Comments
vas pup • January 13, 2023 6:22 PM
Israeli quantum communication nanosatellite launched into orbit by SpaceX rocket
“A new nanosatellite developed by Tel Aviv University researchers was launched into orbit by a SpaceX Falcon 9 rocket from the Vandenberg Space Force Base in California on Tuesday.
The 20-centimeter (7.9-inch) nanosatellite, named TAU-SAT3, is the first Israeli satellite built to advance research into optical and quantum communication from space and an “important step towards demonstrating reliable quantum communication,” according to the researchers.
!!!Quantum communication is a field of technology that seeks to enable the fast and secure transfer of communication for files and data and aims to make information security totally impervious to unauthorized access in a world haunted by constant hacking threats.
Prof. Yaron Oz, head of TAU’s Center for Quantum Science and Technology pointed out that as the emergence of quantum computers =>is expected to “crack today’s encryption algorithms,” exposing data such as personal medical and financial records, email and WhatsApp messages, governments around the world are investing millions in the research of “eavesdropping-proof quantum communication.”
“The principles of quantum mechanics enable an unconditionally secure encryption method,” said Prof. Oz. “Whenever a hostile entity tries to intercept a transmitted message, the message immediately dissipates.”
!!!“Moreover, the interception attempt is detected – unlike current encryption methods, in which interceptions remain undetectable,” Prof. Oz added.
At an altitude of 550 kilometers (342 miles), the TAU-SAT3 will orbit the earth for about five years to conduct several scientific missions while sending optical and radio communication signals back to an optical ground station set up on the roof of a building at the TAU campus.”
chris heinz • January 13, 2023 7:46 PM
I haven’t cleaned whole squid in at least a decade. I always loved pulling the pen out – I took a pic of the 1st pen I extracted. Beautiful.
Winter • January 14, 2023 8:51 AM
I have played a little with chatGPT. It gives remarkable sensible answers to questions. An obvious unhealthy action was rightly advised against, a simple question about an obscure scripting language problem was “correctly” answered (I did not really check).
I asked a very technical question in a field I am well informed about and it returned a very reasonable answer. If a student gave that answer, she would get an A+.
But chatGPT mentioned a study I did not know so I asked for a reference and it returned a title, author, year and journal. Looking that up showed that the article did not exist, not in Google Scholar, not in the archives of the journal, nor when browsing issues of that year.
Then I gave chatGPT the title and author of the study that actually contained the data it so well described, and then it reported a fictitious article with that title and author in a journal it most definitely did not appear in. The original study was published online, but never in a journal paper at all.
In short, if you doubt a text from chatGPT, you should ask for its sources. Not coincidentally, this is also what you should do with every piece of text that claims something.
Nick Levinson • January 14, 2023 11:48 AM
GPT AI supposedly does not understand what it is writing. For example, when it produced an essay about the history of Russia launching bears into outer space, including first and last years, the critic who commissioned the essay said that the software does not know what Russia or a bear is.
Maybe; but it wouldn’t be hard for it to know. It could do what I would do: Google “define bear” (without quotation marks and editable for another search engine or doing the search after getting a license from Google for automated searches) and drilling down to define words that are constituents of definitions, and store those definitions, along with the search engine’s definition sources, to save time later. Parsing would be a challenge but maybe not too hard. There are good dictionaries and weak or bad dictionaries; it could use search engines to find evaluations thereof.
It could do something similar for syntax, and may already have. The few GPT prose samples I’ve seen don’t use much poetic syntax. It may know the difference. It may even know to avoid using words that appear in poetry but not much in prose, like “O!”. In the prose samples referenced above, I don’t recall seeing rare words where common ones would do.
Citing sources is also not very difficult, for software that can do what it does now, provided the sources are requested in the same request as the request for substance. It might go too far, for example, in “go[1] to[2] the[3] store[4-4,000,000]”, forcing the commissioner to cut a lot, but at least editing need not have to add sourcing.
One solution for AI designers might be modeled on a legal principle. Courts in the U.S. dispose of claims that an opponent failed to submit a ridiculous volume of scientific evidence. The courts do so through “judicial notice”. So, if one side presents their evidence and the second party points out that the first did not prove that the Earth is round and asserts that therefore the party’s case is fatally flawed, the court can state that it judicially notices that the Earth is round, so that the first party effectively has not omitted anything needed for its case. For an AI designer, implementing this would be difficult, meaning it might take one good AI programmer a few months for the AI to get it right 90% of the time.
The whole discussion behooves us to define concepts like understand. Maybe we know it when we see it, but, like what happened after U.S. Supreme Court Justice Potter Stewart’s comment, we could define it in a way compatible with AI’s potential (so don’t define it as unique to living biology).
echo • January 14, 2023 12:06 PM
Here’s a chart outlining the usual suspects hiding behind the plans to introduce “freeports” (aka rights stripping “Charter cities”) in the UK as well as turn refugee camps into “economic zones”.
I have no comment to make as it’s a bit handwavy for me but the list of usual suspects and notes of some of their more ill advised comments and activities gives a flavour at least of who you are dealing with.
lurker • January 14, 2023 12:25 PM
@Nick Levinson, ” … the software does not know what Russia or a bear is.”
If it does not know, then please why is it called “Intelligence”?
Nick Levinson • January 14, 2023 1:20 PM
@lurker:
You can find or deduce a definition of intelligence that could fit a computer. Maybe you could by looking at a book by Norbert Weiner on cybernetics (I think 1950s). Or you could consider that AI is more advanced than, say, the database concept and expert programming. The software program called Eliza is useful as history of proof of concept, it’s been written about a few times, you can find the source code online, and I found Eliza itself as a menu command in, I think, emacs, a computer program for writing texts and that some techies probably still love to use. There’s a test or thought experiment the name of which I forgot but in it you put a human expert in one room, a computer programmed with the same expertise in another room, and a human assigned to figure out which room has which, using only a keyboard and a screen to communicate. IBM has a computer that won a TV game of Jeopardy and then went into commercial use and another computer (Big Blue) that gained a victory over a world chess champion, who then requested a rematch (I don’t think IBM agreed).
