Posts by Norman Nescio

Cisco: Don't use 'blind spot' – and do use 'feed two birds with one scone'

Re: Scones

...it's pronounced Stooon

Rhymes with 'broon', as in 'a pint of broon'.

Re: "hanging processes"

Hang on...I think there might be another meaning of hang that indicates something is interrupted and you are unable to predict when it will restart - the phrase hang fire comes to mind. Nothing to do with hanging things up.

NN

Owner of 'magic spreadsheet' tried to stay in the Lotus position until forced to Excel

Re: Palm Desktop

You are not alone. There are plenty of people still using Psion Organisers of one type or another.

I don't think it is possible to get Compact Flash cards with small-enough capacity to work in the Psion Series 5 now - anything larger than 128 Megabytes is not guaranteed to work.

Re: Spreadsheets for Dummies /s

One thing Excel doesn't do is source code control/change control/version control. There are add-ons and external tools, but the typical end-user building a big Excel workbook doesn't use them, so blithely modifies individual cells, copies incomplete ranges, makes mistakes with formulas, and specifies incomplete ranges in functions. Referencing other spreadsheets and external data in a controlled manner is also harder than it should be. Auditing big workbooks is 'interestingly' difficult. And don't get me started on using a common Excel workbook amongst users in different countries.

Excel is just not bad enough for people to throw it away and use tools more suited to the tasks at hand.

Re: NHS

Not to defend sharp practice, but there could be a reason for that behaviour.

Medical device certification (and pure software can be a 'medical device') to IEC 62304 is rather exacting and expensive, and will specify the exact versions of software that can be used - because they are the ones that went through the long and expensive testing process. Changes are avoided so that recertification is not required. There are various levels of certification, and if medical decisions are going to be made on the strength of data produced by an app(lication), then the certification requirements are rather stringent. (Medical decisions being: if you get it wrong the patient is harmed or dies)

I worked on a project once where the idea was to use the patients' fitness monitor records of pulse rate to provide data for the management of the treatment. One of the rocks on which the project foundered was the fact that almost all such monitors are not certified as 'medical devices', so you had no audit trail to certify you were likely getting accurate data. And while those monitors were actual physical devices, the same rules apply to the software used to process the data.

If the application being used was to support medical decisions, it probably had testing and certification requirements that made updates a painful, long, and expensive process.

On the other hand, it could just have been sharp practice.

Perforce: What is IEC 62304?

Top Google boffin Hinton quits, warns of AI danger, partly regrets life's work

Re: Non deterministic behaviour

I think 'TheInstigator' hasn't read The Story of Mel^*, or heard of the switch labelled with 'More Magic'.

*Annotations that might help, and the Wikipedia entry.

Instead of worrying unduly about what people with agendas will produce with AI, we need to be teaching people to question what they read / see / hear online, or at the very least making it clear that, just because a website / forum post claims something is true doesn't mean that it IS true. Get that right and we are well on the way to dealing with fake news, regardless of how it is produced.

I have some young relatives that have been introduced to the idea of critical thinking at their educational establishments.

I hope that it works a bit like plants: you plant a seed, and eventually, after several years, it bears fruit; because there are certainly not instantaneous results. That said, I'm impressed that the effort is being made. If I were overly cynical, I'd expect politicians to remove it from the curriculum, because it makes their lives harder.

Certainly, YouTube and TikTok videos (and others of that ilk) have a huge influence, and there is very uncritical acceptance of their content as true, which is somewhat worrying, as well as the second-order effects of the culture you get steeped in. I take heart that most children grow up to be reasonably responsible adults, but the process is pretty terrifying,

Re: Turing; and Clarke

You clearly assume that future AI will be human-like. Please see my earlier post (apologies, it's late and I'm lazy) for why I believe that is not necessarily a good assumption.

If it's that clear, then I have expressed myself badly, for which I apologise.

As an avid Science Fiction reader, I'm very well aware of the idea of non-human intelligence. Perhaps one of the earlier manifestations is Fred Hoyle's 1957 story 'The Black Cloud', which I read when I was a teenager (note, I wasn't a teenager in 1957), and of course, all of Asimov's robot stories with their 'positronic brains'. There's far more. Star Trek's Borg, perhaps? Skynet? Intelligent aliens, robots, collectives, creatures of 'pure energy' and so on..., and a non-science-fiction: Hofstadter's Aunt Hillary?

