Forum: Author Hangout

dumb and lazy to have a back-up to the computer systems

It's actually not even that. There are so many systems in factories today that are running on 30 year old computers and software, because it's not possible to upgrade them.

Just to make it more fun, when you upgrade to computer control systems like this - which they've done years ago - it's entirely possible there's no one left working that knows HOW to do it the old way.

It's really hard to justify the expense to the corporate accounting/finance/executives until the old system fails completely bringing the business to a stand still.

Not unlike the issue with Y2K. My mom was a programmer, and knew of that issue way back in the 1970's. But at that time, computers were expensive and memory even more so, so they just lived with the 2 digit limitation as they wrote code. Only coding for 4 digits if required.

An additional 2 bits of data per line may not sound like much today. But at that time, even corporate mainframe computers like the IBM System/360 only had from 8-64k of RAM. In an intensive accounting program, those extra 2 bits per line could crash the code with a memory overflow.

Plus, it is more than just the direct cost. You have to train all of your support staff in how to support the new hardware-software, then your staff in how to use it. That can be even more expensive than the upgrade itself, and consume thousands of work hours in the transition.

Add in, the time and cost of QAQC and compatibility testing. Especially if one program is supposed to talk to another. Program A does a job, and reports it's data to Program B, which handles the accounting. Then passes that along to Program C, which does the billing. You have to test to make sure each one of those still works, and may have to upgrade more than one program in the process. Adding to the cost and time required.

And yes, I have run into this first hand many times. Like in 1996, when we were in the process of working with DirecTV to upgrade their network from Novell 3 to Novell 4, and from Windows 3.11 to NT 3.51. The suits upstairs did not understand that they could not keep their old version of Word suite on NT, and would need a newer version. And that they should instead have us testing for NT 4, and forget about 3.51 as it was already on the way out.

We wasted so many hours making test builds, that was a combination of NT OS and Win 3 software. And many just did not want to "play nice", as NT had security and Win 3 did not. Plus one was a hybrid 8-16 bit OS, and the other was a full blown 32 bit native OS. Many programs had to be written just to work on NT, but when done it performed much better than those for 3.X.

A year later I moved on to another company, and they were still battling out how they were going to do the upgrade. The suits had finally agreed with NT 4, but were still causing us grief over our insistence that 90% of the software needed to be upgraded as well.

An additional 2 bits of data per line may not sound like much today. But at that time, even corporate mainframe computers like the IBM System/360 only had from 8-64k of RAM. In an intensive accounting program, those extra 2 bits per line could crash the code with a memory overflow.

Actually, it was core. (And people should note that's "k", thousands, not mb and the OS was in there too.)

And it was the disk space cost. Disk was very expensive back then so saving two bytes for each date stored was a win (until Y2K).

Memory and cost of disk space were no longer that big of an issue way back in the 1980s.

I was responding to "My mom was a programmer, and knew of that issue way back in the 1970's."

In the 1970s, it was core (not memory) and core was very very expensive. And disk was also expensive. Tape was the common media back then where the computer room (mainframes) was filled with tape drives and the operators would run around changing tapes. I remember one company had their operators on roller skates.

In the 1970s, it was core (not memory) and core was very very expensive. And disk was also expensive. Tape was the common media back then where the computer room (mainframes) was filled with tape drives and the operators would run around changing tapes. I remember one company had their operators on roller skates.

True, but listing cores makes little sense to most that do not understand those old systems, so I used a similar term that most even today can comprehend.

And keypunch was still a common input media into the early 1980's.

Memory and cost of disk space were no longer that big of an issue way back in the 1980s.

Late 80's, true. Early 80's, still very expensive. Even in 1982, a home 5 MB hard drive was still over $5,000 US. The price really did not drop until the mid-1980's.

That was when the reliance upon old style mainframes started their decline, and businesses started to give up their dumb terminals and mainframes in exchange for server based systems with PCs. But even those still often worked in conjunction with the old mainframes until the late 1990's.

When I first worked at Hughes Aerospace in 1995, they still had hundreds of mainframes at their El Segundo campus. By 1999, there was only one left. And it was the last legacy system, and was only used for a special high speed check printer. They had determined that replacing that one system was not cost effective, so they did a Y2K patch on it and it was still in use in 2001 when I left.

The last System/360 they had in operation. Only doing a single job, because even checks were rapidly becoming obsolete and they could not justify the cost to replace it with a newer system.