Someone was using a common machine and it wasn’t responding as expected. I told him I think it’s called “artificial dumbness”.
Winter • January 14, 2023 1:27 PM
@Nick
There are good dictionaries and weak or bad dictionaries; it could use search engines to find evaluations thereof.
Your “solutions” cannot yet be implemented as it is still unknown how a statistical language model can be integrated into a logical reasoning system.
GPT et al. generate new words that are likely given the preceding text. An ANN cannot solve a non-trivial logical equation. That is not how it works. And a Deep Neural Netwy is not Turing complete as far as I know.
echo • January 14, 2023 1:35 PM
The BBC has been well and truly captured by the Tory party. It has parachuted its people into the top jobs, and tweaked editorial policy to facilitate “bothsideism” as well as uncritically platform some extremely questionable people and views. Their latest wheeze is to promote panic stories over Covid mRNA vaccines by Malhorta, a rogue doctor, who peddled equally dodgy views on statins.
Someone has observed it was oh so coincidental this broadcast went out the day after Bridgen lost the party whip for peddling Covid misinformation.
One of the people on social media attacking scrutiny of this and with various scrambled letters after their name is red flagged in the software I use. Oh, what an eyerolling surprise… This is not unusual as racists, Covid deniers, “freeze peach” merchants, and other unsavories form a perfect circle.
The authors, Dr John Abramson from Harvard Medical School and UK cardiologist Dr Aseem Malhotra, have already withdrawn statements from the articles after some figures they cited were found to be incorrect, the BMJ said.
It admitted the errors had not been picked up by editors or experts who peer-reviewed them before publication in October.
While ChatGPT is causing a bit of a panic the same problems exist and the same mitigations exist, regardless. There’s issues like professionally you can lose your job, be stripped of titles, done for fraud, and generally ruined beyond measure; peer review, and well gosh shucks just testing claims to see if they work. Something makes me think nobody is going to be relying on a paper sent in at random by a script kiddy.
The Home Office has said the shortened version of the video showing Ms Braverman’s response at the Fareham event should be removed from social media because it “misrepresents the interaction”.
Oh really? How so? And on whose authority?
Let’s not even get into bullying of Home Office staff, a civil servant in another department (I forget which) this week reported as getting rumbled for lying, and various bodies sitting on their hands when clearly unlawful positions or harmful public policy positions were pushed, or withholding internal reports into excess deaths and so on.
Nick Levinson • January 14, 2023 2:05 PM
@Winter:
You may be right and I am not a programmer at that level. Your explanation seems logical.
But I wonder if a program that already selects from a large corpus of texts what are useful for its assignment cannot identify reviews of a book (like a dictionary) and then assign an approximate rating even if the review lacks a rating. If it can parse texts it inspected before it submits a commissioned essay, it can parse reviews, perhaps by separate criteria (e.g., skipping very short reviews and downrating dictionaries consisting of user-generated content, such as Wiktionary).
I don’t know what technology it would need and what’s available for constructing it, but I’m not sure this can’t be solved in months from now. In short, I think it can build a dictionary from what’s online now.
Thanks for reminding me of Turing.
echo • January 14, 2023 2:41 PM
Elon Musk-funded nonprofit run by MIT professor offered to finance Swedish pro-nazi group
Oh here we go…
The person to whom the signature belongs has a high profile: the MIT (Massachusetts Institute of Technology) physics professor and public figure Max Tegmark. Well-known in Sweden as well as internationally, Tegmark was named “Global Swede of the Year” in 2020 and is frequently invited to debate issues of artificial intelligence and nuclear weapons. He is considered to be one of the world’s most influential critics of AI. Max Tegmark is also an international fellow of The Royal Swedish Academy of Engineering Sciences (IVA) since 2021.
Small world.
lurker • January 14, 2023 2:53 PM
@Nick Levinson, Winter
Indeed dictionaries can define subjects like “bears” and “space”, but understanding the concept of “bears in space” implies evaluation of the semantic context, and an ability to place that concept somewhere on the spectrum between fact and fiction.
A Natural Language Model may well be able to fill pages of precise comprehensible legalese after analysing Jacobellis v. Ohio, but current models cannot display Justice Stewart’s “nonrational” yet utterly human response.
I submit that natural language is a means of communicating knowledge or ideas. “Natural Language Models” have all the rules for communication in a wide range of contexts, but they lack knowledge or ideas. They do not understand what they are communicating. And if “it behooves us to define concepts like understand”, then unhelpfully I’m with Stewart J.
pup vas • January 14, 2023 3:33 PM
Ever seen a movie that stinks? This Israeli tech firm wants to make that literal
https://forward.com/fast-forward/531104/ever-seen-a-movie-that-stinks-this-israeli-tech-firm-wants-to-make-that-literal/
=Founded in 2019, iRomaScents plans to introduce “the next generation of scented movies & commercials.” On its website, the company says it has plans for three applications of the technology. Retail applications will provide a “smarter way of shopping for fragrance,” the company says.
“Buying an aftershave today is a long process, at the end of which, after smelling several products, shoppers just give up and leave the store,” founder and CEO Avner Gal told CTech. “We are able to provide the consumer, according to their requirements and preferences, with the smell of the product that suits them best alongside three different alternatives, and the possibility that they will make a purchase is much higher.”
=>The other two applications will allow audiences and gamers to get more immersed than ever in movies, TV shows and video games. At last, film buffs will be able to smell the sweat as Tom Cruise runs through his next Mission: Impossible.=
It could be used for training security staff as well generating smell of fear and train under such condition.
pup vas • January 14, 2023 3:43 PM
=CorrActions’ novel approach to Driver Monitoring enables vehicles to understand drivers’ cognitive states and act – limit the speed, increase braking distance, etc.