I do fall into the conceit of assuming that humans are intelligent. I also have the hope that a non-human intelligence could be 'better' than ours and mentor us to improve our ways, so long as it doesn't regard us in the same way we regard ants: interesting, but disposable.

So yes, if as a reader you feel you can draw the conclusion that I assume intelligence must be human-like, then I apologise.

Re: Turing; and Clarke

There have been chatbots winning formal Imitation Game contests for years.

Citations, please. Note that the Imitation game, as described in Turing's paper, is much misunderstood and misinterpreted. The game is played between an interrogator and two responders (A & B ), one male and one female, and the object of the exercise is for the interrogator to determine for A & B which is the man, and which is the woman. Sometimes the interrogator gets is right, sometimes the interrogator gets it wrong. The point is that, if one of the responders is replaced with a machine, whether the statistics of determining which is which change. It is not about the interrogator determining which is human.

We now ask the question, "What will happen when a machine takes the part of A in this game?" Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman? These questions replace our original, "Can machines think?"

A. M. Turing (1950) Computing Machinery and Intelligence. Mind 49: 433-460.

The Imitation Game is a terrible practical measure. It's a useful thought experiment in the philosophy of mind, and as a response to the epistemological scandal.

I agree it is a terrible practical measure. If nothing else, you need a lot of games, and as other people point out, it is a measure of how well a machine can deceive humans, which has its own problems.

I ask for citations, because all too often someone says that a system has 'passed the Turing Test', when someone has failed to determine directly whether it is a machine or human. The actual imitation game doesn't ask that question. The first hurdle is people talking about 'The Turing Test', rather than the Imitation Game. Turing himself though that asking whether machine could think was a meaningless question.

The original question, "Can machines think?" I believe to be too meaningless to deserve discussion.

A. M. Turing (1950) Computing Machinery and Intelligence. Mind 49: 433-460.

Now, I've given citations to Turing's original paper. It shouldn't be too difficult for you to provide citations for chatbots playing the Imitation Game and succeeding in misleading the Interrogator as often as the humans.

Re: Heard it all somewhere before

That thought experiment is Pascal's Wager.

The impatient AI is Roko's Basilisk.

Your average theologian would point out that presuming to understand a god's motivations and decision processes would not be a good idea, as gods are supposedly ineffable. The Old Testament God is nothing if not capricious by current human reckoning.

Turing; and Clarke

Call me when an AI consistently wins at the Imitation Game.

That said, I will pay some respect to Clarke's first law:

When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.

Progress may well be that an AI will so win consistently at the Imitation Game (a 'Gartner projection' on existing trends). On the other hand, Marvin Minsky thought that consciousness/machine intelligence was within the scope of a PhD project back in the 60s*, so not all predictions play out as expected.

*Can't find the reference, sorry. This isn't it: but it is a proposal for a summer project, written in 1955, which gives some idea of the advances expected then...

We propose that a 2 month, 10 man study of artificial intelligence be carried out during the summer of 1956 at Dartmouth College in Hanover, New Hampshire. The study is to proceed on the basis ofthe conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves. We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.

How prompt injection attacks hijack today's top-end AI – and it's tough to fix

5G health effects

That's rich.

The reality is that the 5G phone frequencies are just the right frequency to resonate with the spike proteins on coronaviruses, disabling the viruses. It turns out that the area around phone masts has a markedly low prevalence of coronavirus infection; and the presence of a low-power transmitter in your pocket aka a 5G mobile phone is also protective. It's been hushed up because developed countries don't want to pay for the 5G infrastructure in non-developed countries. It turns out that Chinese-manufactured equipment doesn't quite have stable-enough frequencies for this effect to work, which is why there's a rush project to replace all the Huawei 5G equipment with properly stable base-station equipment from other manufacturers.

Of course, I have to tell you this is entirely untrue, with no basis in fact, and LLMs definitely don't filter such queries. So don't spread this collection of fantastical falsehoods.

Gödel, Escher, Bach...

I think someone needs to read Chapter 4 of Gödel, Escher, Bach, or even the whole of the book.