I largely laughed at Y2K, because I knew it was not the problem many feared it would be. I was installing new systems in 1995 that we knew were not Y2K compliant. But with an in-use lifespan of only 3 years, we knew they would be replaced with a new system before then so it did not matter. And most major businesses started almost a decade before migrating systems away from the mainframes to Y2K compliant software and systems.

The biggest problems I saw in the final year were in small businesses. I switched pager companies because in February 1999 when I went in to pay my 1 year subscription, they told me they could not take the reduced 1 year fee because the system would not allow it. I talked with the owner and actually gave him a dirt cheap upgrade price (they were actually still using XT workstations with a Novell 286 network system). He turned me down, to expensive. So I went to another company for my pager system.

8 months later I found out they had finally gotten a rush upgrade from another tech, at almost 10 times what I had quoted him for. And I was glad I had changed, a lot of companies tried to ignore that as long as they could, and it ended up biting them in the ass.

When I first worked at Hughes Aerospace in 1995, they still had hundreds of mainframes at their El Segundo campus.

We must be talking about different size mainframes. The companies I worked for back in the 1970s had a few IBM 360s and then 370s. Not hundreds. And when I left American Express in 1995, they had two 3090s. Just two for one of the largest mainframe computer operations in the world.

We must be talking about different size mainframes. The companies I worked for back in the 1970s had a few IBM 360s and then 370s. Not hundreds. And when I left American Express in 1995, they had two 3090s. Just two for one of the largest mainframe computer operations in the world.

Yes, some were fairly basic, some were much more powerful. But not all IBM mainframes, they also had WANG, UNIVAC, and others.

Hughes Aerospace in that era was still a huge aerospace company. Their headquarters in El Segundo was massive, and covered many square miles. RADAR, missile systems, satellite, space systems, and many others. Along with subsidiaries like DirecTV, AC-DELCO, and more. It was the largest aerospace conglomerate at the time, and not only did the company have their own mainframes, but there were others dedicated to each sub-part.

That was one of the reasons why I was part of their first "Corporate Computer Rollout" team. Before then, each department and section bought their own computers, as they saw fit. But by 1995 their IT infrastructure was a mess. They settled on a combination of Dell and Apple systems, and each computer in the place was replaced with those.

We were literally pulling out Apple II, Commodore PET, HP 100 and 9800, and almost everything else made from the early 1970's until that time. And the departments had done the same with mainframes prior to that. If they thought they needed one for a project, they bought it.

One thing a lot of people get messed up with is the mainframe computer systems were usually on or two mainframe CPUs with dozens of tape drives, disc packs, RAM memory units, multiplexers, card readers, and a printer or two attached. Thus when you walked into the computer room there were machines everywhere, but 95% of what you saw were the peripherals and not the actual mainframe unit.

One thing a lot of people get messed up with is the mainframe computer systems were usually on or two mainframe CPUs with dozens of tape drives, disc packs, RAM memory units, multiplexers, card readers, and a printer or two attached. Thus when you walked into the computer room there were machines everywhere, but 95% of what you saw were the peripherals and not the actual mainframe unit.

Oh, I know. I have been in many "fishbowl rooms with raised floors" over the years. I even went into one both before and after it had the mainframes pulled and turned into a workspace.

For example, at Hughes Satellite, they had such a room for each bird they were making. And back in the day, there was at least 2 mainframes in each one. One was their own, and the second was a clone that would be then sent to the final owner of the satellite after it was completed and before it was launched.

And back in the day, for security reasons each major project or department had their own. Or in the era before rapid distance communications, when putting one in each location just made sense.

I know the military had over a dozen of them on Okinawa in the 1980's. Many were PDP-10 based systems, even into the early 90's most of the Marine Corps was still working on that platform and pretty much every base had at least 3 of them.

And I remember that because when I left North Carolina to go to Japan in 1990, we were still doing "SneakerNet" with 8" PDP disks for all of our administrative, maintenance, and supply data. And they were still doing it when I left Japan. But when I returned to North Carolina, they had already transitioned to a new AN-UYK 80386 based TEMPEST desktop that must have weighed 200+ pounds.

We still did the old SneakerNet game however, as networking was still many years away. But they were now away from mainframes for that job and doing it on more contemporary server clusters.

