Posts by Cynic_999 2855 publicly visible posts • joined 15 Aug 2013
"
how long do you think the pressure in the gas lines will stay up once the pumps are turned off? And how do you think those pumps are controlled?
"
Gas pressure is maintained by gasometers, which work by gravity, so pressure will stay up until the gasometer is empty even if there is a total power outage. When the gasometer needs to be filled I should think the pumps are (or can be) operated manually.
"
The Russians are unlikley to crew over their own sat nav system, are they?
"
Of course they will. The jamming signal is effective over a limited area - if that area includes enemy troops but not your own ... Plus there is the possibility that they have an encrypted channel just like the US does that will not be jammed.
"
There's more, but if you understand why the clock, you can figure out the rest.
"
I assume you are referring to using celestial navigation, which requires an accurate time to determine longitude. However firstly there are a multitude of radio time signals, it being unlikely that ALL are jammed, so a super-accurate clock is desirable but usually unnecessary. Celestial navigation is not very satisfactory. You cannot get an instant position during the day (because the only visible celestial object will only give you a one-dimensional LOP at any given time), and you cannot get a position fix if the weather is overcast - which could last for days.
It is possible to get a reasonably good position fix using any polar orbiting satellite if you have the frequency and orbital data for that satellite by measuring change in doppler shift during its pass. Also requires an accurate time and works in all weathers.
"
Have you ever tried navigating with GPS when not on a road?
"
Yes, both in fixed-wing aircraft and helicopters. Also when trekking in Nepal. No problem navigating whatsoever. Perhaps you were using a vehicle GPS which will try to "correct" its GPS position to put you on the nearest road?
Apart from doing a cross-check to see whether both AoA sensors are giving the same information, it would be entirely possible to perform a sanity check using other sensors. There is a mathematical relationship between airspeed, mass, pitch angle and AoA. Knowing any 3 allows you to calculate the 4th. The flight computer should have the current mass of the aircraft (which changes as fuel is burned), and airspeed and pitch angle can be obtained from sources completely unrelated to the AoA indicators.
But I question whether any aircraft that can become irrecoverably unstable should be certified for civilian passenger carrying duties.
Given a blind keyboard input to an unsupervised PC, I'm pretty certain I could do a lot of mischief without needing to see the VDU so long as I knew the OS it was running.
For example, if it is running Windows, then entering the "Windows" key is likely to get the OS to a known screen. From there you can run any program already on the PC by entering a known series of keystrokes. You could for example run a hex editor and upload a program by entering hex bytes of the binary, saving it then running it.
In many countries all sorts of places ask you to hand over your passport. Hotels, car rentals, employers, contractors etc. Never do it. I usually tell them that British law makes it illegal for me to give my UK passport to anyone else (in case it is used for terrorism), so sorry, I cannot do that. But they can have my British library card instead, which is just as good. Another excuse is to tell them that I do not have my passport because the hotel/car rental/police is already holding it (they're not). In quite a few countries you have to take a photocopy of your passport and fill in a form with details of your stay if you want to buy a local SIM card or for several other services & purchases. I always take several photocopies of my passport and a few spare passport photographs with me so that I don't have to hand my passport to the shop to photocopy.
Best company I worked for owned a Cessna light twin aircraft. So I got a pilots' license (PPL) on company expense, and "drove" myself to all on-site visits around UK, Germany and other places. There are numerous small airfields all over Europe, and bound to be one within a taxi-ride of where I'm visiting. There's no waiting around at small general aviation airports and no fixed departure times or cancellations to cope with. Plus I love flying. Bad weather meant I'd have to fall-back to car or public transport, but that happened surprisingly infrequently - visit dates were usually arranged with the customer after seeing the 5-day weather forecast. Also impresses the client when you tell them you are coming by private aircraft.
I was once stationed in a military compound. The buildings all had corrugated iron roofs that were supported by pretty large close-spaced beams. I eventually found out that the buildings were designed to be built with sandbags covering the roof to protect against mortar attack etc. The architect had done all the calculations to ensure that the roof beams were strong enough to support all the heavy sandbags. However what he had failed to take into account was that it would not be strong enough to support *wet* sandbags. So the buildings went up as designed, but it was decided not to fit the sandbags.
I used to think that the military back-pack VHF radio set my old company made was grossly over-engineered. Until I saw it being used as a base-block for the jack when changing the wheel on an armoured vehicle.
One of the batch tests it underwent was to drill a hole in the case, fit a valve and pump it up to 120psi. Then drop it in a tank of water - if any bubbles came out of it, it failed.
"
And the only way an extension will be granted is if the UK can show it will materially change something.
"
I disagree. I expect the EU will readily agree to an extension at the rate of, say, £350 million per day. Which we could finance from the NHS budget, perhaps.
"
... albeit it with a 58.2 per cent accuracy.
"
So a bit better than flipping a coin, though not much more. And this type of kit is what will be used to decide whether or not to arrest and detain random people in a crowd monitored by "AI" cameras. Surely cheaper and no less accurate to simply arrest everyone wearing a blue shirt.