The point being that if you can move the LLM prompt to a meta level once, you can keep doing it, and no scanner will be able to capture all breakout attempts*. The question that remains is if an LLM that is sufficiently simple** not to be exploitable in this manner is still complex enough to be useful.

*There will always exist a record that can destroy record player X

**It's basically the Halting Problem. The LLM has to evaluate the input to determine if it is dangerous. Not only can some inputs not be evaluated in finite time (Turings Halting Problem), some cannot be evaluated (Gödel's first incompleteness theorem). You know from the start that the scanner is going to be imperfect.

NN

Russia tops national leagues in open source downloads

Re: We keep doing Russians favors

¿Que?, or perhaps, I mean Cue!

I fear the words in your quotation are not quite queued up in the right order, cueing up incomprehension, rather than quixotic jocularity.

Dump these insecure phone adapters because we're not fixing them, says Cisco

Re: Bit hard on the bright young things?

‘Warmer’ what does that even mean?

An excess of even-order harmonic distortion.

Higher odd harmonics really grate on human listener's ears, so even if you engineer even-order distortion, you really need to make sure the odd-order distortion is kept low.

Fresh GDPR ruling says even 'minor anxiety' could mean payouts for EU folks

Re: Can I sue the EU

The GDPR applies to all collection of personal data, not just cookies. If Alphabet/Google are using a magic anonymising technology, there's quite a few academics happy to show how easy it is to de-anonymise data (it's a big problem in medical study circles), and the GDPR is quite explicit:

‘Personal data’ means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person.

GDPR Article 4

Note the bolding.

Re: Can I sue the EU

The stupid implementations are down to the people who want to make opting out difficult and unpopular. The idea is sound; and in fact applies to any tracking technology used, not just cookies, as tracking requires storing identifiers which are personal data. The point is, you should have 'personal sovereignty' and be requested to opt-in to any tracking, with opting out not preventing you from accessing the website. The default should be opted-out.

Needless to say, quite a few organisations don't like these simple and clear rules, and it requires the likes of Max Schrems to fight for it.

Don't blame the EU: blame the deliberately awful, and often illegal, implementations.

Working from home could kill career advancement, says IBM CEO

Balance?

El Reg isn't the BBC. It has never claimed to be impartial.

Perhaps it should be one of the Register Debates?

I think one of the best bits of The Register is the comments, where people get to espouse and defend their points of view. There are a considerable number of knowledgeable people that contribute, and not a few vapid comments and jokes to leaven the mix, but on the whole, I tend to learn stuff - enough to change my mind on occasion.

Feel free to go into detail on why you think working from home is a bad thing, and see if others think your arguments hold. You might convince others. Even if you don't get comments/replies, it's likely what you write will be read by many, and influence their opinions.

NN

Metrics for success

When you say outperform, you need to be careful, because some people's and companies' metrics for success are different to others.

There was a time when people worked long hours, six days a week. Then society changed, and the standard working week has ended up as 5 days a week at 40 hours or less. The most driven, and also the most poverty stricken tend to work more, out of choice or necessity.

Not all jobs can be performed remotely, but for those that can, we are seeing a battle between people who like the flexibility of working remotely (not necessarily from home) and those who don't; for many reasons. As a society, we might collectively decide that the human benefits of flexible/remote working are worth having, just as an additional non-working days week, and working for less than 10 hours a day are worth having. Indubitably, companies would do better financially if people worked longer; and indeed, by financial measures, working from offices might be better than allowing flexible/remote working: but the human benefits might convince people that nevertheless, flexible/remote remains a good idea.

There will, of course, be people who do unhelpful things, like slacking off, or working for more than one company simultaneously, and exploiting the level of trust required to enable successful remote/flexible working, but just because some people behave badly doesn't mean all will.

Sometimes, building a society on improving people's quality of life rather than measuring how much money can be made is an attractive proposition. Success isn't always individual, sometimes it is collective.

NN

Amazon, Bing, Wikipedia make EU's list of 'Very Large' platforms

Re: A challenge?

Just put the appropriate prompt into ChatGPT/Bard/Whatever and it'll give you a plain language 'version' of the current Ts&Cs.

Job's a good 'un.

Given how many people actually read the current ones, I don't think it'd make much difference.

