Posts by Dr Dan Holdsworth 592 publicly visible posts • joined 16 May 2008
Assuming XX == female and XY == male is correct over 99% of the time. Naturally-occurring faults in the genetics of gender generally don't breed true, hence are heavily selected against by evolution (unless a different form of kin selection is at play; see also Social Darwinism).
That we now have a few edge cases merely means that an exception needs to be put into the software; something on the lines of "This person falls into the *other* category".
A very famous computer department in Manchester used to have a rooftop garden above their main machine room, complete with a pond. No prizes for guessing what used to happen on a regular basis; the pond leaked.
The only highlight of this was that every year for a couple of decades, one lone duck would build a nest in that rooftop garden, the fledging chicks getting a free ride out into more suitable habitat when the time came; photo-opportunities with the Computer Science duck were a great honour.
These days, the machine room is a big office. The pond is a skylight.
It still leaks.
Ahhh beancounters, where would we be without a bunch of cost-obsessed maniacs to get in the way?
Probably in the place of a certain northern English university that I'd better not name (not that I work there, but still).
This university had a sports complex with a flat roof. Not the smartest option ever designed for use in England, and like every flat roof ever, this one developed a leak. The University Estates people were called, and failed to fix the leak to nobody's surprise since University Estates tend to have the collective IQ of yeast.
So, the job was put out to tender, and the Powers That Be unfortunately included the beancounters in the decision. The problem was that there were only a couple of contractors locally big enough and reliable enough to do the job, and they were well aware of this and tended to cartel-like pricing. A third group also tendered, at a much lower cost; let us call them "Del-boy Trotter & co". The beancounters saw only the cost, and insisted on this bunch being hired to do the job.
Thus it was a week or two later that what looked like the contents of several scrap yards, plus exhibits from the Museum of Dodgy Roofers turned up. Shortly after Security tried to turn them away fearing a gypsy invasion, it was determined that these were the new roofers. Quite quickly, they got to work and started messing about with boiling tar, roofing felt and endless cups of tea, and a fortnight or so later pronounced the job done and asked for payment. In cash.
This was refused until after a couple of good rainstorms showed that yes, the leaks were gone, so Del-boy and hist motley crew were paid and duly disappeared, never to return.
Several thunderstorms later the leak re-appeared. The loft space of the sports complex was investigated, and it was found that the large plastic bin placed under the largest leak by the roofers was now full of water and overflowing.
One of the usual suspects for roofing was duly engaged, after it was discovered to nobody's surprise save for the beancounters' that Del-boy & co had vanished.
To this day the beancounters of this university have their every decision questioned with the words "But remember the sports hall!"
Remember all the calls for encryption products to have a back door in them for the use of police? Remember how our Government are completely sure that this is a good thing, completely sure that nobody will abuse the back door and utterly certain that nobody will do horrible and illegal things like install non-backdoored software?
This is more of the same.
This is politicians trying to change reality by averring that such a thing is so, when in fact it is not. Age verification absolutely has to identify the person who is having their age verified, otherwise it is useless. Therefore age verification will identify who looks at what, and said details will be stored on a government computer somewhere, and later lost on a train by some gormless civil servant.
Anyone with any sense will therefore give this age verification system the bargepole treatment, and obtain a VPN from an off-shore supplier. There are lots of different VPN suppliers to choose from, quite a few of which keep no records of what traverses their systems whatsoever.
Somewhere, therefore, an MI5 man is crying into his beer as formerly passable sources of Internet metadata go opaque. Even the fact that someone was using an off-shore VPN was (and probably still is) a useful indicator of either paranoia or nefarious deeds (barring business use, of course).
Alter the pr0n laws so that a VPN becomes a necessary adjunct of anyone who fancies a spot of executive relief, and all of a sudden VPN traffic becomes so common that it isn't a marker of dodgy deeds any more. Hey presto, that's another easy source of intelligence ruined by the politicians.
I would second this.
I heard a tale of a peacock sold by one farmer to another neighbouring farm. Periodically the peacock would decide to "go home", and would set off across the fields.
It would be found, stuck in the same too-narrow gap in the same hedge every single time. It never learned not to enter tight holes in hedges, and it never remembered that that particular hedge was the problem.
Truly, peacocks are incredibly thick birds!
Criminals are generally not all that smart, and suicidal terrorists are especially not-smart. However, most of the suicidal religious lot have worked out that the secure way to communicate is by meeting up face to face.