And yes, in one of my stories set in that time period in Japan, that was how they really did it. Each evening, a driver hit all the major bases on the island, picking up the 8" disks and taking them to the central IT facility (Camp Butler). At Lejeune, it was only base wide, but doubled as you had 2 major commands (2nd Marine Division and 2nd FSSG). Then all the disks went into cassettes, and were fed into one of the three mainframes, SUPPLY, ADMINISTRATION, and MAINTENANCE. They would churn over all that data, and then the next morning the courier would return with our disks, and any printouts we needed.

And when you are the USMC, buying 3 PDP-10 systems for $1.5 million total is not a huge problem. Plus it gives valuable redundancy, as if one fails you still have 2 others you can work with.

I am probably one of the last still in the military who remembers how we used to do things 40 years ago before computers took over most of our daily operations.

I am probably one of the last still in the military who remembers how we used to do things 40 years ago before computers took over most of our daily operations.

Don't worry, soon Skynet will resolve all of those problems.

I don't understand that point. Why is it not possible to upgrade the computer and software?

I have had a lot of experience in this.

We call these old systems "Legacy Systems". And the reason they may be in place are many. And it can be both hardware or software related.

To start, most of those made before the mid-1990's used hardware like ISA cards or "dongles" that plugged into a serial or parallel port. They stopped making computers with ISA cards slots and hard wired ports almost 20 years ago, so if you are running hardware that requires such a card, you are stuck with old systems.

And no, most of the "plug and play adaptors" that you find actually do not work. They basically emulate those old ports, for use with something like a printer or modem. They are not "true ports", and as such the security devices refuse to work with them.

Also, the software. Most of that was written on DOS or Windows 3, and the companies that made it are long gone (or moved on to other things and no longer support it). They might have even paid tens or hundreds of thousands of dollars to have it written for them back in the day, and those that wrote it are long gone.

I have seen this so many times in my career I can not even count. In 2007 seeing an expensive water carving machine to etch rock that cost them a bundle. Yet the interface card was ISA, and the software only worked on Windows 3.1 I could not even rebuild it, as it used an MFM hard drive (early generation, incompatible with any modern machines) and they had long ago lost the disks. And the company that made the software was also long gone.

All I could do was give it a good cleaning, and encourage them to start looking for a replacement as I literally could not fix it if it had a catastrophic failure.

The problem is that a lot of this equipment was very expensive when it was made, and either nobody makes a more modern version, or if they do it is even more expensive as they are the only ones that do. That is why many ATM machines still use O/S 2 Warp, and many point of sale systems are still using Windows 3, NT 3/4, or even DOS based. The cost to upgrade is very expensive, and most companies simply decide to keep working with what they already have.

Plus, you also get into other issues. Like the fact that Windows 95 tends to get unstable when run natively in machines over 1 GHz. And for each generation of the operating system past the one it was made for, the less likely it will run. Good luck getting more than a handful of XP programs to work on Windows 10. Some will work, most will not. And for many of those control machines, emulation is not an option as they have to have full control of the entire machine to operate.

people arguing that an EMP could not take out a pipeline.

The pipeline itself would still be there. The computers that control the valves would be shot.

In other words, those people were wrong. (As usual.)

In other words, those people were wrong. (As usual.)

I am one of those that said wrong, but for different reasons.

EMP is greatly over-estimated. It is based on some very specific tests done decades ago, and almost all of the damage was to very specific things. And it is also remarkably easy to shield electronics from EMP.

Also, I seriously doubt that a country would waste nukes detonating them in space for a "maybe-if" damage, when they could just drop it on an actual target and do real damage. In this case, why waste a nuke in space, when you can just nuke the damned pipeline itself?

In this case, why waste a nuke in space, when you can just nuke the damned pipeline itself?

For the same reason you use neutron weapons or nerve gas. You kill the people, without destroying the infrastructure. Without getting into the whole EMP discussion again, if it's possible to disable an enemy so that they are forced to use resources to deal with civilians that they COULD use to fight back, when your conventional military moves in, the defender will find himself on the short end of the stick.

I think you seriously underestimate the threat. As for the testing you refer to, I hope you don't mean the EMP commission testing. That was a complete farce designed to give the Congress critters a warm fuzzy.

No, I am talking about how simple it is to actually shield equipment from the effects of EMP.

It's known as a "Faraday Cage", and we have been putting them inside of military equipment for decades. Originally as an anti-espionage measure, then to also protect it from EMP (stopping emissions from coming in is the same as stopping them from going out).

