Schneier on Security

Clive Robinson • August 18, 2018 1:35 PM

@ bttb, ALL,

“Botched CIA Communications System Helped Blow Cover of Chinese Agents”

There is a bit more to it than just “China” we know that the system was previously used in the Middle East but there was no apparent “push back” at the time.

It’s safe to assume that both Russia and Israel and several EU nations were more than aware of the system and it’s defects. However as “The CIA is friend to no man or deity” it’s “quite likely with a high degree of probability” that various SigInt entities just looked the other way…

But there is more to it, than that for a couple of reasons. Firstly various Middle East nations are no slouches when it comes to ElInt and SigInt and the CIA and various other US IC entities have continuously under rated them and been caught with not just their jaws dropping… But secondly, the CIA has a history of being technically not that bright in house, thus over estimating technology vendor promisses… Which has resulted in over ambitious technology rollouts…

It is therefore more than likely that it’s rather more than the Middle East and China this system was/is used for.

To be blunt I’ve said it many times before on this blog and other places,

The Internet as it is currently used is not nor can it ever be secure

It’s a mistake Tor users and the Dark-Net are begining to learn.

There are ways the Internet can be made more secure and I’ve mentioned them a number of times when talking about Tor’s deficiencies. But the one thing you can not hide is the fact there is a node connection to both the Internet and more secure overlaying network.

Which raises the question of how you might anonymously connect to such a node, the answer to which most people will not like…

In essence you need a “sacrificial goat” that acts as a “cut out”. The agent uses old school Opsec to “one way” send messages via dead letter boxes and the like to the goat, and the goat then puts them in at a node. The goat like the SOE radio operators during WWII are expected to have a very short life time. Thus these days they would never be given “plaintext” as they will be expected to be tourtured in various ways prior to being “discarded” by those capturing them.

The important thing to remember which everyone seems to forget is “technology rarely has deniability, and is almost impossible to destroy”…

Where as a cigarette paper with an invisable nitrate or permanganate ink on it “burns in a flash to dust, that is indistinguishable from ordinary cigarette paper used with tobacco”.

Such nitrates are very common and get on peoples hands when preparing cured meats such as salami or bacon, or handling tobacco.

The fact something is “new” does not mean it’s better. Likewise because it’s “easier to use” does not make it more secure. Two things the CIA appear to either not know or chose to ignore for some reason.

Oh if people think I’m being a little hard it’s not just the CIA, though the dude with the ginger wig in Moscow was a hoot. But If people think back to the fake / talking rock in a Moscow Park, that was supposedly “bluetooth enabled” or similar, that was alledgedly a British IC device, so yeah getting the “high tech bug of doom” is not just the CIA.

The actual problem in Spy Craft is actually “the last mile” problem. That is how does an agent get the secret message over the last mile to their handling officer? It’s safe to assume all officers like all diplomats are watched 24×365.25 and have their homes and offices bugged when dealing with the likes of the Russian’s or Chinese. Even the cars, cafes, restaurants, toilets within a mile or so are going to be bugged directly or indirectly (laser mics etc) as a matter of normality. As the price of technology drops this surveillance capability ceases to be just the province of the Super Powers, all first world and even some third world countries can now afford to do it…

It’s actually increasingly difficult to solve the last mile problem, some actually believe it’s now nolonger possible to do it either with technology or old school Spy Craft and OpSec.

I’m not so sure, there are after all “stray cats” that could be trained[1] 😉 But the question arises as to just how small can you make an autonomous drone/robot to act as a goat?

[1] Not surprisingly the CIA is reputed to have had a “stray animal” plot to eliminate Castro. Is it true? Quite probably, during WWII thousands if not hundreds of thousands of plots were thought up and drafts put on paper for consideration, few would have made it to the planning stage, less still to operational consideration. It’s actually known that “James Bond” author Ian Flemming was involved with the formulation of a number of plots and part of the planning of a plot to assassinate Hitler, that came within a wisker of getting the operational green light. He was also involved with “The man who never was” which took the body of a Welshman who had killed himself with rat poison and turned him into an army officer carrying important papers who had been shot out of the air. This plot was green lighted and went into operation and reputedly saved many many lives on D-Day.

Clive Robinson • August 18, 2018 2:06 PM

@ Humdee,

I’m not sure either is true but I am eager to hear what readers of this blog think.

Well a hash will match any number of files by simple logic. The probability of a match between two files is often quoted as the square root of the hash size from a simplification of the Birthday Paradox. So from that if the hash is 128bits in size the probability will be 1/(2^64).

From an idetification perspective a hash is a one way function, that is you can find the hash of a file fairly easily, however working a hash back to a file is not considered realisticaly possible.

This has an evidentiary asspect. If I give you a hash you can not identify the file it is a hash of unless you have either hashed the file or somebody else has and you can thus compare the hash you have been given with the hash you are already aware of.

Thus a hash is not by it’s self an identifer of a file, let alone it’s contents. Further there is absolutly no guarantee of uniqueness of a hash, thus there is not a unique mapping even if you do have a copy of a hash of a file, it could be for another file altogether.

Having got that under the belt, people can then make their own idea as to “the burden of proof” and “beyond reasonable doubt”.

One thing you can be sure of though, if a successful prosecution is predicated on a hash uniquness, if the defence brings two different files into court that match the hash in question then the “beyond reasonable doubt” is blown right out of the water and a sensible prosecutor will call a halt to the case, unless they want an adverse judgment and thus case law.

