than a marketing opportunity for "Y2K experts." I do agree that the errors are in the CODE and are time dependent, but a new implementation won't have fewer errors. That just a fact.a new implementation cannot stand up to over 40 years of refactoring.

Re:old != bad

[fahrbot-bot]

I have no idea how common VME developers are, but when dealing with legacy systems you do have to worry about being able to find qualified people to work on your software. Not only are the skills rare, but most people are going to be wary about pigeon-holing their career by focusing on such a obscure system. You will either have to rely on sub-par employees or pay well over market rates.

These days, I'd take more job security over "over market rate" salary and fancy perks. As long as I can pay my bills, the work is fairly interesting, my teammates and managers actually appreciate me and value my skill set - and I'm not micro-managed to death, I'd be happy to pick up some VME skills.

I don't consider myself a "low quality" employee, I just don't have fancy tastes or a tech-toy habit to feed (nor do I have kids). Now 51, I have always lived responsibly, am debt-free (with enough savings/investments to survive another 60+ years w/o a job - not counting SSI). I currently work for a large defense contractor on a three-person team using about 10 programming languages for about 500k lines of code on both Windows and Unix/Linux systems. I have previously worked for the NYT and at NASA.

Overly highly paid and/or low quality workers are not the only options - for many jobs.

Re:old != bad

[kenh]

The code that truncated years to two digits (Y2K bug) or relied on the epoch clock (2038 bug) were not designed to work past year 1999 or 2038, try were built with limitations that were accepted by the developer, testers, and users - even if they didn't understand the trade-off they made or the self-imposed self-destruct mechanism they hard-coded into their program. In effect, they were designed to fail, so I guess one could argue they 'work as defined', they were just poorly defined.

Re:old != bad

[bobbied]

Changed? Sure.

Improved? Well, the jury is out.

If I was on the jury, I'd vote that it's NOT better in the last 40 years.

But.... I'd also say it's not any worse either. Software development as an engineering process hasn't gotten better or worse since it was invented.

Re:old != bad

[Anonymous Coward]

Did you work on the NYTProfiler? If so... wow, nice piece of software!

Re:old != bad

[gstoddart]

They were NOT designed to fail.

Nobody conceived they'd still be running that long, but if they'd been designed to fail they would have done so by now. Having your usable life be decades longer than expected is not evidence of bad quality.

If code written in the mid 70's is still running 40 years later, that's kind of the opposite of doomed to fail. Code written in 1974 was old enough to vote or buy alcohol by the time Y2K came along.

I predict very little code written in 2015 will be running in 2055.

Re:old != bad

[Anonymous Coward]

I have no idea how common VME developers are, but when dealing with legacy systems you do have to worry about being able to find qualified people to work on your software. Not only are the skills rare, but most people are going to be wary about pigeon-holing their career by focusing on such a obscure system.

I used to work in the UK civil service and I used their VME-based system, which is called JSAPS -- Jobseeker's Allowance Payment System. I just had a quick google and was amused to find an (expired) job vacancy which didn't just specify VME development experience: they were only interested in hiring someone with "Experience in design of the JobSeekers Allowance Payment System".

Re:old != bad

[plopez]

The software is the easy part. Figuring out how to replace 40+ years of business knowledge stored in the older software is the hard part. ANd why most upgrades fail in terms of massive cost over runs, damage to business processes, and slipped schedules.

Re:old != bad

[sexconker]

Code doesn't breakdown, if it works now it will work 100 years from now if the hardware is still running.

You've either never supported someone else's Java application or you're a straight troll.
It's rare for a Java application to work 100 days after it was last working with all of the timebombs they build in now.

Re:old != bad

[cactopus]

What we don't know from the original article because there is a bit of sensationalist language in it is if the assumption that the VME system was installed on ICL hardware from 1974 and has existed and functioned ever since without update. Keep in mind that the current product is software-only running on x86/x64 gear from Fujitsu and is called OpenVME. There have been several intervening generations of hardware and software from 2900 to Series 39 to Trimetra/Nova and Supernova. How do we know from this article that they didn't buy or upgrade one or more times yet keep their apps basically the same (what works) on the same development platform? I don't believe the original 1974 hardware unchanged could cope with or keep up with business load for that department... I/O beast that it might have been for the time. It was probably upgraded/expanded in the 80's and 90's at least once or twice. The 70's disk drives alone wouldn't hold much (200MB and 40MB platters). That might not be a problem for a simple OS image, system files, and their apps but I think the data created would eventually start to fill available storage pretty quick.

Re:old != bad

[Antique Geekmeister]

That sort of thing pays a lot of my salary and consulting fees for my group. Code even six months old can be nightmarish to dissassemble and replace when the original programmer is no longer available, or doesn't care enough to remember why they did things. I find myself treasuring the experience that helps me rember _why_ we did things certain ways before a new application or operating system even existed.