Once an attack is in progress, comms don't really need to be secure; if you assume that the security services aren't on the ball enough to know who all the participants are (a good bet if your little jihadi plot has gotten to live state), then you can also assume that they aren't going to understand the comms chatter quickly enough to make any difference.
That was the assumption the French terrorists made: they used completely unencrypted SMS to start and coordinate their attacks, and over that short time scale it worked.
A better idea is to try some sort of separation of fast cars and humans (like we do now with motorways) and in suburbia where humans and cars do encounter each other, switch on a lot of safety features on cars.
The next trick is to ban cars as much as possible from city centres, and provide instead very good car parking facilities around the city centres. A useful step in this direction would be to mandate that all traffic-related fines including parking fines must be paid to central government and not to local councils to prevent them from seeing car parking fines as a cash cow (this then forces them to see car parks as the cash cow instead).
Inside a city, use either slow autonomous electric vehicles or variations on bicycles, including hire bikes with electric assist. For in-city deliveries, use autonomous and slow robot vehicles, but human-controlled ones outside cities.
The only place a system that gets upset by people being in its way would work is on a motorway, and that is also the one road network that would greatly benefit in terms of congestion reduction from lots of automation of vehicles. Apart from that, about all you can do is improve vehicle safety and try to iron out some of the more moronic driver behaviours in software.
Take one moderately intelligent teen, with a smartphone. Let said teen Google "How to get to blocked sites", learn that a thing called a VPN is needed, then go looking on the Google apps installer on their phone.
There they will find the Opera free mobile VPN, and on installing it will find that porn sites become magically visible to them.
This knowledge will spread, rapidly. Teens are nothing if not extremely gossipy, and the knowing of how to get around stupid government restrictions will rapidly become common knowledge.
At which point one begs the question of why bother, if people are simply avoiding the inconvenient and frankly risable law?
Yes, ABS, traction control, frontal radar, lane-keeping and emergency brake assist all step in when the human has gotten into a bad spot and guess what, these systems all work and all save lives.
Semi-autonomous cars will not be designed by Daily Fail idiots, but by a combination of automotive engineers, actuaries and software designers, all of whom are ruled by intelligence, not by raw emotion. The changes will be slow and mostly not noticeable to the average driver, much as the huge leaps forward in car safety have not been noticed by the average driver, but they will be there.
What will happen is that the robotic systems will creep in and take over the bits of driving that humans find hard, and leave the easy stuff to us. Quite frankly we should be amazed that a bunch of apes that cannot run anywhere near 20 MPH can safely control vehicles travelling at ten times that speed.
The researchers here are not implying that semi-autonomous cars (which a lot of us are already driving, myself included) are 100% safe all the time, but instead that adding in protection against human drivers doing something bloody stupid does tend to reduce accidents and save lives.
If you take a broad view of this, then ABS braking is a primitive form of semi-autonomous car system; it removes the task of optimising the brakes from the driver. Emergency Stop assist systems do the same; drivers often don't jam the brakes on hard enough in an emergency. Frontal radar reduces the number of accidents where through inattention the driver hits a car in front.
Motorway Lane assist systems are another form of semi-autonomous car control system; they reduce the number of "dozy driver wandering about motorway" sort of accidents. This is the sort of thing the researchers were arguing for; not fully autonomous vehicles but systems which reduce the number of just plain stupid things a motorist can do through inattention or misapprehension.
One thing to note with the recent Samsung and Apple phones is that these devices are really, really fragile and are made of slippery shiny glass with minimal amounts of alloy. So, the people who are likely to see screen burn problems are also the people using these phones outdoors, without a case, in sunlight.
In other words, the people most likely to see problems are also the people most likely to accidentally drop their phones, shatter the glass and screen and get both replaced under warranty or insurance, thus rendering the screen burn issue moot.
This sort of thing goes on everywhere, but racecourses seem to be a magnet for it.
Years ago, my father and I were working on York racecourse as bookies, on the cheap side where the big hospitality tents are. York as a course has peculiar betting patterns; the punters bet like mad, flat out for about twenty minutes before each race, then about five minutes before the off everything goes quiet, and you'd better have a balanced book by then or you're stuck standing something.
Anyway, we were standing, hoping to get a few quid more on a mid-ranker horse returned at 8-1 with us, 10-1 elsewhere. No great matter; punters rarely compare odds. Up comes some chap with a tenner, rather more than we wanted so we told him he'd get better odds else where, even pointed out the better odds. Nope, wanted 80 for 10 with us, so we took it and made a backbet with next door of 80 for 8.