The thing is that we know that at some point two files matching a hash is going to happen. Not just because it has already, but simple logic predicates it happening, just as the mile counter in your car will clock over to zero and count up again if you go the distance.

Clive Robinson • August 18, 2018 6:39 PM

@ Anura,

According to the comments section, this is not a cryptographic hash but a proprietary algorithm that tries to determine whether two images are the same

If you go back and read my comment I did not mention cryptography. It was quite delibetate as was my final comment about a cars distence recording device.

Befor generic crypto hashes there were ordinary simple mathmatical or logical hashes used for various purposes. Primarily to provide a short equivalence of a “primary key” of the file or data object of larger size, thus would save both RAM and time for various functions.

The point is those “general properties” of hashes “one way” and “semi-uniqueness” or “collision resistance” still apply to the MS image hash, thus the rest of my comment still holds as a general comment.

What I did not want to do was get into the speciffics as they can be used to add the “smoke and mirrors” of “faux technical argument” that is used by certain legal types to confuse non technical members of the public of which judges are just one quite small but important segment.

I don’t know if you were around shortly before the turn of the century when “Digital Watermarking” was the latest buzzword looking for venture capital or buyouts by the likes of Disney etc. Put simply they added noise to the immage that was the equivalent of LPI radio systems using Direct Sequence Spread Spectrum (DSSS). It appeared to those with a software background to be a “neat idea” and those with a radio engineering background who sounded notes of caution became the targets for accusations of being “Buzzkills” at best and NIH idiots or new market wreckers. As wiser heads predicted it was a faux market the final nail as far as many were concerned came from Ross J. Anderson and his Cambridge Computer labs researchers. The program they developed twisted the image in two dimensions. Not enough for the human mind to notice but enough to make the DSSS signal effectively usless as it had to be sufficiently above the image noise floor as to be quite noticable to the human eye to be readable after the manipulation.

What has not happened is the chasing of the DRM Holly Grail of perfect digital watermarking as undoubtedly there is big money to be made if it can be done. Obviously adding signals had been shown to be a failure, so people started investigating things like fractal compression to pull out image features in a broad sense. Comparing the fractal compressions would give an image related fingerprint rather than a watermark, thus various techniques were tried to turn fingerprints into inherent watermarks, generally with poor results.

From what you have said the MS hash sounds like it might be a spinoff of the fractal or similar compression technique then followed by a space reducing hash function.

To be honest there are quite a few ways you can do this, however there is a considerable gotcha. Whilst the final hash may have quite good collision resistance the preceading image compression system may well not. Worse it is quite likely to be variable in nature that is certain image factors will have high collision resistance whilst others very low collision resistance.

Such failings if there are any will more than likely be “covered up” under “commercial confidentiality” or similar. In a lot of cases that might not be of importance, but lots of people have been forced into guilty pleas by bad forensics before.

One area is human finger prints, people forget that matching them is not a science, just opinion from a human or AI system.

However few fingerprint matches are ever challenged in court (you get told judges do not like such behavior…). Oddly the few times they have been chalenged, they have been found quite wanting.

Now this could be the fact that only the truely innocent challenge them thus the identification was false to start off with but quite a rare occurance. Or it could be that most matches are bogus in some way and it’s only the fear of death penelties, long jail terms, bankruptcy or representitives refusing to challenge the evidence that lets such a bogus system survive.

It’s interesting to note that the rich defendents get representatives that tenaciously challenge evidence, and that the rich are far far more likely to be acquitted than the poor. When you add to that the information comming out about unreliable or non existant evidence from “justice projects” that take up the cases of those who appear to be wrongly convicted that you start to wonder what the heck is going on in the judicial process.

Then when you start to take a good look at “forensic science” you quickly start to ask yourself “Where is the science?”, then after a little more digging “Where is the integrity?”

Way to many “forensic examiners” get a pass on all sorts of nonsense due to the “in awe” effect of cop shows, so much so it’s got it’s own term “The CSI effect”.

But… As with “going dark” the “proffessionals” put a reverse spin on it to try to excuse poor performamce or get more resources etc.

To see this consider this from the Wiki page on the CSI effect,

It most often refers to the belief that jurors have come to demand more forensic evidence in criminal trials, thereby raising the effective standard of proof for prosecutors. While this belief is widely held among American legal professionals, some studies have suggested that crime shows are unlikely to cause such an effect, although frequent CSI viewers may place a lower value on circumstantial evidence.

That last lines gives the game away, the CSI effect has caused jury members to quite rightly disregard much circumstantial evidence[2]. However it’s also been found that to much regard has been given to expert witnesses by juries when the expert looks and sounds credible in a sciency way. That is they fit the jurors expectations in various ways (just as con artists do when going for the sell).

From what you have said I suspect the method is not just new, but also not independently tested, let alone “time tested” as I said this may not matter depending on how it is used, but these sorts of thing tend to suffer from “mission creep” as a larger usage means more returns / profit.

One of my many interests is looking at forensic testing and working out how to cheat it for both false negatives and false positives. I’ve found out a number of times new tests are ridiculously easy to find not just fault with but how to falsify the results. I think I will in the near future be casting my eye over the MS system with interest if I can get sufficient information to be able to do so.