Re:old != bad

[Anonymous Coward]

VME includes the binary date routines ICL9LGGDATEC2B AND ICL9LGGDATEB2C which convert dates to and from a four byte binary number and character string. they Predate VME and were written with the 4,100, 400 and 4000 year leap year rules and the Julian Gregorian changeover (localised if the system was being installed in a country where the changeover date was different). The Y2K problems on VME kit were purely down to sloppy design where analysts wanted to read dates on file dumps rather than having to use a macro to convert that word to a date. Idiots also used 6 character dates to save space instead of that 4 character word.

It would be nice...

[Anonymous Coward]

...if the new system's requirement was to run for another 40 years. Somehow I don't think that's going to happen. I wonder what kind of system that would be. Probably not something based on a Microsoft OS...

VME is not an operating system.

[Anonymous Coward]

It is a bus and connector specification. Motorola did have a proprietary OS called VersaDOS that
was often used on VME systems, but we also ran Unix and embedded OSs like PSOS.

Good times.

Re:VME is not an operating system.

[Anonymous Coward]

Shut up idiot.

https://en.wikipedia.org/wiki/...

Re:VME is not an operating system.

[Guy Harris]

It is a bus and connector specification.

"VME" can refer either to the VME operating system or the VMEbus.

Does it still work?

[91degrees]

If so, why fix it? What are the tangible benefite of a new system?

Re:Does it still work?

[Lunix Nutcase]

Government contracting money.

Re:Does it still work?

[Dr_Barnowl]

Depends what the requirements are.

Usually, this sort of thing happens because requirements are changing faster than the old system can be maintained to keep up.

I wouldn't be surprised if this is to help automate the swingeing series of "sanctions" that are carried out to remove the benefits from job seekers in this country.

Things like suspending their payments for...

* Being late for an appointment at the job centre (by approx 2 minutes).... because they were attending a job interview
* Not attending a job interview
* Applying for 6 jobs one week, and 3 jobs the next, and not realising that the directive to apply for 4 jobs a week is not met by an *average*
* Applying for jobs on Monday and Friday, then being sanctioned because the accounting is done Tuesday and the count of jobs on Monday wasn't high enough

Re:Does it still work?

[TheGratefulNet]

new systems will have backdoors and privacy leaks.

older systems did not have to conform to the 'we demand to wiretap any computer or network box' bullshit.

and if its been running so long, its obviously well designed and they know what to expect if things fail. today, once code compiles, they ship that crap and worry about the next 6mos of bugfixing. today's systems and code sucks badly compared to the slower and more careful design and testing they USED to do.

time to market ruined things. or, was a major cause in huge falling of software and hardware quality.

Re:Does it still work?

[Anonymous Coward]

Ability to hire people who know how to work on it. Ability to buy hardware off the shelf. Those are the two that jump out at me in the time it takes to type this comment.
Oh, reduced power consumption. That's three. I'm going to click submit now.

Re:Does it still work?

[Firethorn]

What are the tangible benefite of a new system?

Keep in mind that all of these are only possibilities:
1. Reduced operating expenses. Modern computers are much more power efficient than old ones
2. Faster response time. If you keep the visual wizz-bangs to a minimum a modern system should be able to serve up a search faster
3. Cheaper hardware replacement, edging towards 'actually able to replace it'. Remember NASA hitting garage sales up for old parts? The old hardware tended to be very robust, but it still fails on occasion, and it's not made anymore. What replacements you can find tend to get more and more expensive. Perhaps worse, you don't know how long the replacement itself will last, because it's often just as old as the part it replaces.
4. Easier to find skilled personnel - you're not having to lure some old fogey out of retirement to maintain it or just to train new people who don't even have the old concepts anymore, you can hire people with experience.
5. Expandability. Some of these old systems are getting up towards their maximum capacity with things like population increases and more years of records.

Re:Does it still work?

[quetwo]

Ability to find parts on for the aging system is probably becoming more and more difficult. Regardless of the software platform, the OS is aged enough where certain parts are bound to be harder and harder to obtain should they need a replacement.

There is also something to be said about reviewing all the business rules and updating them to meet their current needs. That usually happens during a revamp of the system.

Re:Does it still work?

[plopez]

1) Bogus, increases in hardware capabilities has always been gobbled up by bloated software.
2) Yeah..... right. The first impulse will be to slap on a shiney new GUI. And the monkeys hired to code it will probably use bubble sorts or worse code.
3) Emulators
4) How about 'growing your own'? Remember, the hard part is not learning to program, that can be done in a couple of years. The hard part is understanding the business rules.
5) If you use an emulator you have plenty of hardware space, though you might have to write your own paging routines to take advantage of larger address spaces. But even in the 60's and early 70's there were mini-computers with 32 bit words, such as the PDP6 and PDP10

Re:Does it still work?