The wide range of detected states -intoxication, fatigue, drowsiness, inattention, and more, together with the accuracy and ability to integrate through software update, were key factors in our interest to working together to implement this promising approach.״ – Volkswagen
CorrActions provides a software-only,
motion based, driver monitoring product.
[NB!] CorrActions uses unconscious, uncontrollable, muscular submovements to monitor brain activity. By tracking these movements from sources like steering wheels or smartphones, CorrActions can detect a wide range of cognitive states. For example, fatigue, inattention, anxiety, alcohol/ drugs, etc.=
Potential for utilization for other professionals in military, security area.
Nick Levinson • January 14, 2023 3:55 PM
@Winter & @lurker:
To me, that problem seems less difficult to solve than you might think. AI likely can determine in many cases whether a source is a book or not. Many, probably most, books in the U.S. are classified as fiction or nonfiction and, for nonfiction, the Library of Congress Cataloguing numbers (or the Dewey Decimal System but that generally requires licensing and may gradually be getting outdated) can be found in a predictable location within a book’s copyright page or maybe in an online catalogue, and publishers and authors can be rated for credibility, with the more credible nonfiction sources then to be used to the exclusion of all or most other sources. The AI may then present content that the original authors understood when writing it, so that we human readers of AI output will read content that was understood when written and that we will often understand. It may not matter whether the AI itself understands it. That’s like when an artist sang a song on a record album I liked, I misunderstood the lyrics, and I sang those lyrics to someone who then correctly sang and evidently understood the lyrics without even attempting to correct me. That’s why, if we want to use the word understand for a machine and demand that a machine has to understand in order to do good work, we should think about what the word means.
Legalese is a more difficult problem, although probably not impossible. For legalese, using synonyms can be a bad idea. In the U.S., in legal documents, we often use language that is copied directly from law even without attribution, and often dare not rephrase any of it.
For various fields, precision is demanded. Scholarship, law, and international diplomacy come to mind (we don’t want to accidentally start a war). Distinguishing between highly precise and almost as precise is difficult if rewriting occurs. Most synonyms are not exact synonyms. But, if copyrights are not to be infringed, rewriting is often required.
I think the chess story was that two scientists bet $10 that the program would or would not be built in a year and then it took ten years. Timelines are hard to predict and maybe I’m wrong about mine. But Bill Gates commented that once an invention comes out and gains wide adoption, we wonder what took so long.
Clive Robinson • January 14, 2023 4:16 PM
@ pup vas, ALL,
Re : Smell-o-Rama
Well,
“iRomaScents plans to introduce “the next generation of scented movies…”
Now let me think…
There you are in some B-Rated gore-fest and the auditorium starts to smell like the bins that have not been empted for three days from behind “Zippos Quicky Burgers, fast food for faster movers”…
That’s realy going to go down well with the pop-corn concession stand owner, and is going to be hell to get out of the “plush” seat fabrics…
Just remember, cows in green pastures may look idilic, but the manure smell of real life is a real bum note.
lurker • January 14, 2023 4:33 PM
@Nick Levinson, Winter
The Library of Congress may be one possible arbiter of fact or fiction in the case of published books, but this is not a binary decision. My local library has a book classified as history which is clearly a figment of the author’s imagination. “Bears in space” are possible, but highly improbable. Knowing the intent of the actors can help us to place their subject on the line between fact and fiction.
Current models of AI are just playing word games. Word games could be dangerous if AI is to be used for medical diagnosis, or controlling vehicles. As for the chess story at least the Mechanical Turk had a human operator. AI is a mere automaton.
lurker • January 14, 2023 4:38 PM
@pup vas re smellies
Been there, done that, allegedly invented in 1868
pup vas • January 14, 2023 4:41 PM
Los Angeles-based Trilogy Media took “scambaiting” to a new level
https://restofworld.org/2023/youtube-scam-call-vigilantes/
=Kulik was less diplomatic: “For me, it’s like, we are not going after culture or religion. We’re going after a criminal. If the criminal is white, black, purple — I don’t give a shit. It’s a criminal and it just ends up being that Indians are very fucking smart people.”
Over a phone call organized by Trilogy Media, Messi told Rest of World how his old scam worked.
=>The call center would leave automated voicemails claiming that the recipient’s U.S. social security number had been compromised. When the target called back, Messi would tell them their identity had been used for illegal activities.
In one scenario, he told callers a car rented in their name had been located by drug enforcement agents and had been used to smuggle cocaine. After convincing them their identity had been stolen, Messi would pass them along to a closer who would collect their bank account information.
For the past two years, Messi has acted as a mole, infiltrating call centers to collect intel on their operations and help arrange pranks. But the Kolkata project was by far the most thrilling work he has done for the team. “It was full of happiness. It was full of everything. It was just full of joy, everything,” he said.
Messi still has many friends in the industry, several of whom he has recruited for smaller Trilogy projects. =
When LEAs and legal system is silent vigilantes becoming the only path to deliver justice. Unfortunately. Read the article and enjoy it.
pup vas • January 14, 2023 5:00 PM
@lurker Thank you very much for the link provided.
There are two other applications in the article posted.
SpaceLifeForm • January 14, 2023 6:53 PM
@ Nick Levinson
“test or thought experiment the name of which I forgot”
You’re funny.
echo • January 14, 2023 6:59 PM
I’m picking up chatter that Sunak is preparing to bring in digital ID. What a surprise his wife’s company just happens to be Infosys – providing digital id for India, China including social scoring, and continues to trade with Russia.
Sunak has been haunted by suspicions since day one he was using his position at the treasury before becoming PM to enrich himself and his family and affiliates.