[2] One reason that the legal proffessionals get het up about the disregard for Circumstantial Evidence is that it’s the type of evidence most likely to be presented to a jury for various reasons you rarely get to here the whole truth about. That is what they don’t like to mention is it’s also the evidence most called into question in appeals because it’s sufficiently often bogus[3]. Or carefully selected to hang an entirely false narative on by prosecuters of those known to not be able to aford an independent representative. That is rather than have primary indicative evidence the prosecution has conjecture and slight of hand presentation. Which lets be honest should never meet the burden of proof of beyond reasonable doubt.

[3] One classic example of just how bogus circumstantial evidenve is, is “the line up” where a witness points a finger at the defendent and says “It was him” or equivalent, it looks dramatic and in the past used to swing juries to the prosecutions story.

The fact is that juries being skeptical of such witness IDs is actually a good thing for justice. Because independent testing shows that given correct identification procedures only 10-25% of people can be positive about their identification selection even when they get it right… Which often very few people do when given ten sufficiently similar in build, hight, hair colour and clothing style people to pick from. Other investigations have shown that the most disheveled or cowed person in a line up is more likely to be selected. But the scary result is the body language, tone of language and words spoken by those officers accompanying the witness at the line up is often used as guidence by the witness…

Clive Robinson • August 19, 2018 6:43 AM

@ Australian Maths…,

Evidently the legislation requires “assistance” but unless something nasty is known about RSA and AES, I can’t see how assistance can be provided for the few end-to-end encrypted solutions.

As has been said befor “Beat cops, want message contents, intelligence operatives want traffic flow”.

The legislation thus courts are slowly starting to move towards “traffic as evidence” as well with the burden of proof thrown onto the defendant to show their activities are innocent. With a little thought you will realise that with the distinct lack of modern device/OS/App security putting an individual in an impossible situation is a lot lot easier than obtaining real evidence of illegal activity.

Traffic activity is easy for the authorities to requite and nigh on impossible for an average user to stop. All that has to happen is that the point you get your Internet connection your packets get “tagged” by your ISP etc. You can not see such tagging but we know from various sources that it is actively being perused by major carriers wanting a slice of the Google Money. Even VPN suppliers have been caught out doing this sort of thing. In the US those tagging records are very definitely “business records” and you have no rights what so ever ovet them. In most other juresdictions they are likewise covered as “engineering” or “technical” records over which you have no rights.

But in most cases tagging is not required at all. All that is required is your DNS request records. Even for end to end encryption on peer to peer networking your system usually goes through a “discovery” process to find the remote end, whilst it might not be the DNS servers you use, you will in most cases end up using a third party service “in plaintext” as traffic or for the search on the server it’s self.

Which brings me around to your closing comment,

I sleep well knowing I am not actively or passively undermining anyone’s technical solutions, ironically including those of government.

Don’t “sleep well” as even “inactivity” is “activity” especially when it comes to privacy and freedom. If you do not actively defend your privacy and freedom then others without your interests at heart will assume you do not care about having them, and by the time you start to protest you do value them it will be way way to late.

I could give a whole list of quotes about eternal vigilance, refreshing trees with blood, and more prophetic poems about those who come for others, but you can rest assured that those that do not learn the lessons of history condem themselves and others to relive it.

Clive Robinson • August 19, 2018 7:40 AM

@ Sancho_P,

Also Orin Kerr didn’t detect the serious flaw in Judge Ho’s opinion when quoting:

Yup, the fatal flaw in all the legal argument that it is not a search is that “hashes are most definately not unique” nor can they ever be (see Cantor diagonal argument for why).

The simple example most can relate to is, as I said the simple mechanical milometer / odometer in a vehical. It has a limited range of numbers (just under a hundred thousand miles) and they can and do roll over. As there are way way way more vehicals with odometers built than the hundred thousand mile display ability then simple reason alone will tell you that by far the majority of odometer readings are not unique. As the number of veihcles made is easily well above the hundreds of millions the odometer would have to be a lot lot lot larger, but still you would have nonunique values in the majority of vehicles.

Hashes of their various forms all have this nonunique failing. Thus knowing a file has the same hash is not nore can it ever be a guaranty that the file contents are the same, or worse still approximately the same.

In fact if you read the various articles you come across another point that both the Judge and the Proffesor either did not know –very unlikely– or chose to ignore.

A pornographic image is only poronographic in the eye of the beholder. That is under law it is the opinion of the viewing officer as to if it is pornographic or not.

Thus if the officer has not seen the image the officer can not make that determination.

So if the hash says it’s “possible porn” and the officer decides under the rules they work to it’s not porn where does that leave us?

Well at the very least the officer had made an unjustified viewing of the files contents. The only way that should happen is by use of a warrant and presumptive evidence submitted to the signing judge or witness.

But this is where it breaks down, the officer presenting presumptive evidence has to have a chain of evidence to support the presumption.

In the UK there was a harassment case by an individual against British Gas. British Gas’s defence boiled down to “the computer says”. Law lords ruled on the case and made it abundantly clear that it was not a valid argument let alone defence. Put simply it was pointed out that computers had no mind let alone a directing on, thus were incapable of forming an opinion. Computers were designed by man, programed by man to do mans bidding. Thus the responsability was firmly and irrevocably in the hands of man. Thus any mistakes made by a creation of man were mans respondability.

So a computer can not stand as witness, nor can the product of it’s programming. The responsability lies with an individual to check the computer output…

Don’t check it, and it’s found to be false, in a criminal case presenting false representation is a serious matter…

The officer had no way of checking if the hash was to actual pornography or not, and should most assuradly have known that as should the judge and proffessor. There fore the responsability was on the officers shoulders to view the file. If a warrant or other legal authorisation was not used then the activity of the officer was illegal and they should by rights be punished for their behaviour.