[Firethorn]

1. False. Frequently true, but not 'always'. I've seen rooms freed up in exchange for two cube shaped servers under a desk, with the operators praising how fast things now were...
2. See initial disclaimer (only possibilities) and I put the visual wiz-bang specification in there for a reason.
3. You need somebody to program the emulator. They aren't always available.
4. I actually mentioned 'grow your own' - it can get really expensive, because new people don't want to be locked into your legacy system because it wouldn't help them be employable anywhere else.
5. There's a reason I said 'Some'.

Re:Does it still work?

[Eunuchswear]

This would be the case, except that ICL VME runs under Windows on a standard Intel server now.

It is called good coding.

[www.sorehands.com]

Many people are shocked that computers/systems for 20 years still run, but is says a few things:

1. That people are used to crap code that can't keep running.
2. That people are used to crap products that can't last for more than a couple of years.

If it ain't broke, why fix it? They sent man to the moon on less CPU horsepower than my Nexus 6. Voyager has been running for more than 35 years in the harshness of space.

Re:It is called good coding.

[LWATCDR]

News Flash.
The US's ICBMs are from the 1960s and the US still uses tankers and strategic bombers from the 1950s.
Good stuff lasts.

Re:It is called good coding.

[Lunix Nutcase]

If it ain't broke, why fix it?

Code churn. Can't make contracting money by not rewriting things all the time.

Re:It is called good coding.

[TheRaven64]

Not really. We're not surprised that systems from 40 years ago are still able to do the things that they were designed to do, we're surprised that the requirements haven't morphed beyond all recognition in this time. If you spend three years developing a software system, and at the end of those three years your requirements look even remotely similar to the ones you started with, then these days you consider yourself very lucky. The idea that you could deploy a system and 40 years later your customer would still want the system to be doing the same thing sounds too good to be true.

Even good code starts to struggle after the requirements have changed completely for the 20th time...

Re:It is called good coding.

[Dr_Barnowl]

The requirements in those fields don't change.

"Drop the bomb on the target" is a problem defined by the laws of physics. I've seen artillery pieces with old brass analog computers that still work perfectly.

"Make a system that automates the processing of the asinine new rules for Job Seekers Allowance" is a moving target.

Re:It is called good coding.

[DerekLyons]

They sent man to the moon on less CPU horsepower than my Nexus 6.

I wouldn't be too sure of that. While the Apollo guidance computer didn't have much horsepower, it didn't *need* much horsepower... it was mostly a crude control system that performed only very basic calculations. All of the heavy number crunching was done by multiple mainframes on the ground and the results uploaded to the vehicle. Or, to put it another way... the CSM and LM computers were basically peripherals.
 

If it ain't broke, why fix it?

Because "not broke" doesn't mean "works well". It also doesn't mean "cost efficient". Etc... etc...

Re:It is called good coding.

[Todd Knarr]

They have. But they didn't do it overnight, they did it small bits at a time and those 40-year-old systems were patched or updated and debugged with each change. The result is a twisted nightmare of code that works but nobody really understands why and how anymore. And the documentation on the requirements changes is woefully incomplete because much of it's been lost over the years (or was never created because it was an emergency change at the last minute and everybody knew what the change was supposed to be, and afterwards there were too many new projects to allow going back and documenting things properly) or inaccurate because of changes during implementation that weren't reflected in updated documentation. As long as you just have to make minor changes to the system, you can keep maintaining the old code without too much trouble. Your programmers hate it, but they can make things work. Recreating the functionality, OTOH, is an almost impossible task due to the nigh-impossibility of writing a complete set of requirements and specifications. Usually the final fatal blow is that management doesn't grasp just how big the problem really is, they mistakenly believe all this stuff is documented clearly somewhere and it's just a matter of implementing it.

Re:It is called good coding.

[Firethorn]

The US's ICBMs are from the 1960s and the US still uses tankers and strategic bombers from the 1950s.

1. They still have a lot of upgrades since then
2. We're running up to some timelines where we're going to have to spend a lot of money to replace them, especially the ICBMs, because they just can't be extended anymore and the equipment to manufacture replacements no longer exists. More importantly, the skills to make replacements using the old techniques no longer exists in many cases.

Re:It is called good coding.

[Anonymous Coward]

Pretty much this. Hurts so much.

Most don't fucking test their software any more.
Barely anyone even runs BASIC tests at least. (usually leading to data corruption)
So many developers rely on googling and wizards to help them do SIMPLE things. It is awful.

Ever since the software industry became this casualized, it went to complete shit.
It saddens me half these people get jobs but people ACTUALLY qualified are left to starve.
Gotta meet those quotas!

Re:It is called good coding.

[__aaclcg7560]

The useful life of the B-52 bombers got extended to 2044.

Re:It is called good coding.

[Firethorn]

Which is why I said 'especially the ICBMs'. We're actually trying to buy new tankers right now.

Re:It is called good coding.