Hell, at Hughes we even had entire sections of buildings that were giant Faraday Cages. Walk into one, and your pager or cell phone immediately stopped working. And it is nothing miraculous, we have known the principal for almost 200 years now. Hell, I bet everybody reading it owns one and does not even know it.

How else do you think you can watch food cooking in a microwave oven, and not get cooked yourself?

This is stupid-simple, and can be created with early-19th century technology. Most government, military, and military-government contractors have been using them for decades.

How else do you think you can watch food cooking in a microwave oven, and not get cooked yourself?

Because the holes in the shielding around the microwave are smaller than the wavelength of the microwaves generated.

You're right in your last sentence.

Most government, military, and military-government contractors have been using them for decades.

You're simply refusing to accept one relatively minor, nearly insignificant detail. CIVILIAN BUSINESSES DON'T CARE! It's great and wonderful that all of the military equipment has it. The military isn't who built and runs the infrastructure of this country. It was all done by the lowest bidder of whatever contract was put out, to build whatever project it was.

Hey, we need to build a pipeline! Hey, we need to build an office building! Hey, we need to build a television studio! Hey, we need to build a hospital! LOWEST BIDDER!

Do you really think that everything out here in the civilian world is EMP shielded? Hell, half the stuff sitting over on Tinker isn't EMP shielded. I know damned good and well their computers systems aren't - they've had issues over there in the past. About the only building and section that really is EMP shielded there is the Cybersecurity building, and that's because they built it that way. The rest of the buildings were put up LONG before anyone worried about EMP, and they've never retrofitted them.

Because the holes in the shielding around the microwave are smaller than the wavelength of the microwaves generated.

Hence, a Faraday Cage. That is actually the technical term for it.

And you can make one at home using nothing but a galvanized steel trash can. Some paranoids even install Faraday Mesh fabric (a real thing, under $20 a square yard) in rooms in their house.

Heck, a lot of older computers really were Faraday Cages. It was a requirement to prevent leakage by the FCC. And a lot of your higher end server racks still are Faraday Cages.

You're simply refusing to accept one relatively minor, nearly insignificant detail. CIVILIAN BUSINESSES DON'T CARE!

Oh, they go by the fancy name "RF Shielded Server Racks", but they really are the same thing. Not at all uncommon, or rare. And putting a second Faraday Cage around a server cluster is also very common, as it also protects from a lot of other things, like all the RF thrown out if a transformer blows up or extreme sunspots (or the placement inside of a transmitter to be used in "man in the middle" attacks).

You have to realize, I have been dealing with these for decades. And civilian businesses do care, because it is cheap insurance against many security issues. Even more so now as many are moving to locking faraday cage racks as part of their cybersecurity system.

https://hollandshielding.com/RF-shielded-racks

The cost to actually add a Faraday cage to a 10 yard by 10 yard server room? Just a few thousand dollars, it really is not that expensive.

Hell, I have even seen "accidental Faraday cages" on more than one occasion. One of the funniest was when a construction company brought me out, because their new wireless router could not talk to a building only 50 yards away. I went there, the router was literally inside of an on-site construction office trailer, inside of a sheet metal building. Faraday Cage. I got an external antenna, put it outside the trailer, and it worked.

And I have heard those arguments before. But it is almost never true, especially when you are talking about the IT infrastructure. There is a reason many IT Managers in smaller companies are also the Disaster Manager. And why IBM ruled desktop computers for a decade (and mainframes before that). Sure, a System/360 cost a hell of a lot more than a PDP-10. And an IBM AT cost a hell of a lot more than a Compaq.

But as the saying goes, "Nobody ever got fired for buying an IBM computer". Companies shell out a huge amount of money for these systems, and absolutely rely upon them for daily operation. "Spending an extra $500+ per rack for shielding? Bah, don't even bother bringing it up, just order the thing if you think we need it."

Oh, and notice I did say "military-government contractors", as the government requires most contractors to take these safeguards. And many do it themselves, to help prevent not only hacking, but corporate espionage.

And most equipment does not need to be shielded, just the critical stuff. Yes, a lot of stuff will need to be replaced, like say the electronic switches on the pipeline itself. But not the system that runs it.

And I can promise that more is shielded than that, because TEMPEST requirements go all the way back to WWII. When Bell Telephone proved to the military that they were able to intercept clear text transmissions of coded information by analyzing the power use of the systems doing the encryption and decryption.

Faraday Cages, TEMPEST, and a great many other technologies still used today long predate the recognition of EMP. Even going back to when an "atomic bomb" was just a figment of the imagination.