As others have noted the learned proffessor appears to have such a high ivory tower his reasoning takes flights of fancy due to the rarefied atmosphere atop it…

Clive Robinson • August 20, 2018 7:07 AM

@ echo,

All of a sudden the news starts getting serious again…

The MOJ thing is an expected result. It should be clear to most people by now that the likes of G4S, Serco and all those other barnacles on the UK economy are only in existance because of a political mantra that is akin to saying “anti-fouling is not a worthwhile activity”. Thus Gov abdicates on it’s duties to it’s citizens and that “glorious free market” mantra creates gross ineficiences and overloads the workings causing early failure. But the mantra has to go on. A bit like the hand of a drowning man holding aloft a camels straw to wave like some great hope for recognition and being pulled to safe harbour.

Thus the summer news “silly season” that waxed so well starts to wane again. But before it goes too far your comment of,

mansplaining about evidence collection,

Caught my eye. Whilst you have my sympathies for the predicament you were in, you also have my admiration for resisting the urge to take the polish of your toe cap on the offending bum.

It’s actually the invented words that are a kind of itch with me “splaning” is a word with atleast two hundred years of history in it and thus stands without the need for prefix, of which bro- sis- mo- nerd- elder- woman- and many more have been coupled to it since man- most have happened within a small handfull of years and almost all have had their meanings or intent changed atleast once.

In times past we have rightly criticized “managment speak”, “Politically correct speak” and many more. George Orwell for all his other observations had great cynicism for such behaviour and it’s attempts to not just divide society but actually fracture it thus making it easier to conquer by authoritarian behaviours.

But it also misses the point, many of these word couplets come into existance due to “hurt” and the “you just don’t get it do you” feeling. Thus they start as linquistic “pull ups” then become weapons at the hands of others who then encorage them to become dividers for their gain.

What it says about the current state of society in WASP nations where it predominates I’m not entirely sure, but to me it does not feel at all healthy. Thus I see it as in effect a symptom of a deeper insult in society, which is most likely a threat to the wellbeing of society which inturn threatens societies security.

Clive Robinson • August 20, 2018 6:42 PM

@ Bob Paddock,

As the Thomas article notes,

According to Department of Defense Directive S-3600.1 of 9 December 1996, information warfare is defined as “an information operation conducted during time of crisis or conflict to achieve or promote specific objectives over a specific adversary or adversaries.”

Is a fairly broad definition and thus it’s scope is the bounds of the “information operation” which depending on how you view “information” and “operation” could be wider than many assume.

For instance I’ve discussed information on this blog several times in the past and said that for human interaction it has to be impressed or modulated onto energy or matter to be,

1, communicated.
2, Stored.
3, Processed.

The implication of this is that whilst the laws of the physical universe apply to energy and matter there is no requirment yet given for them to apply to information. But also there is the implication that the physical universe we live and interact with is infact a proper subset of an information universe.

With such a view of “information” the implication is “no bounds” for information and only the laws of physics as bounds on the energy and matter information gets impressed upon.

Thus the problem moves on to “What is information?” and this appears to be well neigh impossible to answer in a practical or engineering way.

As I’ve pointed out with “data shadows” information does not actually need to be there, it’s absence is also information. For instance if you have a “listening post” you will scan the EM Spectrum looking for signals you can detect. Detecting a signal is present provides one bit of information, but conversely not detecting a signal provides the same bit of information but of opposit state.

However regardless of the state, the bit of information is insufficient. The detector can and will create false positives and negatives due not just to external noise but also due to internal issues such as metastability. That is our ability to measure information is far from perfect.

But it gets worse as I’ve pointed out a few times in the past complex modulation whilst determanistic in origin may appear to be either chaotic or random to a third party observer, who might well decide it is “burst noise” from as far away as celestial object behaviour or as close as devices in the detector, or background noise such as from thermal sources. This is the idea employed by “Low Probability of Intercept”(LPI) systems developed for military and intelligence communications systems and still continues to day with the likes of MIMO systems where multiple signals contain a fraction of the information and need to be received with the correct propagation path / time delays to produce a coherent and information compleate signal.

In effect this causes a “right time and place” effect for the receiver. Arguably everybody in range picks up the information from the multiple transmission sources but it is only coherant in a single place and time.

Combine that with the idea of data shadows, and you can quickly see that we may well be bathed entirely in information at all times but can only occasionally by luck be in the right time and place to be receptive to it. The more people / detectors there are in any given area the more likely the probability of detection and coherent reception…

We can not currently investigate let alone validate such apparently “off the wall” ideas, because ironically we lack the information to do so.

What we do know is that a neurons action potential is very small and requires tiny fractions of energy to become activated, further we have demonstrated both magnetic and voltage fields considerably external to the neuron causing it to be crudely activated.

So far all we have is a simple idea of what goes on with the idea that a neurons output can be defined by,

f([N1…Nn])

Where N1…Nn is the inputs to the neuron that subjects them to some unknown function f.

Our Machine Learning so far is limited to simple linear additions of the input signals, however it is far more likely that the response will be to some power curve of each input signal.

We can say this from the simple observation that the input signals are effectively pulses of variable repetition rate and that for any given integration process there needs to be a discharge process to ensure stability. In nature such discharge processes are usually “constant percentage” or exponential in nature.