Two quid profit and I still don't know why the guy was so insistent on betting with us, and not someone else. The horse lost anyway.
Actually there are mechanisms for coping with toxic heavy metals in pretty much all organisms. In humans, metallothionine proteins are one of the main methods; these simply grab onto heavy metal ions and sequester them. Generally, a person's bones contain most of their sequestered lead, mercury, radium and so on, meaning that crematoria chimneys need fairly effective scrubbers to prevent the more volatile heavy metals like mercury from being re-emitted.
Reminds me of the several incidents that happened to a major UK University I could, but won't name.
First off, why you don't cheap out and put only VoIP phones in the datacentre control, when the networking kit isn't on secure power: power goes down, phones go down, management have no way to harangue the operators but the operators have their own mobile phones and can quite easily pass over terse instructions to the management.
Secondly, datacentres are air conditioned, and need aircon on all the time. So, when the power goes off, and the UPS generators kick in, that is not the time to wonder why everything in the datacentre is a bit less noisy than usual and seems also to be getting rapidly warmer. Cue very rapid machine shutdowns all round.
Thirdly, when you decide to turn the now rather obsolete datacentre into one absolutely gigantic office, it is unwise to assume that all the various odd machines that used to be in there will all migrate to the new, pay-for-space datacentres you've hired. No, they end up under various academics' desks, in comms closets and otherwise scattered around the place in silly places, and worst of all you don't know where they all are, so cannot apply blanket security policies without random roastings from, for example, the Professor of Cruel and Unusual Geography.
I'm not sure how Nissan managed it, but the Juke, despite being a smaller vehicle than the Qashqai, has worse real-world MPG figures.
As to the infernal beeping, odds are the machine thinks a door is open somewhere. Open and slam all of them, and that ought to do the trick.
One final thing: if you have a car with keyless entry, remember that you are the owner of a system that relies on "Security by Proximity". If a car criminal happens to turn up with a box of tricks which can amplify both the signal from the keys to the car, and from the car to the keys, then someone can fool the car into thinking the keys are right next to it, when in fact they are in your house several metres away.
This trick (and I rather think sophisticated criminals are building the boxes of tricks, then hiring them out to dimmer, bolder criminals) lets criminals get into a car and ransack it; it also lets them start it up and move it a few metres or so. Just far enough to get to the low-loader round the corner...
A better way to work is to remember that when you're hunting criminals, you are not hunting super-intelligent encryption-geniuses, but rather the less-able twerps of this world. As such, you simply have to accept that some of their communications won't be accessible to you, and there isn't a magical McGuffin that will let you get around this.
This is the same thinking process that police had to go through when DNA evidence was first introduced; all DNA actually shows is that at some point, the person whose DNA is present was in contact with whatever the DNA was detected on. Thus the old criminal trick of picking up cigarette ends outside dodgy pubs, then scattering one or two in prominent places when committing a burglary only works if you have stupid policemen around.
Another example is of some burglars who targeted country houses and operated as a gang. Their modus operandi was to meet up at a motorway service station near the target, turn off all mobile phones then go out to rob the target. Only afterwards did they re-enable their phones. This meant they didn't leave an electronic trail to their crimes, but did mean that they left a huge great signal that they were about to commit a crime (for they never met up, turned off phones then sloped off down the pub lawfully to add distraction to the pattern).
As I say, we're dealing with criminals, not masterminds. Criminals always make mistakes, and police have the manpower to catch these mistakes.
So, forget the phoney prize of being able to break encryption. If it is seen as possible, people will use other methods to get around this problem; unbreakable one-time pads for instance. Or, use encryption known not to have been back-doored.
The only reason that Uber looks good as compared to black cabs is that it is being bankrolled by VCs, who would appear to think that it has a good chance of putting most of the local private hire operators out of business, and possibly hurting the black cab operators as well.
Having said that, a wake-up call for the black cab operators is long over-due, not only in London but also in much of the US. Over there, a licence to operate a taxi cab costs hundreds of thousands of dollars and the numbers are being kept low by the dead hand of union protectionism.
Why on earth they cannot follow Samsung's lead and make a smartphone that, whilst it doesn't have curvy screen edges and incredible look and feel DOES have corner protection, a large battery and a case that will withstand being dropped.
As it stands these days, you buy a phone, take it out of the packaging, admire this thing of beauty and wonderous design then spend twenty minutes making sure it is completely clean before stuffing it into the armoured case where it will have to spend the rest of its days merely to ensure that the expensive thing remains undamaged.