[LWATCDR]

But the Job of the OS has not. VMS is actually a great OS for this kind of system. Frankly it is a real shame that is in the Hands of HP today.

Re:It is called good coding.

[ratboy666]

In the Apollo timeframe, a "supercomputer" would be a CDC 6600 (1964).

3 MFLOPS and up to 10 million instructions per second, 60 bit word size, 262144 words of main memory (~3 million 6 bit characters) -- yes, your smartphone is more powerful. This was STILL the most powerful mainframe in 1969.

Re:It is called good coding.

[Anonymous Coward]

And how do you know this? Have you seen this particular code? For all you know it is a beautifully designed and architected system, with well defined structure for modification. Quit talking about your own code for a change.

Re:It is called good coding.

[Anonymous Coward]

VME != VMS.

VME
VMS

It's paid for.

[sandbagger]

Besides, banks are even worse. They're still running virtual COBOL card systems in their basements.

Re:It's paid for.

[__aaclcg7560]

I did a one-night job in 2005 to convert a token ring network into an Ethernet network at a Wall Street branch office in Silicon Valley. First and last time that I ever saw a token ring network in the wilds.

Re:It's paid for.

[Fished]

Listen up, Junior ...

In some ways, Token Ring was very much superior to Ethernet. A hospital I worked for in the late 90's had a huge (1000 nodes) 4Mbps TR, all as one big subnet, built long before switches came along. If you tried to do that with Ethernet, it would have crashed and burned in a week. This was, on the whole, pretty reliable (if slow). The downside was that if one card in the ring failed, the whole thing would generally die. So it was great until the 10 year old TR cards started failing regularly due to capacitors failing. We ended up replacing the whole thing with 100Mbps switched ethernet, which wasn't really noticeably faster despite a 25-fold increase in nominal bandwidth, and failed more often. :)

Re:It's paid for.

[__aaclcg7560]

Junior?! That's funny. I was the most experienced tech for that one night job. The contracting company hired two fresh out of high school students who thought they were hot stuff for unboxing a Dell computer without looking at the unboxing diagram first. The job was simple: removed the TR cable, plugged in Ethernet cable, and test video app for 300 computers. These two jokers plugged the Ethernet cable into the TR port instead of the motherboard port, and didn't catch their mistake because they didn't test the video app. Worse, the IBM project manager sent them home without checking their work first. Oh, well. I got didn't leave the job until 3:30AM and got four hours in OT pay for fixing their mistakes.

Re:It's paid for.

[93 Escort Wagon]

The downside was that if one card in the ring failed, the whole thing would generally die.

As I recall, that "downside" pretty much single-handedly killed the technology. It's a big deal.

Re:It's paid for.

[Fished]

Nope, that's not what killed it. Ethernet was just as bad before hubs and then switching came along -- even with hubs, one bad ethernet card could take down the whole broadcast domain, and did with some frequency. And with thinnet wiring (coax to the younglings) all it took was one marginal connector, anywhere in the loop, to kill the whole network. Don't even get me started on thicknet.

What killed it was money. Ethernet became very cheap to implement. Once everything moved to a star topology (hubs, then switches) the advantages of Token Ring were not worth the additional cost. Ethernet benefitted from being able to advertise higher bandwidths (10mbps, then 100mbps, vs. TR's 4/16 then, too late, 100) -- the perception was, "why would I want 16mbps token ring when I could have 100Mbps ethernet for less money?" Ethernet wasn't really any faster, and was often slower due to collisions, but everybody just looked at the total bandwidth. Once switch ports got cheap, collisions were no longer an issue and Token Rings fate was sealed.

Of course, Arcnet had a star topology long before Ethernet or Token Ring. But it too suffered from low nominal bandwidth.

Re:It's paid for.

[Duhavid]

I worked on a Token Ring network early in my career.
It was a very good, robust system.
We had a star topology, twisted pair to the desktops, the Ring was all in our cabling area.
We had failures of individual lines to desktops, but that would have happened with Ethernet.
Never had a ring failure, IIRC.

Paging John Titor

[Anonymous Coward]

John Titor... Please report to... Oh, nevermind. Some bloke in a blue callbox has already claimed it.

2038

[pjbgravely]

I bet this is one of the systems that will have to be replaced to fix the 2038 bug. There are a lot more 32 bit UNIX systems out there.

Re:2038

[punman]

I bet this is one of the systems that will have to be replaced to fix the 2038 bug. There are a lot more 32 bit UNIX systems out there.

That's why they're starting now.

Re:2038

[Anonymous Coward]

They'll probably still not be finished by 2038 going by past cases where they have tried to upgrade systems.

Re:2038

[ledow]

They deal in pensions.

Likely their applications were dealing with dates past 2032 decades ago.

The only problem is whether the hardware ticks over but given other comments, it looks like it's a virtualised system nowadays on more modern mainframe hardware.