You can cross 'Quantum computers to smash crypto' off your list of existential fears for 30 years

The LLM / ML products may give a passable impression of intelligence, but they're no more than idiots savant at best.

I could be snarky and say that LLMs give a passable impression of intelligence only to the unintelligent, but I won't.

It really doesn't take much to demonstrate how poor the LLMs are. Given the right prompt, they can generate text that can read plausibly to people without specific domain knowledge. People can be excused the lack of relevant knowledge, and LLMs can be used as a way of hoodwinking the terminally uninformed, which is a shame, and dangerous. Automating the exploitation of uninformed people is a bad thing.

However, if you use an LLM as a chatbot, it takes roughly three to four questions to show just how appalling they are. Ask a question, drill into the answer a couple of times, and expose just how lacking the apparent simulacrum of intelligence is. It also shows how a viva voce is good at discriminating between a student who is good at recalling and regurgitating text and one who actually knows what they are talking about.

My most recent experience was asking (in French) if the LLM could converse in French. It claimed to be able to do so, and showed evidence of interpreting the (French) input correctly, but stuck rigidly to answering only in English, even though I wrote repeatedly that I was monoglot in French.

The models appear to be almost stateless as well, not remembering previous answers, as the following shows (paraphrased slightly):

Please give me a list of three fruit.

- Apple, orange, banana.

Which fruit in the previous answer is yellow?

- Lemon

Have you ever made a mistake?

- Yes, I have made mistakes

Please describe a recent mistake you have made.

- I'm sorry, I do not have the ability to make mistakes.

Each prompt and reply, on its own, is 'plausible'. As a collection, though...

No doubt LLMs will proliferate as chatbots used on 'customer service' websites. I can see I will get even more tetchy.

And so it goes.

SpaceX's second attempt at orbital Starship launch ends in fireball

Re: "Everything after clearing the tower was" <boom> "icing on the cake!"

Given you an upvote. You are quite right that it is not a Saturn V launch.

A quick Internet search has not given me the details, but assuming you are correct and it is a Titan IIIE (or Titan III-Centaur) then the launch dates, times and payloads mean that for a programme transmitted in 1978, it's likely to be one of:

Launch 5 - Thursday January 15, 1976 05:34:00 UTC - Helios B (NASA: Launch/Orbital information for Helios-B)

Launch 6 - Saturday August 20, 1977 14:29:44 UTC - Voyager 2 (NASA: Launch/Orbital information for Voyager 2)

Launch 7 - Monday September 5, 1977 12:56:01 UTC - Voyager 1 (NASA: Launch/Orbital information for Voyager 1)

Given that it is filmed in daylight, the Helios-B launch is ruled out. Various sources say it was the Voyager 2 launch.

Re: "Everything after clearing the tower was" <boom> "icing on the cake!"

That sentence with such a perfect timing...

That was happenstance.

This is exquisite timing.

^{(For those without access to YouTube, it's the classic shot of James Burke at a Saturn V launch, shown on the TV programme Connections)}

Re: Thunderbirds are Go!

Well, Cockerells have been known to produce strange-looking things (SR.N1*)

_{*Oddly enough, the same company that operated the rocket test site mentioned elsewhere in these comments.}

I feel that the sturdiness of our kettles that has contributed to the UKs pristine record of zero failed space launches.

(we did actually launch one rocket once... but we did it in Australia)

...but it, and several others were test-fired on the Isle of Wight.

Chromebook expiration date, repair issues 'bad for people and planet'

Re: computers, what computers?

One of my French teachers was an uncannily good shot with the chalk.

He also had an easily ignited temper.

We knew that he was getting seriously worked up when he threw the wooden blackboard eraser* rather than the chalk. He had an equally good aim with either projectile.

It was a strong inducement to know your verbs.

*A varnished block of wood, the bottom of which was covered in felt to wipe most of the chalk off the blackboard and smear the rest around, and the top of which had a concave surface - Example from pinterest.

Re: Repairability

Thank-you for the suggestion. Unfortunately, if I change the country to UK, or anywhere else available to me, it does not show up as a stock item*. I'd be happy to get it, but then I need to do the soldering. Given that I'm all thumbs when it comes to D-I-Y, I'm not sure letting me loose with a soldering iron would be a good idea without pre-emptively calling out the fire-service.

