Posts by vtcodger

Signalgate lessons learned: If creating a culture of security is the goal, America is screwed

Re: Who is to blame?

This is a fallacy

I beg to differ. Ideally cryptography should be transparent. Otherwise you have the problem of training the users who will in too many cases turn out to be untrainable. Examples to point -- Hillary Clinton, Donald Trump himself, and -- apparently -- Pete Hegseth.

But cryptography alone can't provide security. It's just one component. There are a bunch of other problems. For example: restricting information to those with a need to know. Or communicating secure information in a timely manner. And cryptography and similar tools are both costly to implement, and costly when, as all too frequently happens, they fail "safe" and folks who need information don't or can't get it.

Downward DOGE: Elon Musk keeps revising cost-trimming goals in a familiar pattern

PR Problems

Somewhat off topic: Elon seems to be generating a few PR problems with DOGE. Then there's this tesla-whistleblower-says-musk-wanted-to-deport-her-team-for-raising-brake-issue/

Re: Just another Trump lie

You know, there's a proven way to reduce government waste/fraud/inefficiency. That was done in the 1950s. Hoover Commissions are named after the chairman of the first such commission -- ex president Herbert Hoover. The basic idea is that the commission of experts studies some aspects of government then produces reports with a list of recommendations. The recommendations that require legislation are sent to Congress. Those that can be implemented by executive order are (mostly) ordered with the caveat that there is a 60 day window for Congress to nullify them. Most of the recommendations were actually implemented.

The consensus is that they worked pretty well. It's been 60+ years since then. Very likely high time for another set of Hoover Commissions.

Instead, we've got this preposterous DOGE thing.

NPR and PBS are largely funded by user donations and advertising (although they don't call it that). As I understand it, the deal in 1969/1971 was that the government would fund the initial establishment of the Radio (NPR) and TV (PBS) networks, but that the government funding would slowly fade away. The increasingly unreliable internet assures my that NPR is down to 1% government funding and PBS is around 15% government funded.

Small ocean swirls may have an outsized effect on climate, NASA satellite shows

Color me confused

If anyone is like me, quite confused after reading the Nature article about what SWOT is up to and how it differs from conventional satellite altimetry (Jason), there's a much clearer summary of the mission and the instrumentation at https://swot.jpl.nasa.gov/mission/overview/ and the linked pages. Basically SWOT has a wider field of view than conventional satellite altimeters and flies in slightly uncoventional orbit(s) designed to optimize its mission of surveying Earth's surface water altitudes in detail. Conventional satellite altimetry is more focused on obtaining an accurate height estimate for global sea level

Tesla fudged odometer to screw me out of warranty, Model Y owner claims

Re: Vehicle odometry, ..

Assuming a mechanical odometer rather than a digital one, here's a good article on how they work https://auto.howstuffworks.com/car-driving-safety/safety-regulatory-devices/odometer.htm Lot's of ways for it to fail completely and just stop working. Not easy to see how it could be way wrong and still work after a fashion. Maybe it was replaced at some point with an odometer for some other model with a different transmission and gear ratio?

Re: Cockup?

GPS can yield quite accurate latitude longitude positions (elevation isn't as good for reasons I won't go into). But it requires multiple readings. For high precision, differential GPS which requires a signal from a nearby site at a known location is often used. For mobile applications, there are a bunch of problems if buildings, terrain or even really dense foliage block signals from some satellites. In urban areas, multipath reflections off buildings can lead to substantial errors. There also are, or at least used to be, some GPS "dead spots" where GPS works poorly or not at all for no very obvious reason(s).

Don't get me wrong. I think GPS is terrific. But as with much to much modern technology, many people tend to give credence to the marketing hype and ignore the worst case problems.

That said, I'm curious why one would use GPS for speed estimation when wheel speed sensors on all four wheels are presumably needed for ABS braking. I'm 98% certain our 2017 uses those sensors to drive the speedometer/odometer.

Re: Cockup?

"See Kalman filter on Wikipedia"

We're off in the world of conjecture here as I don't think we really know how Tesla mileage metering works. But I should point out that Kalman filters although much loved by mathematicians, don't necessarily work all that well in practice. Basically Kalman filters build a model of whatever is being analyzed using statistical techniques. They then reject "noise" -- which is to say data points that seem improbable based on the model. That can work brilliantly. If it didn't, Kalman filtering wouldn't be used much. But they can have a problem if the data rejection isn't set up properly for the particular problem. In simple terms, the problem is described as "tracking the model". What happens is that the filter works well unless the conditions change. Sometimes, if conditions change too quickly, the filter can fail to adjust it's criteria fast enough. It will then reject the new data as being too unlikely to believe.