Chances are they're the one place that won't care about 2032.

Re:2038

[Anonymous Coward]

> I bet this is one of the systems that will have to be replaced to fix the 2038 bug. There are a lot more 32 bit UNIX systems out there.

This has nothing to do with Unix.

They are running VME-B Operating System.

Big in Japan is as good as small in Japan.

[Anonymous Coward]

If the new british IT will be like british made cars, then they better stick with that 1974 model year Toyota, oh I mean Fujitsu!

Re:Big in Japan is as good as small in Japan.

[TheRaven64]

It was a 1974 ICL, which was a British company. It wasn't bought by Fujitsu until much later (2002).

Re:Big in Japan is as good as small in Japan.

[Anonymous Coward]

> It was a 1974 ICL, which was a British company. It wasn't bought by Fujitsu until much later (2002).

The hardware is unlikely to still be the original ICL 2900 VME system that was installed in 1974. Later 2900 and Series 39 machines used Fujitsu cores microcoded to run VME. Even later ICL bought in Fujitsu Nova mainframes microcoded to run VME and rebranded them as ICL.

Re:Big in Japan is as good as small in Japan.

[Anonymous Coward]

It was a 1974 ICL, which was a British company. It wasn't bought by Fujitsu until much later (2002).

Fujitsu had an 80% shareholding in ICL in 1990. It was the sole shareholder by 1998. 2002 may well be when the ICL name was dropped.

Well it would be an ICL mainframe not Fujitsu

[Anonymous Coward]

Fujitsu didnt takeover ICL until the 1990s.

Also anyone going to ICL.com will soon be directed to the Indian Cricket League webiste. So this is brilliant obfucation. The average age of DWP coders must be in their 60s.

Orange Leos?

[sysjkb]

I wonder if they are running "orange Leos"? Here's a post from alt.folklore.computers in 1998. Terribly impressive. I'm not sure his age estimate is necessarily accurate, though: the final incarnation of the Leo ceased to be manufactured in the latter half of the 60s, so it may be a bit younger.

From: Deryk Barker (dbarker@camosun.bc.nospam.ca)
Subject: Re: Multics
Newsgroups: alt.folklore.computers, alt.os.multics
Date: 1998/11/09

[...]

When my wife was working for Honeywell, in the 1980s, one of the
customers she had dealings with was British Telecom.

BT, at one location, had what they called the "orange Leos".

Now, for those who don't know this, the LEO was the world's first-ever
commercially-oriented machine (1951). Even more amazingly, the Lyons
Electronic Office was designed and built by the J Lyons company,
best-known as manufacturers of cakes and for their nationwide chain of
corner tea shops.

Anyway, an "orange Leo" was an ICL 2900 mainframe (they came in orange
cabinets), emulating an ICL 1900 mainframe, emulating a GEC System 4
mainframe emulating a LEO.

30+ year old executable code over 3 architecture changes....

Re:Orange Leos?

[Virtucon]

I miss the days when we had distinctive designs in the data center. Big blue mainframes, orange and blue DEC 20s and 10s.. Now it's all racks and the only blinking lights are on the switches :-(

Re:Orange Leos?

[Anonymous Coward]

> Anyway, an "orange Leo" was an ICL 2900 mainframe (they came in orange cabinets), emulating an ICL 1900 mainframe, emulating a GEC System 4 mainframe emulating a LEO.

The System 4 was an English Electric machine (not GEC*) based on RCA Spectra 70. English Electric computer division had taken over LEO and Marconi Computers to form English-Electric-LEO-Marconi in 1964. This merged with ICT and Elliot to form ICL in 1968. I joined ICT and later that month worked for ICL.

The ICL 2900 range could emulate an ICL 1900 by running DME-1900 operating system. It could also emulate a System 4 by running DME-S4 OS but these two emulations were distinct and unrelated. A 1900 (hardware or emulated) could _not_ emulate a System 4. A System4 (machine or emulated) could not emulate a LEO. The 2900 could (given appropriate peripheral hardware) run LEO programs directly under the DME-LEO Operating system using an emulator implemented on the 2960 in microcode.

That is: the 2900 range (2950 and above) could run one from a range of Operating Systems: VME-B, DME-1900, DME-S4, DME-LEO. There was also VME-K which was not generally released. The emulation was done by microcode.

* The remainder, non computer parts, of English Electric was later merged with GEC and the EE name was lost but this has nothing to do with computers.

Re:Orange Leos?

[Anonymous Coward]

Gosh! Nothing you put on the internet ever seems to go away!

As the author of that original posting I should mention that the GE System 4 appeared in 1965.

I never claimed that BT were running an original 1951 Leo and a little further research on the Leo ite (http://www.leo-computers.org.uk- don't know if it existed bac in 1998) reveals that the GPO (General Post Office) as it then was received their first Leo III in 1964. What I have yet to find out is how much architectural difference there was between the Leo I and the III.