Thus all we can realy say is that our current ML models are unlikely a match for neuron behaviour but might at some point be broadly similar in performance, oh and we still have no clue as to why they produce the results they do, even though we can mathematically model them…

Clive Robinson • August 21, 2018 4:38 AM

@ Wesley Parish,

Uri Avnery was one of the very few, perhaps the only journalist in Israel to decry the actions of the IDF on “Bloody Monday” this year when the US President and others of ridiculous demeanor were cavorting around the new US Embassy.

Uri Avnery pointed out yet again the IDF lies and propaganda, that appears to be the only story allowed in Israel’s MSM by the current leadership. Used to keep the Israeli citizens and residents in an Orwellian NewSpeak world of “glorious victory” afar.

But it is not afar an hour or twos drive at most to where the carnage was being created. It is unknown to the rest of the world just how many rounds of ammunition the IDF fired at unarmed protesters well away from the IDF imprisoned in what is a concentration camp created behind the fences and barricades the IDF have errected. All we know is that over a thousand struck and injured or killed those unarmed protesters.

So no “glorious victory” but a massacre and diplomatic protests from much of the world (but apparently not the US) flooded into Israeli embassies. The IDF were thus forced to end their sensless shootings, but not their violence.

God alone knows how many times this has happened and how many times it will have to happen again. Those who lead Israel currently are proving two points.

Firstly they are confirming Einstein’s definition of madness, secondly they are yet again failing to learn histories lessons.

Uri Avnery knew this as much as any man, and spent his adult life trying to get the leaders of Israel to get off of the “wheel of pain” they have put the citizens of Israeli on for so many years.

Peace is only achived by the existance of equality of respect, opportunity, and trade between peoples that leads to both trust and peaceful coexistence. Something Uri Avnery tried to bring about right to the very end of his life.

May he and those he fought so hard for find ultimate peace.

Clive Robinson • August 21, 2018 10:23 PM

@ echo,

I can tell you conversation becomes 50 shades of boring… At least I get better service in the shops.

Hmm there is an odd exception where you get both…

Late last year I had the misfortune to bend over to get a 2ltr bottle of fizzy spring water at one of the new chains of supermarket. I came too sprawled on the floor, with a shop assistant who was the store first aider kneeling beside me.

She informed me that an ambulance had been called and that I should remain laying on the floor, in front of what was the lower cost crisps / potato chips display… Which I was to find out was not a good place to by lying around.

Well what do you talk about after you have exchanged names? The old British stand by of the weather was soon exhausted. Meanwhile women with small children in tow were passing by, the children could see me but the women well, they just stepped over me to get at the crisps for their little bundles of stickiness (or worse).

Now whilst I am not quite the 600lb gorilla or elephant in the room I was definitely getting “the wall flower feeling”. I had a sense of unreality I was getting less attention than the crisps were.

Time draged on and after an hour of laying there my conversation level had dropped way through the bottom of even the deepest of sunk barrels. There is a sort of artificial bright brittleness of tone you hear in the voices of people trying to sound chearfull and keeping their end up as it were, and you could hear it getting into both the first aiders and my voices. Worse I think the conversation had turned at that point to the merits of local primary schools, not that either of us had a need of such establishments… We were just compleatly out of place as people just wandered by with their trollies and off spring looking at the shelves around us as though we were in some alterative universe.

Then another member of shop staff appeared, she had just had some form of asthma attack and poor thing she had serious wheezing when breathing both in and out, and did not look at all well. The security gaurd went and found her a chair to sit on and some colour came back to her cheeks. So there the three of us were making a large island in the aisle with a steady stream of shopers passing by like waves lapping around a sea wall. There was a bright spot at this point when one of the other staff came to say that an ambulance was definitely on it’s way… Well there was the first aider still on her knees talking to the other member of staff who nodded replys, so I at least could settle back on the now icy stone tile floor and start counting the ceiling tiles in my head and pretend I was not where I was.

It was around this point that one child pointed at me and said “mummy Santa Claus” and pointed at me… Well what do you say? I felt like I should leap up point back and declare in a low menacing base voice “You’ve just made the naughty list”… But no I just sighed and smiled resignedly whilst the mother draged the offending mite out of sight around a corner. But atleast we had a new conversation topic “What are you doing for Xmas”… Which we must have got a good fifteen minutes out of with the poor lass in the chair nodding along and gasping out the occassional answer. But there are only a handfull of ways to “stuff the turky” so that conversation petered out…

And so it went on, two hours had gone buy and I only half jokingly said that I’m supprised some one had not come around and stuck an over due parking ticket under my chin. The conversation had got that low. Anyway the girl in the chair was beginning to look almost normal and I was getting a nasty cramp in my back when finally the ambulance crew came around the corner and asked who the patient was. The first aider pointed at both of us and as a last ditch effort to raise my spirits, I said that we were a special offer two for the price of one, which atleast got a smile from the girl in the chair at last.

Well the medics started taking observations and asking the usuall questions and I spotted a worried look from the girl who was checking me to her colleague. So the got me up on the bed by a series of undignified posses and a second ambulance with a specialist appeared and some how I refreained from making the bus joke. Well the last I saw of the young lady in the chair she was being given oxygen as they wheeled me through the trollies at the checkout and out the door, again with hardly a look from the shoppers. I was definately getting that 1975 “Stepford wives” vibe.