If an applicant were to state "yellow" it could be an indication of a liver disease. It's not unreasonable to factor in a possible serious medical condition when deciding who to hire. If the job involves carrying heavy loads, it would also be advantageous to employ a person who sees themselves as a young mule.
There's not many things that would cause an aircraft that size to enter a nose-dive. Deliberate pilot action is one, and a sudden shift in cargo is another. It can also be the eventual result of an aerodynamic stall, which in turn can (and has been) caused by making a mistake when programming the autopilot with neither pilot monitoring the instruments as they should. Nevertheless, when the stall-warning and stick-shaker activates almost all pilots would react in a way that prevents the nose dropping as much as it apparently did here.
Yes, the cargo is on pallets or in bins, but they are on tracks in the fuselage. There are stops on the tracks that must be raised in front and behind each bin or pallet to prevent it sliding. If a heavy bin/pallet has a space on one or both sides (very common when a small but very heavy pallet is close to the maximum weight allowed in its station), and the loader has forgotten to engage one or more stops, it can and has caused a lot of damage. As an example - the locks have been engaged on the tail end of the pallet, but not the nose end, and the pallet is situated close to the tail end of its load station (compartment). All is fine during takeoff and cruise, and also cruise-descent which is a very shallow dive. But as the aircraft is pitched further down for a faster descent to land, the friction between the pallet and rails is finally overcome and it slides forward building up speed until it crashes into the pallet in front at the next station. The shock overloads the stops which break and now there are two pallets rushing forwards. Etc. Besides mechanical damage that may be caused to the aircraft frame, the C of G moves suddenly much further forward than the amount that can be compensated by the elevator, and the aircraft noses over and dives into the ground PDQ.
Not saying that's what happened, but it's a plausible scenario that requires only one point of failure (a forgotten bin-stop) that would be completely symptom-less before the sudden catastrophic event. Bins/pallets can each be loaded with several tons of cargo.
Another scenario we have unfortunately seen before - a suicidal pilot.
The flight recorders will soon make clear what happened, I'm sure.
"
Two engines out at that height wouldn't give them a lot to play with.
"
Total power failure would result in a glide followed by an emergency landing or ditching, not a nose-dive and crash. Given that they were over an open area of shallow water & mud, an emergency landing would have been perfectly survivable. Maybe you recall a total power fail situation at a similarly low height that occurred near the Hudson river a few years ago?
I don't understand either. It's well known that Windows operating systems are used on warships and other areas in the military, albeit not in direct control of weapons systems. I cannot see that the use of a company's generic product to help in military training is any different. Heck, there was a time when broomsticks were used to help train infantry personnel, is that a reason to castigate broomstick manufacturers?
"
Sorry but selfdriving vehicles are able to recognise much more than human
"
Is that why one failed to spot a huge frigging articulated lorry parked across the road?
Sorry but an "AI" can only recognise things and situations that it has been programmed to recognise. Humans (by the time they can drive) have had at least a decade being programmed to recognise a a far greater number of objects, object-sets and complex situations.
"
So the NNs do the same thing we do
"
Except we have not mapped only one or two data sets, but many hundreds of data sets. Often we need to make use of correlations learned from one data set and apply it to a situation that is normally the domain of a different data set.
For example, how we react while driving and an unexpected object appears in the road depends whether the object we see maps into the data set of "very hard solid things" (e.g. big rocks), "Soft inanimate things" (e.g. a bin bag blowing in the wind), "Small animals" (e.g. fox), "possible human" (e.g. runaway pushchair) or "Something abnormal" (e.g. very big pothole, open fissure, collapsed drain etc.). We also react based on our knowledge of human behaviour - e.g. a ball bouncing across the road is itself no danger, but may well be followed by a child chasing after it.
We can also recognise and appreciate the difference between a road that has a soft verge, a road that has a ditch next to it, and a road that has a sheer 200 foot drop next to it, and this will influence our decision on the best course of action to take in order to try to avoid a collision.
The "A.I." in a driverless car would be able to recognise a tiny fraction of the things we are able to recognise, because its "knowledge" might be quite deep, but is nowhere near as broad.
"
because $10/mo for 1TB of cloud storage isn't a terrible deal.
"
You think? Really?
You can get a USB 1TB HDD for the price of 5 months subscription if you need portability.
Or a 1TB NAS for the same price as about 1 years' subscription.
Both solutions are at the least 10 times faster, and will work fine when the Internet's down (which usually happens just when you urgently need the data).
And I trust the reliability of a HDD more than I trust "the cloud" as well.
"
Keeping the machine powered on is Computer Forensics 101
"
In theory perhaps, but in practice it's no more practical to bring a computer forensic expert to every police raid than it would be to bring a cardiac specialist to every ambulance callout, so it's only done when the suspected crime is extremely serious.
The search officers are told to take a note/photograph of the screen contents then turn all desktops off by the mains or remove laptop batteries without shutting down in case there is software that will wipe the HDD unless a special shutdown process is used.
The internal speaker was driven from a single binary output - it was either fed with a voltage between its terminals or it was not. Beeps and tones were created by rapidly toggling that voltage on & off at a suitable frequency *in software* (via a timer interrupt if the programmer wanted to be sophisticated). Thus no way of adjusting volume (until the CPU became fast enough to feed a supersonic PWM waveform, at which point it could produce full sounds rather than just beeps).