Example: An aircraft flies in a straight line for a long time at constant speed. A statistical filter smooths radar data observing the aircraft and, once everything settles down, it comes up with much better estimates of position than the single radar data points would give. However, if the aircraft abruptly changes direction and or speed the model may not "open up" sufficiently quickly to accept the new data. The model will then wander off projecting the aircraft to be somewhere it isn't.

I scanned the Wikipedia article and didn't see that addressed. My guess would be that it is addressed, but not necessarily in terms that are easily understood.

Is it possible that a Muskmobile could have a faulty shiny new fangled digital technological odometer? Seems credible to me

Guess what happens when ransomware fiends find 'insurance' 'policy' in your files

Re: What people are willing to pay to avoid ...

But aren't those transaction backups going to be encrypted along with everything else the ransomware can get to? Not being argumentative, just curious. I suppose one can try to make the transaction backups read only. Probably could be done? But likely not as easy as it sounds?

Re: What people are willing to pay to avoid ...

Backups are fine. They (usually) aren't that hard to do. Although they can be hard to test without a spare machine to test them on. You really don't want to load them back onto your system if they turn out to be corrupt or unreadable. They'd work well for me and many businesses.

However, they aren't going to work so well for online businesses because all the transactions between the last backup and its time of reload will have vanished into cyberspace.

Furthermore, it's probably only a matter of time before the malware folks start poisoning backups and delaying the encryption and ransomware demands for a week or two. What then Kemosabe?

So what's the answer? I don't have one. I don't think anyone else does either. Except maybe a complete, thorough, and no doubt incredibly painful rethinking of what it is safe and reasonable to allow over a worldwide communication system where every scumbag on the planet is your next door neighbor.

EU gives staff 'burner phones, laptops' for US visits

Re: Good practice

"So the Pilgrim Fathers received an invitation ..."

Actually, they did. It's a long story, But basically a member of the abenaki tribe named Samoset who had been to England wandered into the Pilgrim settlement one late March day and asked if he could have a beer (No, I'm not making this up).. Samoset had been to England, and apparently spoke some English. He explained the political situation with the local tribes, left and came back with some furs and brought along a member of the Wamponoag tribe named Tisquantum (Squanto) who had been to England and Spain and apparently spoke English pretty well. A treaty was negotiated and Tisquantum ended up settling in with the Pilgrims as a translator and advisor. For more details see the Wikipedia article. https://en.wikipedia.org/wiki/Squanto.

Not that this has anything to do with America's far right's weird obsession with immigration and their bizarre determination to evict the people who are doing the jobs Americans down't want to do like picking strawberries all day in the hot sun,

Apps-from-prompts Firebase Studio is a great example – of why AI can't replace devs

Re: AI Coding Agents: Failing to Do the Most-Common Programming Tasks

Actually, I think creating tests is something AI might be useful for. The job is difficult,tedious, and -- in my experience -- very few people are much good at it. Again, in my experience, it is rarely done well and frequently doesn't seem to be done at all. I think AI might, and I emphasize MIGHT be able to do as good a job and maybe even a better job.

OTOH, I'm quite skeptical of the ability of AI to create even mediocre programs. If for no other reason than that specifying what a program should do is another task that is monumentally difficult and rarely done well.

Infosec experts fear China could retaliate against tariffs with a Typhoon attack

Re: China only needs to wait a few days.

I'm inclined to agree with you. Another reason is that China seems to be almost as reliant on the internet as the US. China attacks the US in cyberspace and the three letter kids in DC retaliate. Mutually Assured Chaos.

The difference might be that China MIGHT make some serious effort to harden their digital infrastructure. The US -- especially with the collection of crackpots Trump is assembling to "advise" him (Since when has Trump listened to advice from anyone?) is unlikely to do any hardening that impedes immediate profits.

Laser-cooled chips: Maybe coming soon-ish to a datacenter near you

Re: A question

Looks plausible. Thanks.

Re: A question

Possibly answering my own question, it looks like it MIGHT work as follows.

1. The laser photons excite Gallium Arsinide atoms,

2. The atoms then return to their normal (ground?) state emitting a photon.

3. The emitted photon is at a shorter wavelength than the laser photon was (fluorescence) thereby carrying off all the energy from the laser photon and some kinetic energy extracted from the GaAs

3a Igor or whatever Google's AI agent is called informs me when describing fluorescence that the emitted photon should be at a longer wavelength than the absorbed photon and therefore carry less energy. Shows how much Igor knows.

Or maybe Igor's right and this works some other way.