So maybe my "30+ year old code" was an overestimate, but still 20 year old code wasn't bad.

New Platform

[Dr. Evil]

Port it to minecraft. There seems to be some good 1970's CS work happening there.

Useful lifetime

[NotQuiteReal]

Any old building that is in disrepair due to neglect, wasn't worth saving. Q.E.D.

You imply it is a waste to build a cheaper building, while one could argue it is a waste to over build, and commit to future expenses.

Buildings have a useful lifetime, and become prohibitively expensive... obsolete plumbing, electrical, heating cooling, insulation.

Re:Useful lifetime

[csnydermvpsoft]

Buildings have a useful lifetime, and become prohibitively expensive... obsolete plumbing, electrical, heating cooling, insulation.

Those things can all be upgraded. Within the past few years, I've upgraded all of those systems in my century-old house to modern standards, spending orders of magnitude less than we would have to construct a new house of comparable finish and quality.

All mechanical systems have lifespans far shorter than the structures themselves. Even copper or plastic pipes will fail with time. Repair is almost always cheaper than tearing down and rebuilding.

It is broken (probably)

[sjbe]

Many people are shocked that computers/systems for 20 years still run,

Only people who don't know much. It's not shocking that such a thing would happen or that hardware can be made that robust. What IS shocking is that people put systems in place without any thought whatsoever to what people might want to do 20 years later. Seriously do you REALLY think it will be efficient or practical without problems for you to use the PC you are reading this on today in 20 years? Why would it be any different for a business or government?

If it ain't broke, why fix it?

Because it probably IS broken in a multitude of ways. Just because it can get a specific job done doesn't mean it does so efficiently or without problems. I've driven a lot of beater automobiles over the years and while they usually got me from point A to point B they were unquestionably broken if a number of ways. I have PCs that are 10-15 years old here in my company doing specific jobs and they definitely have problems. Yes we still get some productive work out of them but that doesn't mean I shouldn't think about replacing them when I can.

They sent man to the moon on less CPU horsepower than my Nexus 6.

Because that is all they had at the time. Nobody would even dream of doing that way today because we have better options now. Why limit yourself to yesterday's technology if you have a choice?

Voyager has been running for more than 35 years in the harshness of space.

Which is relevant how? You're comparing a spacecraft that human eyes will never see again with a earthbound computer system that we can modify or replace any time we want.

Re:It is broken (probably)

[hcs_$reboot]

a spacecraft that human eyes will never see again

Well, until the aliens throw our junk back at us

Public Sector.. uhuh

[Anonymous Coward]

Not a surpirsie this is still running. Like all things in public sector it was probably built.for a single purpose. When a new project to replace it comes along, the bells, whistles, cream and cherries need to be added. It's the middle managers getting in the way and ramping up the cost. If it was replaced like for like it could probably be done in an afternoon of code inspection and a couple of weeks of dev.

Re:Public Sector.. uhuh

[eneville]

I think what you're referring to is https://en.wikipedia.org/wiki/..., time being proportional to the number of heads on a project.

no more bugs

[AndyKron]

They must have finally gotten all the bugs out, so now it's time to upgrade!

Extreme example here, but...

[ErichTheRed]

Not all legacy stuff is bad. Not all legacy stuff should be kept around to the point where you can't find people to run it, however,

I've had experience working in die-hard IBM mainframe shops as well as places that used the HP MPE operating system on the HP 3000 minicomputer system. In the 3000 case, the customer was relying on a service provider that was providing an application that was way way way out of date but still worked. All the IBM places I've ever worked have been slowly "modernizing" their application stack, but in most cases, the core transaction processing has remained on the mainframe because that was the best solution. It's extremely rare these days to see an end user facing green screen application, but they do exist as well. (Yes, I work in "boring" old school industry sectors, very few web-framework-du-jour hipsters here, but we're also not old farts.)

The problem I've seen is that vendors love the fact that customers are locked in and will do nothing to encourage them to get off. Most ancient mainframe code can run virtually unmodified on newer hardware, and that backwards compatibility is a big selling point. It allows IBM to go in, swap out your entire hardware platform at $x million, and keep billing you by the MIPS without changing any code.

But...the reverse problem is that "mainframe migration" projects often end up becoming case studies of how Big Consulting Company X was paid hundreds of millions to not deliver a working system. I believe I read about DWP's "Universal Credit" project that has Accenture, IBM or Oracle written all over it. These kinds of projects usually try to port all the business logic and transaction processing to some horrible-to-maintain J2EE monstrosity backed by an Oracle database. They usually fail because (a) no one correctly estimates the work required to pull all that business logic out of 30+ years of cruft, and (b) the consulting companies replace their star team (that travels with the sales force) with new grads in India (who do the actual work.) I've seen this cycle over and over again, and am still amazed that CIOs aren't wary of consultants.