If ever you end up in this situation, don't exact a terrible revenge on your colleagues and management before you go. Resign quietly and politely and maintain a dignified and civil demeanour throughout. This avoids antagonising one's former colleagues and leaves them with an impression of professionalism that may be completely unwarranted, but which means they will be at worst neutrally disposed towards you should you encounter them again.
Content yourself with the adage that Hell is other people, and that this particular bunch of other people will in your absence have been inflicting pain upon each other to a far greater degree than anything you could ever devise. Their incompetence is your revenge, your sanity is your reward. If you really can't help yourself, warn them about the Easter Egg (the one you did not leave, being far too smart ever to leave one) and let them tear the place up looking for one.
We don't really need artificial intelligence, not when humans are still fairly common and cheap to hire.
What we actually need is things that are about as intelligent as a cockroach. That means an ability to find a way around obstacles, enough memory to get bored with going in the same circles all the time, and an ability to recognise simple dangers such as pitfalls and walls, etc.
Do this and do it cheaply, and higher-level functions such as navigation can be dropped on top from conventional programs. This sort of thing is already sort-of happening with robot vacuum cleaners, but needs to get better to be truly useful.
All of this comes down to a trade-off between how strong the system can be, versus how much whine you are prepared to tolerate from the users. Since the users in this case are MPs who are trusted with state secrets and are almost the highest authority in the land, I rather suspect that it is they and their great power which is the main cause of trouble.
From a sysadmin point of view, even just the simple TCP rate limit function provided by UFW is useful, in that it stops single IPs from banging away at a machine. Fail2Ban provides a much better level of protection, especially when the "findtime" is extended enough that somewhat more clever botnet attackers are detected and excluded. The problem with both is that a fat-fingered or dyslexic user will get passwords wrong, and will repeatedly get locked out until they demand that the security levels be decreased for them.
This is why 2FA is so important and so essential; use 2FA and only the dozy users who cannot follow instructions get left behind, and the cure for them is simple: get their secretary to handle all the technology for them a la Tony Blair.
Done properly 2FA isn't difficult either for sysadmins or for users. Banks have successfully managed to get their customers to remember strong passwords and use 2FA dongles, and have managed it without much in the way of screams of agony from mentally-challenged lusers.
2FA for email is similarly not rocket science, and it is also not beyond the bounds of possibility to produce small, laminated instruction cards (laminated to prevent the poor dears writing their password on the card) which detail how to log in using the 2FA dongle. Tricks like this work wonders when you have thick users, or so I am told.
2FA plus Fail2Ban with suitably long time outs on the IP logger, together with intelligently-designed supplementary rule-sets such as a blanket ban on all Chinese, Russian and North Korean IP ranges and a strong and secure VPN for access from foreign climes which relies partly on ssh keys for authentication. Do that, and yes, any random script kiddie can have a pop at a dictionary attack, but no, said random script kiddie isn't going to actually get anywhere.
Suppose our putative Taliban is out on a battlefield, where people are actually shooting already. In this case, several factors both cultural and practical come into play.
Firstly, if our target is busy then he might not even notice bullet impacts around him.
Secondly, even if he does see impacts, he may just ascribe these to random battlefield stray rounds that aren't actually meant for him.
Thirdly, as he cannot see or hear a sniper (too far to hear the muzzle blast, and the rounds will be subsonic by the time they get to him) he may just think he's out of range and disregard the shooting as inaccurate fire that won't get him.
Fourthly, the man might actually be rather stupid, be that from lack of education, nutritional deficiencies early in life or even rampant inbreeding. Certainly anyone smart enough to realise the dangers of front lines isn't going to wander about willy-nilly in front of the enemy.
Finally, there is an attitude prevalent in that part of the world that predestination exists to a greater or lesser extent and that when Allah thinks it is time for you to go, you die; up to then no worries.
All of these plus the fact that he cannot actually see enemy forces might contribute to his apparent unconcern under fire.
A common Victorian design of lavatory for mills and factories consisted of a series of stalls with seats atop a porcelain gutter part-full of water, with a flush unit at one end. Periodically, this would be triggered and the accumulated turds flushed away.
However, a very common trick eventually forced a re-design of this system. The trick was simple: chuck a large ball of lit waxed paper down the hole closest to the flusher, then trigger it whilst some of the other stalls were occupied. This then burned the backsides of anyone not quick enough to stand up as the burning paper came past. Smarter pranksters generally departed rapidly before any scorched-arse victims could find them.