"Spending an extra $500+ per rack for shielding? Bah, don't even bother bringing it up, just order the thing if you think we need it."

I was flat out told: "Emmeran, we've paid $25 thousand just to watch monkeys fuck. Order the stupid thing already!"
(Obviously they didn't call me Emmeran and we only discussed "real money" after that.).

That's the corporate reality and that was a C-Suite meeting. My perspective and approach changed that day.

Anyway, the point of bringing it up was to point out another threat vector. To argue otherwise is pointless. Infrastructure (actual infrastructure and not the genned up bullshit from the left) is sorely exposed to external threats and in dire need of fortification sans the political gamesmanship.

And mostly that is because until the last decade or so, nobody ever really considered such things as a threat since the 1980's.

Other than those who felt the effects of the Coronal Mass Ejections, Solar Flares, Coronal Clouds, and other such events. At that point, a lot of facilities started shielding themselves to prevent that from happening again. Yet, others played ostrich and ignored it.

And most do not seem to comprehend the concept of "mitigation". Nothing will ever remove all threats everywhere. But the idea of Disaster Planning is to take steps to minimize the long-term effects, and prevent the damage or destruction of key components.

A massive CME or EMP could take out all of our power systems tomorrow. Mostly at the local level. That is just time and work to replace, so long as the critical infrastructure remains that is acceptable. And many do not seem to get that is the really important part of the system.

I could see something destroying all the CAT-5 cable in a building. So what? Only work to run more cable, so long as my servers remain intact, that is the really critical part. Even the computers at the desks are disposable, they rarely sit there more than 2-3 years anyways.

More and more datacenters are putting them in for other reasons. CME being one of the biggest ones, as well as things like a near-miss of a nuclear bomb (ground bursts and other things also cause EMP), massive lightning strike on the building, and others. There are a great many sources of EMP, not just nukes in space. For them, they see it as "cheap insurance".

I can only imagine most here have never really worked "in a corporation", so have no idea how they really function. Ironically, many are much more likely to pinch pennies when it comes to security and other areas than they are on crucial IT infrastructure.

Hence, why OS/2 is still in use. Antiquated, outdated by decades, but seen by many in the industry as "bulletproof". It is not kept around from being cheap, as much as it is just really damned stable, and even 20 years after it's first release IBM still supports it. But they run it today on modern systems, not ones built when it was still sold.

I have worked on some, and for some reason when I tell people that they imagine a company too cheap to buy new systems. When in reality, I was putting it on modern systems, where the OS was actually mostly wasted. It can not take advantage of things like 64 bit processing, or the use of RAM above 4 gigs because it is still a 32 bit OS.

They simply felt secure with it, because it is pretty much immune from virus (I can't remember the last time I heard of an OS/2 virus), stable (most software written on it is 20+ years old and long had all issues debugged out of it), and since it is an IBM product it will continue to be supported for decades to come. And in reality, it costs then more to maintain an OS/2 system as less people know how to work with it, and most software needs to be custom made and just does not exist "off the shelf".

I've worked in them and for them in contractor mode. Among them was colonial. They definitely took security in all its forms seriously. Unfortunately, all it takes is one idiot with a thumb drive to fuck over IT security.

Which has always been the case. However, it has nothing to do with my comment, which was aimed at those that think that every corporation is going to be as cheap as possible.

The issue here is that it has not been a threat, and it still involved "thinking outside the box" in a great many ways.

And we ourselves have used this (supposedly).

That is why most military computers have access to data transfer to and from USB ports turned off. And severely locks down any kind of software writing to the hard drive. But this is rarely seen in most companies.

I work for an IT consulting company doing work for a publicly traded utility company. Both my employer and the client block storage devices on computer USB ports. USB mice and other non-storage devices are fine.

And I am not surprised. But is that done for the systems like those? And how many of those systems are legacy systems where it is not possible?

The systems like what? It's done on all company owned laptop and desktop computers.

I'm not sure what this is supposed to mean, it's done at the OS level for all laptop and desktop computers. These systems are all replaced company wide on a 3-5 year schedule.

As to any servers that may be older, but not so old as to not have USB ports at all, physical access to the servers is tightly controlled with physical security.

Legacy System, that is a system that uses an operating system and hardware that is generally considered "obsolete", but is still in use for other reasons.

Such as a Pentium II, or a system using XP. Basically anything not using Windows 8 or 10, or anything older than dual core 64 bit.