Re:Extreme example here, but...

[Ash-Fox]

But...the reverse problem is that "mainframe migration" projects often end up becoming case studies of how Big Consulting Company X was paid hundreds of millions to not deliver a working system. I believe I read about DWP's "Universal Credit" project that has Accenture, IBM or Oracle written all over it.

I work in the consulting industry, I have always delivered. That said, some of these can be absurdly difficult to deliver. You have to work with 3rd party vendors that are obstructive because they see they're on the way out. When in government you have to deal with various policies to do with change requests, importing of new software to the network, hardware and software acquisition and through no fault of your own, these organisations are very slow moving and can take months in some instances approve. Never mind when you get rejected and you have to spend excessive amounts of time trying to convince the client that they do want it because it's in their best interests and there are no alternatives that are sufficient.

Generally people assume the contractors are just messing about and fail to deliver a project because of poor organisation and such. The problem I find more common is that the client ties the hands of the consulting firm and obstructionists prevent the firm delivering much of anything.

Selection bias

[sjbe]

I am always amazed how buildings constructed thousands, or even hundreds, of years ago are still standing although often in a state of disrepair due to neglect.

Then you are a victim of selection bias. Those buildings that are still standing after hundreds of years are the best built ones. The ones that weren't built so well aren't around anymore. So you think that they used to build them better back in the day when in fact they mostly built them worse if anything (they didn't exactly have building codes back then) and the old crappy buildings simply aren't around to compare with anymore. Saying "they don't build 'em like they used to" is true but not in the way it is usually intended. Building something to be more durable than it needs to be is wasteful - just in a different way.

Ship of Theseus

[sjbe]

The US's ICBMs are from the 1960s and the US still uses tankers and strategic bombers from the 1950s.

B52s have been rebuild and upgraded and refurbished so many times they may as well be the Ship of Theseus. Furthermore the munitions they carry aren't really the same either these days in most cases.

Why migrate?

[Mostly a lurker]

I assume these applications are not running on the original hardware. They should still be working fine on current Fujitsu mainframes. There may be a valid reason to rewrite part or all of the applications because additional functionality is needed but, too often, money is wasted replacing systems (especially mainframe systems) that still meet most of the enterprise's needs, Often, "more flexible reporting" is used as an excuse for hugely expensive rewrites, when a periodic data extract into a separate data warehouse can meet the need much better, much more cheaply, and without disruption to existing production systems.

use an emulator

[rewindustry]

should be simple.

am willing to bet they will make it complicated, for profit.

Hello, Big Bank Here....

[tekrat]

And we are running an IBM Mainframe. Yes, it's been upgraded to zOS, but it's fully OS360 compatible and I regularly review Cobol code with comments going back to 1980. There's been a push from above to use the Control_M "GUI" interface, but a lot of the folks here are resistant, since we have faster and better control via the terminal (sorta like GUI versus Command Line).

And yes, my Windows workstation is simply a glorified terminal as I spend all day logged in to the mainframe itself (green screen apps).

It is broken (probably)

[Anonymous Coward]

Depending on "yesterdays technology definition" but for harsh environment and requiring very high reliability. They are likely not going to choose the latest generation. I can't locate it but there was a story where a new space vehicle design was discussed including triple redundant computers using 10+ year old processors because 1) it met their computing power needs , and 2) they had a proven track record in harsh environments.

Dust off the CV

[Virtucon]

VME is unfortunately on my CV. What I'm amazed is that they can still get parts for the damn thing.

Re:Dust off the CV

[Anonymous Coward]

They don't actually run on mainframes anymore. Everything was moved to superNova based systems, which runs as a VM on top of Windows and Linux.

Is it broken? Is it unsupported/sunsetting?

[davidwr]

If it's not broken, it gets the job done, it's vendor-supported, and you don't expect that support to end in the foreseeable future, then I don't see the problem.

Age alone is not any reason to declare technology obsolete.

Here's a common example: Stores still sell 4-function calculators for $5 or less. As far as the user is concerned, they are less-expensive versions of the same calculators you could buy from the mid-1980s on, and thinner-and-cheaper-with-LCD-and-button-battery versions of the kind you could buy in the 1970s. In other words, if you still have a 1985 four-function calculator and it's still working, it still meets your needs, and you can still get batteries for it, there's no reason to throw it out.

Heck, even the lowly manual typewriter is still better than a pencil, pen, electric typewriter, or computer+word processor+printer in certain situations. In others, the best choice is a pen or pencil.

Power cycle it

[Anonymous Coward]

just power cycle it, it should be good for another 40 years

Amost Yoda

[Anonymous Coward]

Many people are shocked that computers/systems for 20 years still run ...

Just switch "many people are shocked" with "shocked are many people" and you're there!

A quote from the Minister of Upheaval....

[magusxxx]

"Don't worry, we'll make sure the replacement will run Doom."