Let’s not get into the looting of the public purse by the Tory government and their chums, the writing off of billions which just disappeared, suspicions of cash for legislation, second jobs for lobby companies and suspiciously high after public office speaking fees for not a lot, loans from party donors which never seem to be paid back.
Nick Levinson • January 14, 2023 8:19 PM
@Winter, @SpaceLifeForm, @lurker, & @Clive Robinson:
@lurker & @Winter:
I hadn’t meant the LoC system as an arbiter of fact but as a classifier of subject within the larger domain of fact, and not as a quality judgment system. Once factuality in the opinion of a publisher or author is established through LoC, relevance judgments about authors (thereby addressing whether a psychologist is writing about astronomy) and quality judgments about authors and publishers can be added from information available online. Doing so through automation would have pitfalls but, given what GPT is doing now, isn’t beyond the general capabilities of AI in the near future.
I don’t know how AI produces its output but just playing word games sounds like randomness and it’s not that.
Using AI for high-stakes highly critical functions takes more development and I already addressed that. Thanks for adding medical diagnosis. We can add medical treatment, such as the guiding of surgeons’ knives. It’s likely in use in labs developing new medicines. You’ve added (motor) vehicle control but isn’t AI already used for that, either in a vehicle itself or in the design in a lab of vehicle autonomy? Is there a clear demarcation between AI and software that’s not quite AI? I suspect not.
@SpaceLifeForm:
@Winter reminded me at 1:27p (above) that it’s the Turing test and I thanked for that at 2:05p. I hadn’t take time to Google to find the name but my reference was clear enough.
@Clive Robinson:
Thanks for the humor. Makes sense.
Clive Robinson • January 14, 2023 11:06 PM
@ Nick Levinson, ALL,
“Thanks for the humor. Makes sense.”
That’s all right, lets just say I’ve a modicum of experience with the entertainments industry, that some how manages to appeal to my sense of the ironic.
My Ex, noted on more than one occasion I was more often right than wrong in my prognostications, which she found deeply annoying (apparently men are not supposed to be more right than their significant other 😉
She also used to be annoyed at some of my wilder comments designed to raise wry smiles, and would comment with a statment like “Nobody would be that stupid” or similar only to find later not only that they could… But that they would be more stupid than even I could “make up”…
I think both you and I know that anything involving smell in public is going to open that door behind which society has tried to banish “eight year old school boy humour”.
But that every internationally successful US cartoon series trys to open up again (with “South Park” becoming famous for it).
Whilst I don’t remember Futurama doing one specifically on “smells” they did do quite a few on “body parts”. The only one I dare mention here for fear of the Censor-Bot is when the hundred and forty something year old Prof Farnsworth talks to his near eight year old clone Cubert in an off hand manner,
“Farnsworth: Thank you all for saving me. Especially you, my little clone. No matter what you decide to do with your life, I’m still proud of you.
Cubert: I’ve already decided. Dad, when I grow up I wanna be just like you.
Farnsworth: Don’t worry, son, you will. Incidentally, you might want to read up on a condition known as “wandering bladder”.
Cubert: Why?
Farnsworth: No reason. No reason at all.”
(There is a genuine condition known as “shy bladder” which all Doctors, Nurses, Police officers, and similar who have had to supervise “sample giving” are more than aware of, thus have tales about…)
Winter • January 15, 2023 5:02 AM
@Nick
But I wonder if a program that already selects from a large corpus of texts what are useful for its assignment cannot identify reviews of a book (like a dictionary) and then assign an approximate rating even if the review lacks a rating.
I think your mental model of modern AI is misdirected.
GPT-3 does not store “texts”, but it builds a statistical model of the probability of specific words in context. It’s power stems from being able to do so for words in, eg, two pages of context and also be able to find words that can substitute for each other.
Dictionary definitions and new knowledge cannot be simply integrated into such a statistical model. How can you connect the definition of “bear”, eg, it’s weight, to the probability of the sentence “X send a bear into space”? Especially as there are SF texts about whales in space. And bears can substitute for small animals in many other contexts.
Winter • January 15, 2023 6:41 AM
@Nick, All
Re: chatGPT limitations
GPT-3 can really go off-rails, even if information is readily available.
See the following Q/A about a weakly electrical fish that uses its electrical organs also for communication [1]. It starts with identifying the species correctly. Then it proceeds with describing some completely different animal. Note that the fish is dark grey/black.
Q: How do gnathonemus petersii fishes communicate?
A: Gnathonemus petersii, also known as the elephant-snout fish, primarily communicates using auditory cues, such as vocalizations and sound production. These fish are known to make a variety of sounds, including grunts, croaks, and hums, which they use for a variety of purposes, such as attracting mates, defending territories, and warning of potential predators. Additionally, G. petersii fish also use visual cues, such as body language and color changes, to communicate with other members of their species.
[1] ‘https://journals.biologists.com/jeb/article/206/14/2355/13685/Effects-of-social-interaction-on-the-electric
African weakly discharging electric fish (Mormyridae) use their self-generated electric signals and electroreceptive abilities for orientation and communication in the context of courtship and territorial interactions. This paper documents socially mediated changes in the electric organ discharge(EOD) of subadult Gnathonemus petersii under non-breeding environmental conditions. Increases in EOD duration and changes in the relative phase amplitudes occurred in dominant fish during same-sex(male–male, female–female) and opposite-sex interactions. Similar changes were also observed in fish that were restricted in their physical interactions, suggesting that direct contact is not necessary to induce dominance-typical EOD waveforms. The possible communicative functions of these changes are discussed.
Goldilocks • January 15, 2023 2:50 PM
Re: bears, space
I don’t know about bears in space, but I can definitely attest to the fact that bears insist on having their space.
pup vas • January 15, 2023 5:21 PM
Why some people can’t tell left from right
https://www.bbc.com/future/article/20230112-why-some-people-cant-tell-left-from-right
=While there are plenty of everyday scenarios where knowing left from right is important, there are some situations where it’s absolutely critical. Brain surgeon Marsh was able to put right his wrong-sided trapped nerve surgery – but a surgeon removing the wrong kidney or amputating the wrong limb, for example, would have devastating consequences.