In the ambulance they started the “big questions” such as “next of kin” and “Do you suffer from AF” at which point I knew I was not going to be going home for a week or so… So off we went on blues and twos to the hospital.

But the good news was the lady medic had a pink pig key ring and I asked if she had a name for it, and we got chatting about the merits of keeping pigs as pets, which was a lot of fun and brought some sense of normality back to me after the compleatly wierd and surreal experience of being invisable on the floor of a supermarket for two hours, as though I had a Douglas Adams “Somebody Elses Problem” field around me.

Then at the hospital hard reality hit me like a 2 by 4 to the pit of the stomach there were people waiting at the dock and they rushed me into “resuscitation” where the look on the triage nurses face said rather more than her words.

The on call consultant appeared like a genie from a bottle and the crash cart appeared under her hand and needles went in all over the place as she rapidly called things out. Then I started to feel deeply sea sick and out of breath, cold sweat poured out of every pour and my body wanted to evacuate every thing at the same time, I just wanted to get up before I disgraced myself. Then my hearing went and my eyes went out of focus and it went dark as atleast four people started doing things and I felt as though I was on fire, and all I could here was a distant voice saying “stay with us” as I was thrown flat on my back… Then out of the darkness the whining of a machine in distress came back to me voices calling out things and the eratic rapid beeping of another machine and a voice talking about inverting and then ketamine and I remember saying “I’m not a bl**dy horse” to which I got an amused “Welcome back” from the consultant. Apparently my blood preasure had just fallen through not just the floor but Hades as well. Anyway they got me something towards back to normal and I started to ask questions like why was my ecg rate on the monitor over 200 and dancing around like a flea on a hot griddle. I quickly got the feeling that I’d not dodged a bullet so much as taken a ride on one, and only just fallen off in time.

But all that aside I still renember more of lying on the supermarket floor for two hours trying to have a conversation with nothing to say not even small talk.

But I am left with an itch I can not scratch, I still have the urge to put that little mite on the naughty list for calling me Santa Claus…

Clive Robinson • August 22, 2018 4:31 AM

@ echo, All,

Lawsuit against Christopher Steele dismissed!!!! Yay!!!!!!!

Unfortunately for Christopher Steele that is just one of many court cases he has to fight over the dossier.

For instance he has one pending in the UK where the courts are natoriously plaintiff friendly even after the 2013 act[1] and there is no right of free speech (there are however “public interest” and “National security” get outs along with “privileged speech”).

Thus he has a number of hurdles to go befor being free and clear.

[1] For those that want a little intro to UK Defamation law,

http://kellywarnerlaw.com/uk-defamation-laws/

Clive Robinson • August 22, 2018 8:01 AM

@ MarkH,

The wording of the [4th] amendment says “probable cause,” and the conventional interpretation in U.S. law is “more probable than not.” In other words, 51% confidence is sufficient.

The subject is “illegal searches” which brings in a whole load of questions about how the “probable cause” was obtained and goes back to the,

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated…

As I’ve pointed out above the large corporations are frequently just “taking and searching” users data without their consent or even knowledge and in many cases you as a user can not stop them so it’s the rquivalent of theft. Which I think most people would classify as being covered by “unreasonable” if not “illegal”.

The European view is such taking let alone searching is an illegal act, which in part is also the US view when it comes to illegal searches.

As you note there is a difference between obtaining a warrant and obtaining a conviction, but when it comes to illegal searches or even entrapment, then in both cases the person giving afermation under oath is committing purjury and does not have the excuse of being “unknowing”. Part of that is like the chain of custody on physical evidence, they have to know what they are receiving from Microsoft “follows the evidentiary rules” or it is tainted. This brings back the “beyond reasonable doubt” burden of proof.

Thus the question of if what Microsoft are doing A, follows the evidentiary rules, and B, is free from all other taint.

I would argue not. Microsoft has chosen for whatever reason not to supply “false positive/negative” data. Not just on their hash values but also on if an image is illegal or not (something they can not do due to it requiring “the opinion of an officer” qualified to do so in most juresdictions many of which have varying legislation).

Further Microsofts searches are not “accidental” or “impartial” as the are doing it for gain to every one which makes it the equivalent of a “general warrant” search. When you add in that often this search is done on data that may well have been taken “without consent or knowledge” of the user and it most certainly was not in “plain sight” then the act of moving it potentialy creats a crime thus it is hardly taint free.

Thus the question arises “If an officer knows this” then why are they involving themselves with such crimes as they will have to go on oath… And if they do not know why are they qualified to do the job?…

The questions build up and with each one the process from the get go becomes more and more questionable…

Clive Robinson • August 23, 2018 12:08 AM

@ MarkH,

I’m prone to faintness from time to time — thankfully, not apparently related to any serious health matter.

My advice, keep a close and watchful eye on it. In my case I’ve had “unexplained” episodes of passing out for over a decade. It happened once in hospital but due to the actions of an agency nurse, vital information was lost so things remained “unexplained”. Which if the loss had not occured might have saved me the “busy doing nothing” on the shop floor.

As it was I got several invasive proceadures to insert “stents” around my heart. As with all things there are upsides and downsides. It’s therefore possible that the stents have actually “kept me alive” during what are now assumed to have been AF attacks that caused my blood preasure to fall thus me to pass out (apparently your brain does not like low blood preasure, so it responds by making you lie down the fast way by blacking you out).