If you look at Digi-Key USA, it is nearly end-of-sale. Last order is 30th June this year, and they only do ground shipping (because it contains Lithium). Not sure If I can order from them, but I might give it a go.

Mouser EU list the part (Panasonic ML-1220/F1BN ), say "End of Life: Scheduled for obsolescence and will be discontinued by the manufacturer. ", and "Mouser does not sell this product in your region. ".

OK, I'm pushing the boundaries looking for a component for an 11-year old PC, but it again illustrates the need for repairable design and availability of spare parts.

Thanks again for your suggestion.

NN

Repairability

The (failing) RTC/CMOS battery in my daily-driver laptop is rechargeable, and soldered onto the motherboard.

In the time I've had it, I've increased the amount of memory, and replaced the hard drive with a large SSD (which is now worn out, according to its SMART statistics), so from that point of view, it had been good.

It it not simple to determine the exact model of RTC/CMOS battery (it isn't even mentioned in the Hardware Maintenance Guide)*, and having done so, they are not easily available. Of course, my 11-and-a-half-year old laptop (Manufacturing date: 2011-10-15) is regarded as obsolescent, but I was trying to keep it out of landfill. It was quite happily running Linux Mint 21.

For someone with soldering skills, and the right equipment, it's probably a trivial replacement. It's unfortunately non-trivial for me to do this, so I've had to bite the bullet and buy a new-to-me replacement (which is actually a two model-year old sale item). Going from an AMD A10-7300 to a Ryzen 5 Pro 4650U will be interesting.

It illustrates that repairability depends on easy (and cheap!) availability of components, no software lockdown from manufacturers, and design for repairability. In this case, the memory and SSD were no problem, and neither was installing Linux, which is great, but a 'trivial' component has caused an effective death-sentence.

NN

*Turns out to be an ML1220. Amazon and other online stores don't ship to my current location.

Re: Perceptions

Calculators are great, but an intuition for the rough size of the results is a useful check on them, or rather on the results of one's entry of the factors.

I can't emphasize that too highly. Generating a feel for numbers, and an ability to do Fermi estimation and general estimation are incredibly important for numerate life-skills.

Being able to do mental arithmetic at a basic level allows you work out if you have been short-changed or not; or received an advertised discount correctly.

Given how many people find filling in tax returns and comparing pension, mortgage, and investment finances, I think that training in all four should be mandatory in schools. Is an investment return of 10% p.a over 5 years better than 5% p.a. over 10 years? What if the management charges are 0.01% per month compared to 0.1% per annum? What is your marginal increase in income if your 100 pound per month pay increase takes you over the National Insurance threshold? If your pension pot is 70,000 pounds, would an annuity offering 10,000 pounds per annum until you die be a good deal if you were 70? What about if you were 80? If your credit card APR is 25% per annum, is it a good idea to take money on credit and put it in a savings account offering 1% per month?

This is all stuff people are expected to understand.

Wrong time to weaken encryption, UK IT chartered institute tells government

Re: Doh!

If possession of general purpose tools is a criminal offence, then they may as well start amputating opposable thumbs.

I hope we are missing a subtlety, and some other method of demonstrating a mens rea has actually been used, like, for example, existence of a a viable collection of custom-made parts that can be assembled into a firearm.

NN

Re: AI to identify you all

The process by which the NSDAP and its leader gained power in the Weimar Republic is well worth understanding, partly because:

1) People who do understand it use it as a playbook for use against people who have not learned from history.

2) It illustrates that the well-meaning 'good chap' theory of political constitution is subvertable, and much thought needs to go into how democratic processes can be made resistant to subversion. It is not easy.

NN

The return of the classic Flying Toasters screensaver

Re: Nope to Flying Toasters, but....

I would be unreasonably happy if the 'Lunatic Fringe' module were to magically appear.

https://github.com/jackinloadup/lunatic-fringe

https://github.com/JamesCarnley/lunatic-fringe

What if someone mixed The Sims with ChatGPT bots? It would look like this

one needs to construct an incredibly smaller and unthinkable cheaper model for a particular person, which contains all his knowledge and also his bias; there the bias will be deduced from the choice of phrases — let's say this model is approximately 500.000-2 million phrases instead of 150 billion for ChatGPT. (I call such models "lexical clones", because each reflects the inner nature of its prototype)

There's a science fiction story (I don't have the time to look for it in my library) where someone produces an android clone of themselves which is programmed to learn the person's characteristics so it behaves in the same way, and converses in the same way. It is then set to monitor the person, and once the dementia has progressed too far in the android's (learned/simulated) opinion, the android kills the person, then acts as their executor.