And yrs, there are a huge number of systems out there still. Doing things like this because nobody makes replacement software for the most part.

And same with servers, nobody cares if a server has USB ports, that is completely besides the point.

Where I work, "legacy systems" is almost exclusively used to refer to software applications

That's been my experience too, and IMO is the meaning that makes most sense. However teh interweb disagrees and claims it applies to both hardware and software.

AJ

Where I work, "legacy systems" is almost exclusively used to refer to software applications, so I wasn't sure if you were referring to OS/hardware or not.

And again, nothing that old at either my employer or the client in terms of laptop or desktop computers.

The client pushed Win 10 out company wide less than six months after MS officially announced that they were cutting off patch support for Win 7. Anyone with hardware too old to run Win 10 got new hardware.

This is not common on a great many machines, and I have worked on tons of them over the decades.

Goes right back to what I had stated earlier. Such as a rock crusher and mover I worked on at the US Borax mine. The "machine" was literally built inside of a building, interfaced to the computer with ISA cards, and still used Windows 3.1. Even then they were trying to find somebody to update it, but were having no luck as the program and cards were custom made for them, and the documentation so scanty that they were already working out the budget to start all over again from scratch.

Hardly unique, things like ISA cards and dongles were once common, and if a company relies upon them to function, they will keep using them.

On Netflix right now there is a documentary about the last Blockbuster, in Bend, Oregon. And in the back she has a stack of antiquated systems she cannibalizes for spare parts. She is the last to use the software, and she relies on it to operate her business. Nobody even makes a replacement for it, she is stuck with old systems and working with what she has.

And good luck with the "new hardware", if you need to use an ISA card, or a native hard-wired serial or parallel port. Very few dongles work on add-on cards, been there - done that.

Delivering product when you can't bill for it is a recipe for bankruptcy.

The principal publicised reason for the UK government pressing for 'Smart' Meters in every home to measure electricity usage is that the National Grid doesn't know how much leccy it's supplying at any given time.

Sceptical? Moi?

AJ

3. Easier to implement rolling blackouts that will be necessary to handle inconsistent supply when a majority of power comes from grid scale wind and/or solar.

Some consumers in Texas just found out about that the hard way, with their Smart Thermostats. Middle of the day, high energy use, the supplier moved the temperatures in the houses up to 78. Regardless of whether anyone was in the house or not ...

Lots of them. Here's just a couple.

From MSN

From KHOU

3. Easier to implement rolling blackouts that will be necessary to handle inconsistent supply when a majority of power comes from grid scale wind and/or solar.

That's the one I'm most afraid of. Unless hydrogen overcomes the lunacy of betting the bank on electric cars, when conventionally-powered cars are no longer manufactured and everyone has to put their electric car on charge overnight, there will be a colossal mismatch between electricity supply and demand.

AJ

Americans utilize their vehicles far more than their European counterparts.

I have no direct experience, but colleagues who worked in the States confirmed that. In the UK, a trip to the shops to pick up some milk is usually a short walk. In the US, it's more likely to be several miles.

However I'm not convinced most Americans spend more than 12 hours a day in their cars.

AJ

After looking back, I don't see any mention of 12 hours either.

Lupus Dei wrote "Most cars spend most time parked anyway."

You replied "Maybe in Europe, but not in America."

There are obvious exceptions, as you pointed out, but I believe Lupus Dei was right in asserting that most cars are not in use for most of the day.

AJ

Hydrogen - power cell - electric is the way to go for anything larger than a school bus, but also doesn't quite work for consumer streetcars. And it's not quite here yet anyhow.

I've just read a motoring review of a hydrogen-powering-electric-motors mainstream car. They are here now, although they probably have drawbacks (in the UK, there's currently only about a dozen hydrogen filling stations). Although the range is greater than an equivalent electric car (even before halving the manufacturer's stated 'up to' range because they overstate the practical range of electric cars by 100%), it's still short of a typical internal combustion engine equivalent.

Having been world leaders in hybrid technology, Toyota is now betting heavily on hydrogen. (Backward European manufacturers are still in the process of moving from clean diesel to electric.)

And no, burning oil in cars should be criminal, for reasons unrelated to climate, it is just so criminally wasteful.

Even biodiesel? EU diktat means that 10% of diesel fuel sold for domestic cars must be biodiesel.

AJ

Well, rape seed is aptly named (in English translation).

I used rape seed oil when cooking my evening meal. I don't feel violated ;-)

AJ