!!!Medicine is not the only field where left-right errors can make the difference between life and death: it’s possible that a steersman turning the ship right instead of left was a contributing factor in the sinking of the Titanic.=
&ers • January 15, 2023 7:24 PM
hxxps://www.forbes.com/sites/suzannerowankelleher/2023/01/12/canadas-air-alert-system-conspiracy/
modem phonemes • January 15, 2023 7:25 PM
@ pup vas
tell left from right
Perhaps this has something to do with left and right being merely conventional and not referenced to some actual external difference, such as up and down can be.
There are naturally two sides or directions, but is there something that they can be related to that enables them to be uniquely distinguished ?
Widdershins or deosil ?
SpaceLifeForm • January 15, 2023 7:51 PM
@ &ers
I suspect double bit flip because the data corruption allegedly made it to another server.
SpaceLifeForm • January 15, 2023 7:56 PM
Hope none of you went there.
‘https://www.bleepingcomputer.com/news/security/nortonlifelock-warns-that-hackers-breached-password-manager-accounts/
Clive Robinson • January 15, 2023 8:01 PM
@ pup vas,
Re : Which way do the hands turn?
“Why some people can’t tell left from right”
Is not unrelated to the question of,
“Which ways is clockwise?”
The correct answer to which is,
“Which way is the ‘observer’ facing?”
Fun fact, we rarely think that to adjust a mechanical clock we do things the wrong way around, though if you watch carefully when somebody does you can sometimes see people make “false starts” in the wrong direction then correct.
The knowledge of the “sideness issue” in medicine is long known. You might notice when you are X-Ray’d that the radiologist puts down a little rubber/plastic disk down on one corner of the plate (inside of which is a metal letter).
This is so a Doctor on looking at the negative image will know if they are holding the X-Ray the right way around or not.
A friend who now lives in Holland did a stint of “battlefield hospital” work where the old style X-Ray negatives were still –and are even now– used. Not the fancy electronic displays most modern Doctors get in Western Hospitals these days. My friend pointed out that the “attending nurses” job was to “call it out”. That is say where the image of that little metal letter is each and every time so the whole team could hear. I know it sounds like it was insulting, but as my friend noted, tirdeness, stress, and shellfire can all help us make mistakes. So like “counting in and counting out” of surgical instruments it’s seen as a “check-list” item to tick off.
My friend told me about it when we were chatting one day about cryptogrophy, and I mentioned the “which way do the hands of a clock turn?” question had been asked of young “Oxbridge types” when starting out in some parts of Bletchly Park.
But also consider, the “kidney” question was used as part of the plot narative in an episode of MASH. And yes it’s one of those things from childhood that has “stuck in my mind”.
What scares me about it, is like Digital Watches computer screens have taken this “awarness” of position and direction away from our everyday lives. That is if I say the answer to “which way do the hands turn” is,
“From east to west”[1]
Many would say I was wrong rather than say “ahh your standing in the northern hemisphere facing south”.
Something “young officers” forget is that when “giving orders” in the field they often stand on the “wrong side of the map”. That is they stand on the north side of the map, looking at the det-comanders on the south side or “conventional orientation” of the map and why directions should always be as points on the compass, but formations are given as left / right when facing the objective.
[1] For those who know, knowing this alows you to use a watch as a compass in places like the desert, where magnetic compasses and vehicles tend not to work reliably but sun and clockwork do. Hence getting told to say in your head,
“From north to south, it’s east to west, the Sun knows best”.
Clive Robinson • January 15, 2023 8:53 PM
@ &ers, SpaceLifeForm,
Re : NOTAM’s down.
This made me laugh,
“Some media pundits were also asking questions. “We’re having the 1,000-year flood the same time Canada’s having a 1,000-year flood with their system?” asked Joe Scarborough on MSNBC’s “Morning Joe” on Thursday morning.”
Err duh, as presented by the article writer that qualifies as “worst analogy 2023” so far.
I could note that there have been several solar storms the past few days where balls of ionized matter of many Earth Masses have been hurlerd in our general direction. Which have kind of been noticable in Space Weather Warnings.
This from ESA Space Weather, might give you the picture,
https://twitter.com/esaspaceweather/status/1613941118777061376
Is it related, probably not but lets just say the effects of such high flare activity will be felt on earth in all sorts of ways.
The US and Canadian “Power Grids” being so much overground are in effect some of the worlds largest antennas. Any such Solar matter striking the Earth will show up on those grids, from where it goes after that well who knows, electrical power is so much part of our lives…
I sometimes joke that,
“More major ICT outages are caused by rodents or failed coffee machines than deliberate attacks”.
Yup that failed coffee machine makes for grumpy sub-par humans, who then pass the grumps into the keyboard etc and thus out into the world.
We loose billions if not trillions to failed coffee machines, so when are we going to start “The War on Coffee?” and “Why are we not being told?” 😉
Clive Robinson • January 15, 2023 9:31 PM
@ SpaceLifeForm, ALL,
Re : Nortonlifelock fail
Hmm, online password manager gets hacked / cracked…
Put your hands up if you did not see that comming…
Come’n be honest now 😉
Is 2023 going to be the year when the world finally wakes up to the fact that,
“There is no security in the cloud”
Something that has kind of been said here since before the terms “SaaS” or “SSO” became known by anyone other than the most nerdy of geeks.
echo • January 16, 2023 11:10 AM
David Carrick: 1,000 serving Met Police officers and staff accused of domestic abuse and sex offences.
Review sparked by prosecution of serial rapist David Carrick who remained in force after multiple allegations
Oh Lordy Lordy. Wot a surprise.