New Platform

[Anonymous Coward]

This was the funniest thing I've read/heard all day.

From a proxy insider....

[devilvole]

IDS and his yes men IT tarts are totally incapable of understanding the complexities of what they are attempting. The fact that the systems are VME is simply a reflection on the vast amount of logic that went into developing some substantial business systems (probably more from the 90's than the 70's) "The department's legacy systems still run on Fujitsu mainframes using the VME operating system installed in 1974" is a bit inaccurate, it was ICL in 1974 and has certainly been updated many times since then and the machines are pretty modern, just running VME. All that logic can't simply be magicked into some new technology by a bunch of outsourced IT staff who don't know the difference between Universal Credit and Wonga. IDS still obviously believes it's some trivial bit of code that basically just needs to say "No Soup for YOU!" But once the DWP got rid of all the in-house IT bods, who actually understood the technology as well as the benefits, they gave away the keys to the safe. Expect more billions to be extracted by the UC leeches.

Re:From a proxy insider....

[MavEtJu]

> The fact that the systems are VME is simply a reflection on the vast amount of logic that went into developing some substantial business systems

In my approach: Logic should be in configuration files, not in the code.

Re:From a proxy insider....

[Thiarna]

What difference would that make? If 40 years of business rules are encoded in configuration files how is that any easier to work with than COBOL or whatever language was used?

Re:From a proxy insider....

[Megol]

But then your configuration files are code.

It Works, So Don't Mess With It

[Anonymous Coward]

Sadly, some politiican will probably propsoe spendign 10x the original cost to replace it, and project ovverruns will add 4x to that cost.

If it works and meets the needs, KEEP IT!

The California DMV has them beat.

[funwithBSD]

The California DMV has them beat, they are still using code installed on UNISYS mainframes in 1970 to run the DMV core applications.

It is as old as I am...

Re:The California DMV has them beat.

[ebvwfbw]

The California DMV has them beat, they are still using code installed on UNISYS mainframes in 1970 to run the DMV core applications.

It is as old as I am...

A lot of states had Univac stuff. I worked on them. I didn't think anyone still had one running. I know the University of Maryland blew it out of the building back in the late 1980s and they wrote a lot of Exec 8. They wanted way too much money to run it. Replaced it with Unix systems of course. Far superior.
Cal should upgrade from that old POS. I never considered it good when it was good. Helped put me through college though.

They may have the manual, but ...

[Jumunquo]

... how long will it take to hack in? Days later, the machine will finally respond to the unoptimized hacking code, it will launch a shell, and out-of-memory, lol.

Re:Is it broken? Is it unsupported/sunsetting?

[Ash-Fox]

it's vendor-supported, and you don't expect that support to end in the foreseeable future, then I don't see the problem.

These things tend to be vendor supported at incredible costs.

Simply....

[Anonymous Coward]

1. Create a virtual environment capable of emulating the existing hardware (if any)
2. Port the existing applications and data to the virtual environment, removing the expensive requirement to maintain, back up and develop on legacy equipment and hopefully improving records processing speed

Given how the current system is probably the evolution of decades of undocumented patches and replacing such a system (remuneration is always horrendously complex as the rules have changed so many times over the years and there would have been fiddles done perhaps even on the level of individual cases) would be a total nightmare...

3. Write a new system to handle all new pensions under current rules and be done with it. Eventually all the people on the old system will die and you can archive the lot.

Now it's no longer a technical issue

[davidwr]

At this point, it becomes an economic issue, not a technical issue.

Compare two cases:
You buy/rent a big mainframe with a long-term support contract. 5 years in, the vendor has economic trouble and files for bankruptcy. The bankruptcy court lets the vendor give 2 years' notice before canceling long-term support contracts. The vendor notifies you that in 2 years, it will cancel the affordable support contract and offer you a "take it or leave it" mega-expensive contract (they hope you will say "no"). Except for the money, you are happy with the product and you know that if you say "yes" you will get good service and support.

Case two:
You contract to buy/rent a big mainframe with a long-term support contract. When that support contract is up 10 years later, they raise the rate significantly for a 10-year renewal. 8 years into the renewal, they notify you that the next renewal will be at a mega-expensive level (they hope you'll say "no"). Except for the money, you are happy with the product and you know that if you say "yes" you will get good service and support.

In neither case does the "renew/not renew" decision rest on technical merits. In both cases, it's a pure economic decision: Is it cheaper for us to renew or to replace?

VME is still sold

[Anonymous Coward]

VME is still sold and used by MANY, MANY organizations. It doesn't mean it's obsolete. It's even virtualized into Unix/Linux. Basically this is what IBM does to this day with it's 360 emulation on Linux. In the case of IBM, this is part of what pulled their fat out of the fire in the 1990s!!

This isn't a bad thing: it's actually a major success story for computer engineering/science!

http://en.wikipedia.org/wiki/ICL_VME