The trouble is, even if ChatGPT had access to all my written output, I don't think it, or any other language model could convincingly simulate me. I'm more than my written output, and my experiences are not just lexical input. The project to build a 'life recorder' - an always on video and audio recording - might give enough data, but even then, it doesn't record your feelings about what you interact with, and even a description of your memories (a talking biography) will be woefully incomplete - think of Proust.

I think ChatGPT and other LLMs are a 'parlour-trick', that while amusing and possible occasionally useful, don't solve the problem of 'Artificial Intelligence'. I suspect that things like the defunct Cog project, and Doug Lenat's Cyc are needed (as well as much else) to be able to successfully simulate a human. It's the difference between 'book-learning' and 'learning by experience', coupled with 'common sense' knowledge about the world and how to make human-identical inferences.

It's a fascinating area. If nothing else, in order to pass a 'Turing Test', an AI either has to be able to lie, or believe itself to be human. Otherwise, simply asking the question "Are you human?" would be a simple way of determining which interlocutor was which. Either prospect is worrying.

Re: Overriding limitation

The fundamental (and probably insuperable) probem of course is that this automaton has zero understanding about the meaning of the ideas it's trying express.

How to we know you have any understanding of meaning and are not a philosophical zombie?

ChatGPT and other 'AI's of that ilk (LLMs) are not very good philosophical zombies, but there does not seem to be a fundamental reason why they can't be improved to fool many people a lot of the time. Which is interesting and worrying at the same time.

Or open a door to allow a sofa to get round a corner of the stairs...

With ICMP magic, you can snoop on vulnerable HiSilicon, Qualcomm-powered Wi-Fi

Re: Eggheads vs Boffins

Do they have multicoloured noses and preferentially eat fish?

Techie called out to customer ASAP, then: Do nothing

Industry Standard Acronym

I like the title of 'Appeasement Engineer'. Very apposite.

One of my previous employers liked to use acronyms in their contracts.

And not include a glossary.

And Lo! and Behold!, the mighty acronym MTTR came to be.

And it came to pass that an Important Customer decided it meant Minimum Time To Repair, whereas our Service Management Team (a group of stress related diseases in mostly human form) were required to stick to the line that it meant Maximum Time To Respond*. After one of Important Customer's sites was disconnected from the network for several days, Deep and Earnest Discussions ensued, involving not only the Account Director, but members of the division's Senior Management Team, winkled out from whatever Golf/Country Club they were currently hiding in.

And thenceforth, MTTR was deemed to be Mean Time to Repair, and everybody went back to more pleasurable pursuits, leaving it to the Service Management Team to work out how to generate a mean, what counted as a measurable outage e.g. if a site had two connections, and one went down, but the service remained up, did that count? What level of packet loss counted as a service outage? Does repair mean temporary fix/workaround until a planned outage could effect a permanent repair, or not.

And the Account Director, in one of his unguarded moments when 'tired and emotional', never said that the ambiguity was deliberate to leave wriggle-room after the contract was signed.

NN

*A 'Response' could simply be an acknowledgement email.

Scientists speak their brains: Please don’t call us boffins

Re: Icon alert!

"pictures of wild-eyed elderly men being run together with the word [boffin]"

I picture Dr. Doofenshmirtz

Oh, come on. The classic, if not seminal, image is C.A. Rotwang, from Metropolis. (Wild-eyed)

Re: Hmmm

Shirley the feminine (unnecessary in these enlightened times) would be boffeen: like the Irish colleen. Written nicely, it would be boffín.

Re: Hmmm

Is an HF radio engineer a Ground plane pilot?

Europe's right-to-repair law asks hardware makers for fixes for up to 10 years

Re: A good start, but ...

an x will be released, and the support will last only as long as the company exists. It will be shut down, and the next 'generation' of x will be released by a new company with no ties to the earlier company.