The re-designed system merely had partitions dipping down into the water surface in the gutter to extinguish burning items; mill workers in those dim and distant days didn't have access to metallic sodium and the like.
Good and fairly cheap VR would actually be a winner for business work, programming and systems admin especially. Instead of several large actual screens, the user would have several large virtual screens positioned around a virtual environment of their choosing.
In my case, a virtual "office" under a large, shady tree in the middle of a walled or hedged garden would be ideal and relatively cheap in CPU terms to simulate (not much moving scenery, no long views), as well as being a great deal nicer than a dingy office space.
From the point of view of whoever is paying for this, the graphics hardware is more expensive but only a smallish back-up monitor is needed, plus the virtual screens can be set up to appear to be big, but a few feet away so middle-aged eyes aren't struggling to focus. Status indicators for business systems could be integrated into the VR scenario; a compost heap represents the system's rubbish bin, a flower bed the core business systems and so on.
Best of all, if the office space is a cramped cubicle, then the VR space actually represents a better and more pleasant environment than the physical environment does, allowing the employer to cheap-out on the physical environment.
I feel much the same way, and plan to switch to shorter working hours at some point, but to carry on working as long as I feel able. I rather think that in many cases, work becomes a person's life, and without work they simply have not got very much to give their life structure. So, carry on working for as long as you can, but simply reduce the hours and pressure as much as possible.
A friend of mine had someone call up "This is about the accident you had...". It turned out that my friend had indeed had an accident, which had left him concussed and with very little short-term memory. There then followed a long and (for the scammer) most frustrating conversation as the scammer was mistaken for an online grocery, a vet dealing with the castration of a pet cat, a hospital, someone whom he'd forgotten the name of and so on, over and over and over again.
The scammer ended up beside himself with frustration, yet not quite able to put the phone down since he was never openly abused or mocked.
Nice if you want to indulge in a spot of smuggling, that is.
All you need is an HGV and a set of false plates, preferably plates belonging to a vehicle that the authorities either trust already or will find difficult to check out.
Quick, easy and simple and as long as you don't reuse the same plate too often or get caught physically changing the plates over, you ought to get away with the scam for quite a long time.
This reminds me of experiences I had years ago, training to fly hang-gliders. The group I was with were doing short flights off a valley side in the Dales. The wind was slowly getting stronger, had been all day. When my second turn came round, I did the usual "run like hell downhill" launch, but instead of the expected pull away from the ground, nothing happened.
What had occurred was that the wind speed had gotten strong enough that instead of laminar flow over the opposing valley lip, down to the bottom and back up again, we were now getting break-away rotors of wind peeling off the opposite valley lip. These turned the wind from a strong uphill flow to gusts and occasional dead air.
Visually, everything looked the same. Neither I nor any trainee had any clue that this might happen (although I reckon the trainers had it in their minds to watch out for). Now, imagine you have a random non-pilot in a computer-controlled aircraft, which at some point performs an emergency landing in a field somewhere.
The cause: a thunderstorm visible on radar, not so visible to human eyes. A danger of downwards microbursts, hail and strong winds, so the aircraft HAS to land somewhere to avoid the danger.
Imagine now that you're the poor helldesk techie on the other end of the phoneline as our unclued, over-paid businessman rants down the phone at the hapless operator about missed meetings, broken contracts and the like. Hell on earth as the moron customer is certainly not going to listen to sense, yet if the danger factor were ignored then his surviving relatives would certainly sue.
Similarly the customer whose flight gets stopped or diverted because of a NOTAM for Purple Airspace over where he wants to go. Insta-rant over delays, which is much better than a short, painful visit from the RAF for endangering the life of a royal.
This alone is going to prevent the widespread take-up of flying cars.
Handing over an unpleasant biohazard to a techie without the equipment or the pay to handle such is basically impolite, irresponsible and just plain stupid. I'm with the techie here; dump the bloody thing straight off and perhaps even feed it through a shredder if there's any chance of there being unencrypted data on the laptop.
If you look in your local equestrian-orientated shop and ask for something to stop horses gnawing things, then they will show you a produce called Cribox. This is capsicum plus something that tastes vile plus some sort of smelly stenching agent in a grease base.
It looks bad, it smells bad and it tastes on the far side of appalling, so I am told. It also stings like anything if you get it into an open wound.
I have used it to stop a frustrated squirrel from gnawing a garden shed; I never saw the animal actually taste it, but it stopped the gnawing alright.
Clearly the OP doesn't know how to safely and legally trap wildlife.