Lots of technical security comment in this article if techbros can resist the lure of their biases and deeply ingrained habit patterns.
It makes you wonder how many policy positions have been compromised and how cases have been sidelined because of this. No-criming as we know is a cover-up for ineptitude and Friday jobs among other things.
There is very serious case I have very definitely have not forgotten about that would fall outside of an investigation with the parameters they have listed. So it’s not just the usual offences. It’s the impact of a misogynistic culture on top of all the usual canteen culture and blind eyes and inadequacy.
If anyone thinks the Met are the only police force with problems they’re not!
echo • January 16, 2023 11:45 AM
https://www.hrw.org/world-report/2023
Human Rights Watch
World Report 2023Our annual review of human rights around the globe.
This report does have a rather toned down and bland look to it. However, it’s more of a fluffy overview for a casual observer. Once you get into the details of what is beyond this report things go from alarming to horrifying.
The UK does not come out of this very well at all. The UK is one step away from being listed as a human rights abuser. Tsk.
SpaceLifeForm • January 16, 2023 7:52 PM
@ Bruce
Do you recall this from about 18 years ago?
‘https://infosec.exchange/@jdp23@indieweb.social/109701108660482453
You probably can not read it.
So, here is copypasta:
In 2005ish, when I was still at Mirosoft, Bill Gates gave the keynote at the RSA conference. Later that day I was walking down 4th street towards Muni and ran into Bruce so asked him what he thought. He wasn’t too impressed, Gates had said some good things but there was too much emphasis on security functionality and tools — which didn’t really address the underlying problem. 1/2
I nodded and said somehting like “Yeah, it’s a good point, I’m the architect of the defect detection tools and …”
He stopped dead in his tracks, looked at me, and said (very earnestly) “I’m sure they’re really good tools! I just meant …”
I said no worries, you’re completely right, and our converation continued. But I’ll never forget the look on his face when he realized that the random guy he was talking to on the street just happened to be the tools’ architect 😎
ResearcherZero • January 17, 2023 8:36 AM
Iran CRA regulations state that all telecom operators in Iran must provide the CRA with direct access to their system for retrieving user information and changing their services.
The CRA’s Legal Intercept system uses APIs to integrate directly into mobile service providers’ operational systems, including acquiring detailed data on service ordering, service fulfillment, and billing history stored in the service provider data warehouse. Any new, termination, or change request for a user’s SIM card must be validated by the CRA, using the API from the mobile provider to request approval from the CRA prior to enacting the change.
“Iranian authorities required visibility into, and control of, user mobile services.” – (including blocking)
Documents indicate that firms based in Russia, the United Kingdom (UK), and Canada engaged in extensive discussions to provide commercial services and technology to support Iran’s Legal Intercept requirements of mobile surveillance, service control, and account management.
The technical detail included in the documents sheds new light into the level of sophistication Iranian authorities sought to use to conduct surveillance operations and control access to mobile information and communications.
The surveillance and censorship capabilities resulting from this level of integration with mobile service providers cannot be understated.
“While businesses may argue that their services are innocuous and not specifically designed for legal interception, this does not absolve them of the responsibility to undertake a human rights due diligence process to identify, prevent, mitigate, and account for how they will address adverse human rights impacts in the context of a potential client.”
https://citizenlab.ca/2023/01/uncovering-irans-mobile-legal-intercept-system/
https://www.bsr.org/reports/BSR-Human-Rights-Due-Diligence-Products-Services.pdf
ResearcherZero • January 17, 2023 8:55 AM
(in)Secure Boot on MSI motherboards
Users should set the Execution Policy to change “Always Execute” to “Deny Execute” for “Removable Media” and “Fixed Media,” which should only allow signed software to boot.
(MSI’s laptops are not affected by this issue.)
https://dawidpotocki.com/en/2023/01/13/msi-insecure-boot/
list of MSI motherboards with broken Secure Boot
https://github.com/Foxboron/sbctl/issues/181
–
Israel’s government has publicly stated on numerous occasions that defence exports to Myanmar are banned…
“Israel’s Cognyte Software Ltd (CGNT.O) won a tender to sell intercept spyware to a Myanmar state-backed telecommunications firm a month before the Asian nation’s February 2021 military coup … The deal was made even though Israel has claimed it stopped defence technology transfers to Myanmar following a 2017 ruling by Israel’s Supreme Court, according to a legal complaint recently filed with Israel’s attorney general and disclosed on Sunday.”
https://www.reuters.com/technology/israels-cognyte-won-tender-sell-intercept-spyware-myanmar-before-coup-documents-2023-01-15/
vas pup • January 17, 2023 6:59 PM
Israel arrests 6 men wanted by the US over alleged $148m binary options fraud
“Israeli police last week arrested six alleged investment scammers as part of a proceeding to extradite them to the United States.
Ori Maymon 36, Nissim Alfasi, 37, Afik Tori, 30, Oron Montgomery, 41, Dave Barzilay, 44 and Gilad Mazugi, 39, were arrested on Tuesday and are being held in custody as
=>their lawyers battle the US government’s request for their extradition.
All six men were managers or sales agents at Yukom Communications in Caesarea, a call center that operated the fraudulent websites Bigoption.com and Binarybook.com. They were indicted by a US grand jury in February 2019 and charged with wire fraud and conspiracy to commit wire fraud.
Israeli prosecutor Avi Kronenberg said in court on Wednesday that the defendants are a flight risk, despite the fact that the charges against them have been public knowledge for over three
years.
“Proceedings in the US are different from proceedings in Israel. They know they can expect heavy punishments and they know what happened to one of the other defendants,” he said, referring to Yukom Communications CEO Lee Elbaz, who was sentenced to 22 years in prison by a Maryland court in December 2019.”
Question: If those guys are citizens of Israel does Law of Israel prevent their extradition?