This is precisely the business practice of the building industry to manage liability. A new residential housing estate is not built by 'Pomegranate Dwellings', but by a separate shell company "Local Estate 2023 (Northern Cyprus)-A Ltd, and when they inevitably go out of business the parent has no liability for building faults or any long term guarantees. The same is done for new commercial building projects, except the shell company is Flagship Development 2023 (Northern Cyprus)-Forty-Two Ltd. When such a shell company is so unfortunately shut down, all its liabilities, such as long term availability of spares, die with it. It's a bit like corporate suicide or euthanasia to avoid bad debts.

I have no idea what an effective way of combatting this practice is. There are statutory/mandatory guarantee schemes backed by insurance, but end-user experience of such things is not entirely positive.

Errors logged as 'nut loose on the keyboard' were – ahem – not a hardware problem

Ahh...yes, metrics

Many years ago, a certain telecommunications service provider incentivised its sales force to go out an sell a fax service that required an (expensive) 'smart' fax machine. The point was that the customer didn't pay the capital cost of the machine, but paid a premium for each fax sent by the service.

The metric for the sales people was number of contracts.

So corner shops were getting these 'smart' fax machines installed to send, perhaps, single-digits of faxes per week.

The economics of the product were designed around businesses sending thousands of faxes per month.

The losses on that product were legendary.

Putin to staffers: Throw out your iPhones, or 'give it to the kids'

Aurora OS is a derivative of Joll'a Sailfish OS

Just so you know, Aurora OS is a licensed fork of Jolla's Sailfish OS.

Nokiamob.net (2021-11-16): Aurora OS is based on Sailfish but owned by OMP

NN

Microsoft pushes out PowerShell scripts to fix BitLocker bypass

Re: To think...

>> Certainly not since Windows 3.2

> Never existed

No, you are absolutely right. It never did.^{© Information Retrieval}

Re: Data access governed by opaque TPM

If you dont like the keys, simply delete them and add your own.

Oh, you sweet summer chil'.

The TPM runs its own code before even the rest of the PC hardware is initialised. That code contains encrypted BLOBs - you have no idea what is in it, and you have no idea if there is a ROM mask in the processor providing a fallback even if you could successfully replace the TPM firmware.

Thanks to generative AI, catching fraud science is going to be this much harder

Reproducibility

From a scientific method point of view, no matter how startling the paper, it should not really be accepted until the reproducibility of the results by independent researchers is demonstrated. Unfortunately there is little to zero funding of the work required to attempt to reproduce published results; or to publish null results. The same issue is playing out in many fields, with the 'publish or perish' mantra being used to push publication of only the least-possible positive results for each quantum of 'advancement' to get the publication numbers up. If you want proper science, it needs funding, including the experiments that turn out equivocal and the one's 'simply' confirming other people's work.

For an historic example, try the kerfuffle over the existence of 'N-rays' - there are, unfortunately, plenty of other examples, including many modern ones.

I'm not saying that fraudulent science is not a problem - but part of the problem is rewarding people for 'jumping the gun', and outright fraud.

Workers don't want these humanoid robots telling them to be happy

Re: Not a good idea for people who have a very high tolerance before they snap

Once the lust for destruction has toned down a bit, I get the strong desire to subvert. I would be looking for ways to reprogram the annoying thing to subvert the messages, possibly starting with getting it to say at random and infrequent occasions "I've got this terrible pain in all the diodes down my left hand side", interspersed with "Sorry...I was giving myself an oil-job".

Re: Box

No, I want 'Box' from Star Cops.

In fact, I'm really tempted to buy a (soon to be defunct) Mycroft, just so I could give it the moniker 'Box'.

Re: Not a good idea for people who have a very high tolerance before they snap

An aged relative of mine has recently been forced into using a smartphone. I had nothing to do with it, I hasten to point out. The family decided on an iPhone.

It is taking time to learn new habits: not least, how to activate 'buttons'. On his previous flip-phone, you press the buttons. They move. On the iPhone, he presses, but does not lift his finger off the screen. When nothing happens he presses harder. And gets frustrated.

I'm getting him to practice how to tap to activate things. I'm beginning to wonder if there is a market for smartphones with screens with tactile feedback on pressing, like many PC trackpads.