Best practice for the UK at least means that all snap traps such as the old Fenn mk4 and mk6 (soon to be banned for squirrels in favour of more certain-to-kill traps) had to be set somewhere that non-target species such as dogs, cats, roving network engineers and the like couldn't accidentally set them off. This generally means setting them inside a tunnel made either of mesh, or of some other durable material.
Rats, mice, squirrels and the like generally cannot resist the temptation to have a look inside any tunnel, hole or similar thing they come across, in case there is something to eat in there. This propensity can be improved by baiting the trap with peanut butter, in such a way that the bait is beyond the trap along the route the animal has to follow. This generally ensures a kill.
Mouse traps are a different proposition. Generally speaking, you get what you pay for with mouse traps and the cheap pressed-metal garbage off Fleabay are so insensitive that they don't work. Electrocution traps are best, and some can even be remotely monitored by SNMP to determine when they have caught something. Another interesting design is the Nooski trap, which uses an elastic rubber ring to strangle the poor victim. Not nice, but most effective and doesn't splat guts all over the place.
The final trick to try is the cellulose-based rodent baits. These work by dehydrating the animal to the point of collapse, but only work on rodents so anything else mooching along and eating the bait won't get killed. This bait has the other advantage of not actually being a poison, so no certification is needed to handle it.
Strange how everyone is thinking "Bomb" here. It may well be that some smartalec has worked out how to build an xray-transparent firearm, but the limitation of this is that it is quite bulky. So, the only way to hide such a device is to build it into a laptop.
Hence the ban on things over a certain size that contain various sorts of electrickery and thus look on Xray to contain wires, batteries and so on. The specific danger here is from the terrorist having his plastic gun with him in the cabin; it doesn't matter if the thing is in the hold because he cannot get to it during the flight. It also doesn't matter if some twit is importing a highly impractical firearm into the country; it isn't as if the engineering expertise to build working firearms doesn't exist here.
You don't honestly think that a civilisation that can build interplanetary megastructures is going to be sending actual live examples of its self between stars, do you?
Much more likely, they long ago started going down the cybernetic route and are now more or less indistinguishable from computers, or are even just software ghosts running on a computing substrate of some sort. This transport system you see isn't shifting biological entities, but instead compacted, backed-up software and information.
I have a University-based institutional subscription to the IEEE journal, and I can confirm that the observed inaccuracies are ONLY seen with three-phase meters, and then only in circumstances where large photovoltaic arrays are feeding power back into the grid via Active In-feed Converters (AICs). It seems that these AICs are not subject to proper regulation regarding how much electromagnetic interference (EMI) they may generate, and thus having been built to the lowest possible standard many of these AICs generate quite a lot of EMI.
The dodgy photovoltaic converters weren't the only EMI emitters seen; the drive systems for fans in one farm's barns were also very noisy indeed. Reading between the lines, I would think that quite a lot of electrical equipment on farms especially is going to be very noisy in EMI terms, partly through age and partly because with the old mechanical three-phase meters, it didn't matter a jot.
Having discovered all of this and built a specially EMI-noisy measuring rig, the researchers then went on to test the single-phase meters that pretty much all domestic situations will have. They found no deviation from the specification, and no influence from interference, EMI noise or distorted voltages could be detected.
TL,DR: No problem detected for household meters.
Facebook is an American company.
America has the Patriot Act, which effectively says that plod over there can shout "Terrorists!" then march in and seize data from any American company they so choose, with hardly a murmur.
So, if a person's Facebook record is of such amazing interest, the simple option for our idiot cousins over the water would be to ask the incoming suspect their name, then go over to Facebook, back their database up to that of the NSA and then simply grep through for info on that particular Facebook user.
Simple indeed, and not done because this trawl has been dreamed up by lackwitted goons in the ports, not in the American executive or legal hierarchy.
Worse yet are power users, that is to say people with a little bit of knowledge but not nearly enough wit to realise how little they know.
Many years ago I took a support call whilst working for a particular dodgy ISP. Their main site name was $FOO, but their infrastructure was still named $BAR because nobody could be bothered to change the domain names.
So, when this power luser sees his machine connected to $FOO internet regularly making connections on port 53 to machines in the $BAR domain on his newly-installed firewall, he panics and uses this firewall to block these connections, thinking it to be a hack. Then, as he put it, this terrible hacker must have done something truly appalling to his machine since the Internet went ever so slow.
As slow, in fact, as a machine trying to connect each time to its primary DNS, getting blocked and timing out to hit the secondary DNS server...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