A farmer friend of mine in what was then Rhodesia regularly used his Rolls to inspect his extensive and rather rugged fields. One day after a liquid lunch he drove it into a ditch and broke a half-shaft getting it out. He took it to his local garage, who contacted RR (UK) to order a replacement shaft. After a few days he called the garage to enquire about progress, and the bemused owner told him that a team of RR mechanics had arrived, having flown in from the UK and hired his workshop for a week or two, saying they would sort it out.
When they left, his battered Roller had been restored to mint condition - and there was note left stating, "We found no defects on your vehicle. We have however taken the opportunity to perform a routine service free of charge under its lifetime guarantee. You will find a new Landrover registered to yourself parked on the forecourt - we respectfully invite you to make use of this vehicle, compliments of Rolls Royce for all future farm work, and we trust you will continue to enjoy your Rolls Royce appropriately."
I guess those were the days!
"
They were / are the cheapest effective solution for using windowed applications.
"
And a bloody sight better than the god-awful capacitive touchpads on laptops that are the bane of my life. Let your wrist droop a bit too low while coding, and the cursor is suddenly somewhere completely different which, if you didn't notice the switch, means you insert your new lines of code in a completely different place than where it was meant to go, ending up with a complete mess that takes hours to decipher what you had intended to do.
"
I remember TrueCrypt had something similar with hidden volumes but it all takes space and is detectable if you're looking for it
"
Yes, it takes a little space - but it certainly cannot be detected however hard you look. At most you might find signs that increase *suspicion* that a hidden volume exists, but nothing that can *prove* (even on balance of probability) that it really does exist.
"
Well, that is what happens when the government cuts the police budget by 40%
"
While that may be part of the reason, I do not believe it is what is mostly to blame by a long chalk. I subscribe to a weekly email that proudly gives all the initiatives and new activities the police are involved in, and most of them seem to be using money and resources with on wishy-washy activities with either no clear goals or no way to measure what they have achieved. Just how do you measure the effectiveness of a police initiative to "Improve diversity awareness," - and is it likely to prevent any crime? £20000 of my area's police budget tax was spunked on a web site set up to "help victims of online bullying."
ISTM that easy cases or cases that appeal to officers get prompt attention from car-loads of police, but those that may require lots of routine police work get shelved on the basis of "prioritizing".
"
The user experiences less than advertised connection and (rightly so) doesn't give the least about the cause of it, it is just the operator not delivering.
"
But it needs to be the result of *protracted* testing.
If a 90 second test at 14:15:09 last Thursday gave a result of 70Mbps download speed, would you be quite happy even though most of the time you hardly ever get more than 1Mbps?
While real-World tests are likely to be more accurate than the theoretical *if* they are done properly, the theoretical average performance is likely to be more accurate than a one-off 2 minute drive-through test.
Both data contention and signal path analysis are both very well established arts and give very accurate results *if* the correct parameters are used. Results that are more accurate than one-off snapshots of under 1% of the total area covered, anyway.
ISTM that if you used all relevant and already available data (power, height and aerial type of all the towers, topography, population density, cell capacity etc.), a computer algorithm could predict both coverage and average data speeds fairly readily. The physical test described would be far less accurate for any number of reasons - radio shadow areas caused by the car & equipment itself, temporary data congestion causing falsely low rates at the time of the test, belated cell handovers causing the download test to be done on a weak signal when a stronger signal from another cell tower exists. Etc. etc.
"
Let me ask the stupid question... if you aren't allowed near the printer, or don't know where it is, how do you collect the output?
"
In quite a few organisations I have seen, each print job has a header sheet with the name of the person who sent the job. Each job is looked at by the front-office person when it comes out of the printer to ensure it is bone-fide (e.g. not confidential information an employee is intending to take out of the building), and then the print job is carried to the person who sent it or deposited in their pigeon-hole.
Of course, that only occurs in companies that take security seriously. In many companies the cleaner could copy the contents of their entire server to a few USB sticks, or even upload it to their home NAS without the company even knowing it had happened.
The system itself may not be responsible for diagnosing a patient, but if test results were altered it could surely result in a doctor making the incorrect diagnoses?
The patient him/herself may have a motive for altering their records. e.g. to get cheaper insurance or to avoid having their driving licence revoked.
"
Bringing the casino nearer to the place where its illegal wont help will it?
"
The way I read it, it will help. In the US, it is illegal to gamble across states - IOW if you are physically in one state, a U.S. gambling establishment that is situated in a different state (or country) is not permitted to accept your bet. Phoning your local betting shop & placing a bet is OK, phoning a betting shop in a different state to place a bet is illegal. Casinos were not originally considered to be subjected to that law because casinos at the time required gamblers to be physically on their premises making it impossible to place bets from a different state. Someone has just realised that Internet casinos have now made such a thing possible.
The question I guess is exactly where an Internet gambling establishment is physically located. At the server? At the parent company HQ? At the location of any shared database? At the location of the bank where the bets are deposited?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