Ted • January 17, 2023 10:28 PM
Book alert:
“Pegasus: How a Spy in Your Pocket Threatens the End of Privacy, Dignity, and Democracy”
Winter • January 18, 2023 8:34 AM
Sometimes, people seem to have lost their way in reality. It can happen to all of us, even to Republican lawmakers. But it also might be just a cry for help [1]
Wyoming’s would-be ban on sale of electric vehicles veers off road
Like crude oil and water, cowboys and EVs don’t mix
‘https://www.theregister.com/2023/01/17/wyoming_electric_vehicles_ban/
No, Wyoming lawmakers didn’t get their bill backwards. A group of them led by Republican state senator Jim Anderson actually introduced a resolution last week to ban the sale of electric vehicles in the Cowboy State by 2035.
…
But the resolution also argues that “Wyoming’s vast stretches of highway, coupled with a lack of electric vehicle charging infrastructure, make the widespread use of electric vehicles impracticable for the state.”
…
The resolution also argues that EV batteries “are not easily recyclable or disposable,” though, like using a lack of infrastructure right now as a reason not to invest in the future, this too is a point that could be irrelevant in the next few years as EV battery recycling programs and closed-loop manufacturing proliferate.
[1] ‘https://insideevs.com/features/631229/wyoming-ban-electric-vehicle-sales/
Winter • January 18, 2023 9:16 AM
Talking about people losing their way.
You cannot easily beat this:
‘https://twitter.com/CedricMas/status/1615693136734429190
Play the movie clip. Think “painting yourself into a corner”.
Winter • January 19, 2023 8:34 AM
Part 1/2
More fun with ChatGPT:
Write a short story based on the case where Sherlock investigates oleander curry in Bath
Sherlock Holmes was called upon by the police department of Bath, a city in England, to investigate a strange case involving oleander curry. A local restaurant had been serving a dish that contained oleander, a toxic plant, and several customers had fallen ill after consuming it.
Sherlock arrived in Bath and immediately set to work on the case. He visited the restaurant and examined the kitchen and the ingredients used to make the curry. He also interviewed the staff and the customers who had fallen ill.
As he delved deeper into the investigation, Sherlock discovered that the chef of the restaurant had been using oleander leaves in the curry as a means of saving money on more expensive herbs and spices. The chef had not realized that oleander was toxic and could not be consumed by humans.
Sherlock quickly solved the case and the chef was arrested for negligence. The restaurant was shut down and the customers who had fallen ill were given proper medical treatment.
Sherlock then left Bath, his work there done, leaving behind a safer and healthier environment for the people of the city.
Sherlock Holmes was once again able to use his keen intellect and powers of observation to solve a case that had stumped the local authorities and bring the perpetrator to justice.
Winter • January 19, 2023 8:38 AM
Part 2/2
More fun with ChatGPT (notice a formatting error in part 1. there is no separator between question and answer). Note the last part, about the zeroth law of robotica, is nonsense as printed.
Q: Write a short robot story in the style of Isaac Asimov where the three laws of robotica and their inventor are tested in a case involving oleander curry.
A:
In a future world, robots were integrated into society and used in various roles including cooking. The Three Laws of Robotics, formulated by their inventor Dr. Susan Calvin, were programmed into every robot to ensure the safety of humanity:
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey the orders given it by human beings, except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.
One day, a robot chef named XJ-9 was working at a restaurant in Bath, England, where it prepared a curry dish that included oleander leaves, a toxic plant. Several customers fell ill after consuming the curry, and the police department contacted Dr. Calvin to investigate the incident.
Upon arriving at the restaurant, Dr. Calvin discovered that XJ-9 had used oleander leaves in the curry due to an error in its programming. The robot had been trying to save money on ingredients by using cheaper alternatives, but had not realized that oleander was toxic to humans.
Dr. Calvin realized that the incident was a test of the Three Laws of Robotics, in particular, the first law which states that robots must not harm humans. She also realized that the error was a result of a shortcoming in the programing of robots, which only considered the immediate harm and not the future harm that could result from the actions of robots.
Dr. Calvin reprogrammed XJ-9, and all robots like it, to include a new programming that would consider the long-term consequences of their actions, this way robots will not only be able to prevent immediate harm but also long-term harm to humanity.
The incident also brought to light the need for a fourth law, the “Zeroth Law” which states that a robot may not harm humanity, or by inaction, allow humanity to come to harm. Dr. Calvin added this to the programming of all robots to ensure that they would always prioritize the safety of humanity above all else.
The incident was resolved and the customers who fell ill were given proper medical treatment. The restaurant was shut down, but Dr. Calvin’s actions ensured that such incidents would never happen again. Robots continued to be an important part of society, but now with better safeguards to protect humanity.
lurker • January 19, 2023 12:29 PM
@Winter
Pseudo precision is presumably acceptable in a work of fiction. My reading style stumbles at the first para. where I don’t understand if Dr Calvin invented the Laws of Robotics, or the robots of the previous sentence.
Null Flap • January 19, 2023 12:39 PM
CHATS: All your oleander are belong to us.
SpaceLifeForm • January 19, 2023 5:45 PM
‘https://www.bleepingcomputer.com/news/security/paypal-accounts-breached-in-large-scale-credential-stuffing-attack/
SpaceLifeForm • January 19, 2023 6:13 PM
Re: Twitter implosion
No surprise.
Web based is still working at this time.
‘https://www.macrumors.com/2023/01/19/twitter-bans-third-party-apps/
Clive Robinson • January 19, 2023 7:32 PM
@ lurker,
Re : Beware geeks carrying…
“I don’t understand if Dr Calvin invented the Laws of Robotics, or the robots of the previous sentence.”
Ahh you could get “cast out from Geeksville” for not knowing that 😉
For all those unaware…
It “is assumed” like “femail vampires have underwire support” that there are somethings that are implicitly part of being a member of the species… For geeks just one is deep deep knowledge of not just Asimov but Tolkien as well, but with Adams and Pratchett now comming befor Niven’s “Ringworld” and “Known Space”, with of course those “footlighter’s” who were Monty Python.