At such low blood preasures there is a considerable danger that the heart even in tempory AF could go to VF and crash which is usually fatal if you are not in an area of a hospital setup to deal with it. Also the lack of oxygen to the brain also means the same for the heart, so if you have partialy blocked blood vessels around it, you are way more likely to suffer parts of the heart being irreparably damaged or it stop functioning followed shortly there after by death…

I don’t want to scare anyone but repeated unexplained events should be investigated as they might be early signs of something worse to come.

In the UK and I expect other places men are notorious for ignoring their health and not seeking help, it might account for why men die that much earlier than women on average.

Let’s just say I’ve had it fairly forcefully pointed out to me by those close to me that I should have been on the Doctors case more frequently. It’s not nice especially as they can now also say “I told you so”…

I’m also now facing the prospect of having a box of insecure electronics attached to my heart via electrodes. As I work from time to time with quite high power RF thus very high field strengths more than sufficient to block receivers and low voltage electronics it is not a prospect I am looking forward to. But worse as I point out on this blog from time to time, I know a number of implanted medical electronic devices are less secure than a tent with the flaps tied back…

Clive Robinson • August 23, 2018 1:28 AM

@ echo,

The journalists who know about the story told me the newspapers refuse to print the story because it involves the establishment.

Sadly as I’ve said on this blog before this is all to common.

The example I’ve had close contact with is the UK “Operation Ore” which a friend had their life destroyed by simply because a well known supermarket with attached petrol stations alowed credit card details to be stolen and sent to the Far East where criminals were using pornographic sites to raise income through a US gateway, that processed the stolen credit card details.

The US authorities sent to the UK a long list of peoples details that coresponded with the credit cards used through the gateway along with “falsified data” about the gateway (just another reason to not trust US authorities).

The UK authorities under a disreputable person who had been involved with “Ronny Flanagan” and some quite illegal goings on during the NI “troubles” went on a crusade against the non establishment figures on the list, some of whom committed suicide. As part of that they did not hand over evidence in a timely fashion, thus with the help of another friend who obtained via Freedom of Information Requests the full list of names and thus other evidence.

On the list were not just establishment names but more importantly names of people very close to establishment names that were not investigated.

One was the boyfriend of a very prominent politician against whom the evidence was overwhellming.

This name was publically read out by my friend infront of journalists from all the major UK MSM organisations on a number of occasions. Not once has any of those journalists or their organisations investigated the politician or their boyfriend, even though there is other evidence against them.

I would dearly love to name the politician concerned because they are clearly a stain on the body politic in many more ways than one. However as I’ve pointed out there is no right of free speech in the UK that has any meaning and the defences available under the Defamation Act are narrow in scope.

I can however suggest people look up the list of names the US supplied to the UK –they were put up on a website– and then cross check the names with the known boy friends of major political persons of the time.

My friends name was eventually cleared but the damage was by then long done…

But the person I would most like to see dealt with was the man incharge of Operation Ore, who was “rewarded” with yet another position of trust, and involvment with Microsoft. Who a decade later said,

“I’m proud of Operation Ore today. And I’ll be proud of it on the day I die.”

It’s not an opinion many in child protection services agree with, they point to the many harms and diversion of resources. Whilst others in compiter security and forensics point out the it has in no way diminished online pornography, but it has by alerting those involved with criminal acts how to go about protecting themselves and others so that they have in effect not just become stronger, but flourished.

It is the harms that Operation Ore caused that are rarely mentioned by the UK MSM and even where some of them are mentioned they tend to be down played. For instance you can see this in,

https://www.bbc.co.uk/news/magazine-20237564

Clive Robinson • August 23, 2018 2:55 AM

@ relevant?,

The interesting bit is,

Under Kyllo, however, even an extremely invasive technology can evade the warrant requirement if it is “in general public use.”

It has a weasle word statment in it that is problematic to say the least.

To see why ask yourself some questions about what “in general public use” realy means.

Cars for instance can be either privately or publically owned and operated for either private or public transport.

That is both the citizens and instututions/entities can,

1, own a car.
2, use a car for their private activities.
3, use a car for others private or public activities.

Most would agree incorrectly that a car was covered by “in general public use”. Why incorrectly? Well because there are people who are not alowed to operate a car for various reasons or in some cases to own one, therefor it is only a subset of the general public.

But ask the same questions about a train for instance, then citizens owning and operating is some what different, and you would find not so many people would agree that they are “in general public use”.

Now ask the same questions about the electricity meter that measures the power entering your property,

1, It is not owned by citizens.
2, It is not operated by citizens.
3, The data gathered is not available to citizens.

I could go on, but whilst power meters are certainly “ubiquitous”, they are most definately not “in general public use”. Because it is an institution/entity that owns it, installs it, holds all legal title over it, and owns the data produced by it. Citizens however have no choice but to be connected to a meter, they actually have no right of access even to their own readings as the bill is all they are entitled to receive. Nor do they have the right in many places to fit their own meter so they can contest a bill. As for getting access to other citizens meter readings, the best you can hope for is a court might grant it under very limited conditions.

Thus I suspect few would consider a utility meter of any kind to be “in general public use”.

Which brings us to surveillance equipment such as drones currently like cars both citizens and institutions/entities can own and operate them (though that is starting to change). Just because the can be owned and operated by anyone would you argue that they were “in general public use” if that ment they could either hover outside your bedroom window looking in or use a telephoto lense camera from a distance to look in your bedroom window?