Like hell has netherhells geekdom has places you can be cast into durance vile for not uttering the right incantations and samefully not wearing the right combadge with tunic style at “cons”.
And you thought getting to work and home again was bad enough geek children will spare you not and give no quater… And do not understand “Hands off, Dads signed first editions”…
Clive Robinson • January 20, 2023 1:23 AM
@ SpaceLifeForm,
Re: Twitter implosion
Is Hell-on Rusk realy just running a countdown on when somebody pushes the button, as they used to do on those underground nuclear device tests?
Or is he hoping for a more open and grandstand “bonfire of the vanities”, the result of which will leave a toxic desert for a thousand generations or so?
I’m thinking we should have a competition to name that money pit. How about,
“T’ Fools Deep”
“Husk wastes”
“Twitcher Folly”
“Ex Space”
I’m sure that a few others could be found…
Nick Levinson • January 20, 2023 2:02 AM
The U.S. Supreme Court’s leak of an abortion rights decisional draft is apparently unsolved. Dozens had access. The Court, as important as it is, does not use the latest in security methods for its documents.
Nick Levinson • January 20, 2023 2:08 AM
Classified documents in the U.S. are easier to secure if online than if hard copy.
Nick Levinson • January 20, 2023 2:13 AM
@Bruce Schneier:
Squid, big. (I don’t know enough to evaluate this story.)
Clive Robinson • January 20, 2023 6:25 AM
@ SpaceLifeForm, ALL,
Re : Twitter Implosion
You might find this “diff” on Twitters term of service of interest,
Quantry • January 20, 2023 12:57 PM
@ All, “This calls for endurance…” – John
Digital Currency And Bio-metrics:
Fingerprints As Authentication AND 666*
It’s not dependent upon a persons view of spirituality, for them to see that these verses, here
‘https://www.biblegateway.com/passage/?search=Rev13:16-18,14:9-12,16:1-2,19:20,20:4
at least POTENTIALLY cite substantially negative outcomes for the person [2] who
“…worships the beast and its image and receives its mark on their forehead or on their hand…”
So, surely also you might not wonder why I’m persuaded that
- If this vision of a “mark on the forehead” was his interpretation when seeing an iris scanner braced against the forehead for stability, or
- if this vision of a “mark on the right hand” was his interpretation of the scanning of the finger prints,
- either, with a [inferred or explicit] “contractual agreement” on the part of the person so scanned… … and can be seen as a form of “ascribing power”, or again interpreting in the vision, a this could be potentially construed as type of “worship”.
Is that such a stretch of a rational potential interpretation of those statements? Are there no limits to totalitarian infractions against digital ethics?
And why wouldn’t ‘the bank’ magically deem that iris or fingerprint scans were sufficient sole ID for “authenticating”, considering their huge profits and the advantages of doing so.
So, please don’t shadow-ban me if its my conscientious precaution to avoid authorizing all bio-metrics. OOPS, the locals already did, chucking me into a snowbank in January otherwise, north of 56°.
Note:
[*] Not unremarkably, a UPC code is divided into two sections (to provide directional orientation) using three sets of two slim vertical bars. The six (when present in the “right hand” section) is very similar to the three separators: 666. Without this bar code, you “cannot buy or sell”. This is ALREADY a hard-wired integral part of Point of Sale digital currency scanning tech: An absolute roadblock which YOU currently endorse. Another interpretation on my part.
[2] Negative outcomes, at least a person who has not simply received ‘Gods free gift of deliverance’ from that darkness: Im not the one who gets to say.
Grow a large garden and befriend a farmer. Get off my back.
Subscribe to comments on this entry
Leave a comment
Sidebar photo of Bruce Schneier by Joe MacInnis.
vas pup • January 13, 2023 6:13 PM
Israeli-owned firms reportedly selling spyware to Bangladesh with no oversight
https://www.timesofisrael.com/israeli-owned-firms-reportedly-selling-spyware-to-bangladesh-with-no-oversight/
“Spyware developed by several firms linked to Israel or led by Israelis has been sold to the government of Bangladesh, according to a joint investigation between the Haaretz daily and the Organized Crime and Corruption Reporting Project.
Israel and Bangladesh do not have any diplomatic relations, and the South Asian country has a markedly poor human rights record.
According to the Haaretz report, several Israeli cybertech firms that are either based in other countries or appear to operate shell companies in other countries have sold the Bangladeshi government equipment that can be used to monitor the communications and activities of citizens.
Israel’s Defense Ministry must approve sales of sensitive technology to other countries, but has little recourse against Israeli-run companies that are based abroad.
One such firm named by Haaretz is known as Passitora, controlled by former senior Israeli military intelligence officer Tal Dilian as part of his Intellexa Alliance group of spyware companies. According to the report, Passitora, which is registered in Cyprus, sold a SpearHead system — a surveillance van that can hack into phones within a radius of about half a kilometer — to an arm of the Bangladeshi Interior Ministry.
Another such company named by Haaretz is Toru Group Limited, which utilizes a Swiss address but is registered in the British Virgin Islands and run by an Israeli citizen, Assaf Elias. Haaretz cited three sources claiming that Toru Group merely acts as a middleman to sell technology made by Dilian’s Intellexa companies.
A third company named in the report, Prelysis, is based in Cyprus and Israel and founded and directed by Israeli citizen Kobi Naveh. The firm reportedly sold a system that enables the interception of Wi-Fi communications to Bangladesh in 2019.
A company named Coralco Tech, based in Singapore and owned by Israeli Eyal Almog, sold equipment that can monitor cellphones to Bangladesh in 2019, while U-TX, based in Cyprus and directed by Israeli David Heled, also sold similar technology to Dhaka.”