Because what the court is saying in reality is “yes they can” with their interpretation of “in general public use”…

It’s playing around with such weasle words that alows the scope of legislation to be broadened way way beyond the drafters worst nightmares let alone intent.

Clive Robinson • August 23, 2018 5:20 AM

@ MarkH,

Probably you’re past the days when you needed to tweak up a powered transmitter by twisting a “gimmick” of a picoFarad or so …

Although it’s a small part of what keeps me from being idle, it’s R&D activity. Which means you “have to get your hands dirty” both on the bench and in the field.

Some are probably wondering what the tie-up is between RF R&D and Computer Security. Well I take a broader view of information security which is not limited to information processing, but all forms of information storage as well, and all forms of information communication.

You could say I have the view of,

Energy Energy everywhere but not a drop should link

I also used to design “security equipment” of all forms not just crypto type kit. The surveillance kit was the most fun and was quite often on the leading side of “the bleeding edge”, and still is for some things.

But when it comes to “information leakage” via side channels you realy need to know the fundemntals from a solid practical basis and you also realy need to know how things fail in part or fully. The only “proffessional design” side that used to require that was for “safety systems” especially “Intrinsic Safety” used in hazardous environments or life critical environments. So I’ve also worked in the design of Industrial Control and Automation and the design of Medical electronics.

Which kind of brings us back to High Power Transmitters, few realise just what is inside the various types of body imaging equipment. When you describe it in all it’s raw details tooth and claw people start to look uneasy.

But that is not as bad as the “Beauty Industry” there you realy are talking “load her up” “for maximum smoke”. Those machines that remove unwanted hair near permanently, are realy quite dinky looking. But… they use a Medium Frequency power oscillator to drive a considerable RF current into the hair follicle, which heats up rapidly then explodes under the vapour preasure of the cells contents boiling. Sounds nasty, but then consider that for a current to flow it needs some form of circuit, thus you should ask where the current goes on leaving the cell. Well after going through a lot more cells some of which die as well it ends up using any good conductor such as blood, lymphatic and nerve systems to get to the other electrode that could be the chair you are sitting in or a strap on some convenient point like your wrist… At least with medical equipment used for surgury or physiotherapy they use a patch about the size of a sheet of A4 connected to places that draw the current away from internal organs etc.

But don’t start me on the lasers, they realy are Dr Doom territory outside of medical labs and thearters. I get queasy when I hear about people “going on holiday” for laser eye surgery, tattoo removal or similar. Just remember the same basic laser can be used to cut out cancers in much the same way as a scalpel can but with less blood, the bigger ones have no trouble in amputation or reconstructive surgery.

But… if you Google “lasers in hip replacment surgery” you might get bombarded with lots and lots of adds and advertorial articles with lots of “Five things you should know…” all befor you start getting to “the meat of the subject”. Yes you will find lasers being used in hip and other joint replacment surgery, but dig a little deeper to find out what they are realy being used for 😉

One similar advertorial title that brings a wry smile is,

Which treatment is best for piles, laser or stapler?

Which could bring a new meaning to “pass the stapler” or “put it on the laser”… I guess it’s a slightly safer way than,

https://www.schneier.com/blog/archives/2009/09/ass_bomber.html#c393969

For some reason there is always a morbid curiosity with those performing DIY surgery… Possibly the most famous being Joe Mellen and wife Amanda Feilding both of whom drilled holes in their own heads to be creative (and neither appears to be crazy unlike some U-Tube types).

Clive Robinson • August 24, 2018 2:55 PM

Unsuprisingly Intel are being prats

Do you remember the outcry when a senior Oracle person basicaly said on her blog that only Oracle had legal rights over benchmarking etc? She fairly soon took it down.

Well it appears Intel is trying the same trick but worse much worse, the implications are you can not even benchmark your own code on Intel CPUs if you install the microcode patch that supposadly fixes the balls up they made over Meltdown, Spector and later variants.

Bruce Perens has blogged about it,

https://perens.com/2018/08/22/new-intel-microcode-license-restriction-is-not-acceptable/

Unlike Bruce Perens I am in no way forgiving Intel. As I said some time ago they chose to do things the way they did to get a speed advantage, thus it was quite deliberate. And don’t let any mealy mouthed spin Dr or Marketing / PR Droid try and convince you otherwise.

The fact the same issue occured on other CPUs is very much like those BWM rigged emissions tests, that suprise suprise also appeared in other manufacturers cars. People in the industry get their opponents products or design/manufacture data and go through them to see how they get better specs… Then in effect copy or come up with a slight variation for their own products.

This has been going on for a very long time so the Intel Xmas Gift will carry on giving for some time to come.

Oh and by the way Intel’s restriction is illegal in atleast three places I can think of, so it probably won’t be long before a Chinese or Russian web site publishes the information Intel are desperatly trying not just to not let out but avoid having class action level law suits over…

My guess based purely on this gaging order nonsense is that the performance hit will be closer to 15% than 5%, which means anything over 10%, and I’m thinking I am being a little conservative 😉

Clive Robinson • August 24, 2018 3:16 PM

Remote Execution on FACEBOOK server.

I’ve got to admit as I’m not a Facecrook user this did make me smile,

https://blog.scrt.ch/2018/08/24/remote-code-execution-on-a-facebook-server/

Though I doubt it will effect their share price, which should hopefully make Mr Zleeze-burger happy if of course he can be…