A question

I'm not much good at physics, but I do believe fervently in the principle of Conservation of Energy. My question -- which seems to me not to have been answered -- is "Photons carry energy. Where exactly is that energy and the thermal energy subtracted from the target ending up in this setup?"

I'm not questioning that these gizmos will do something that cools the target. I'm sure they will. But I'd sure like to know HOW. And I'd like some reassurance that the GaAs isn't being "consumed" -- which is to say being irreversibly altered in the process.

Canada OKs construction of first licensed teeny atomic reactor

Re: Small nuclear reactor? That's all very well, but...

Did the folks doing the promising mention that lead underwear might be a good idea just in case your Nucleon gets rear ended?

Re: could power up to 1.2 million homes

In truth, the "grid" has a kind of weird buffer to handle short term peaks. Basically, (almost) all the power sources attached to the grid are rotating generators and many of the loads are motors. Under the skin motors and generators are the same device with perhaps a few tweaks on the motors to help them start and, if it matters, to get them turning in the proper direction, There's a LOT of energy in those many rotating motors/generators. When you turn on a device -- a teapot, blender, TV, microwave, whatever -- the grid management doesn't go off and find a 100 watt or 1000 watt source and add it to the grid. Instead, every motor and generator on your grid slows down a tiny bit to provide just a wee bit of power to your device. When your device turns off, all those motors and generators speed back up a tiny bit.

Kind of weird. But it works. One of the affects of turning on a bunch of appliances (loads) is that the frequency of the AC power decreases. The power company monitors the frequency. If it slows too much they know they need to bring more generation on line. Likewise, if the frequency increases, they drop some power sources offline.

FreeDOS 1.4: Still DOS, still FOSS, more modern than ever

Progress?

While I would agree that there are probably better ways to do computing than it was done three decades ago, I'm not sure that the industry has identified them. The current state of computing software seems to me perhaps more like a third world slum than a shining city on a hill.

Boeing 787 radio software safety fix didn't work, says Qatar

Re: 90 Minutes to install a patch????

This is Avionics and safety critical equipment at that. I'm more familiar with military than civilian, but I expect they are similar. It will probably come as a shock to many of my fellow software types, but in the world of flight critical components for platforms costing $100M and up and where human lives are at risk, folks are expected to exercise discipline, follow checklists and do things "properly". Most likely, the 90 minutes includes reading the instructions carefully, checking the paperwork to make sure the patch applies to the hardware and software installed, removing the radio from a very likely hard to get to cranny in a cockpit designed for efficient flight control, not repair work. Then checking serial numbers to make sure what the paperwork says is installed is what's actually there. Then the equipment -- which probably does not have a USB port on the front (and really, you probably wouldn't want it to) has to be powered and the patch installed. Then it needs to be tested. Then reinserted back into that cranny using connectors likely designed more for reliability than convenience or ease of use. Then the device needs to be tested again. Then, whoever is in charge of making sure things are done right needs to be summoned if he/she hasn't been hovering the whole time. The QA person may well want to run their own test. Finally, if everything went well, the work order is signed off on and the (voluminous) documentation is updated.

Given all that, 90 minutes seems pretty reasonable. At least to me.

China hits back at America with retaliatory tariffs, export controls on rare earth minerals

Re: All is good!

Not to worry. Manufacturing will come back to the US and Canada ... Someday ... Maybe in 3 or 4 decades. Wherever and whenever factories become sufficiently automated that machines do 90% or more of the work.

Jobs? Not so much one suspects.

Mediatek wants to make Chromebooks more like Copilot+ PCs

Re: Chromebooks are student's labor saving devices

"Copying and pasting the AI agent's output is not education."

Of course not. But as far as I can see many, probably most, students are not very interested in being educated. There may be one or two subjects that interest a specific student. They'll probably learn those on their own. And they may be lucky enough to encounter a talented teacher who can bring some otherwise tedious subject to life. But mostly kids feel the same way about school that adults feel about tedious, repetitive jobs. If a Chromebook with AI helps them deal with stuff they couldn't care less about, they'll likely use it to deal with that stuff. It seems to me kind of unrealistic to expect otherwise.

Chromebooks are student's labor saving devices

And if you were questioning why anyone in the market for a Chromebook would want AI capabilities

Possibly because doing the assigned reading for a class then composing a 300 word essay can be so tedious when watching TV, playing video games, or hanging out is so much more fun. Copying and pasting the AI agent's output is much more time and labor efficient. Chromebooks are targeted at the education market after all.

Mozilla is rolling Thundermail, a Gmail, Office 365 rival

Re: It is not the ecosystem

I'm expect that Google would tell you that email to readable URIs is inherently insecure and the industry MUST replace those URIs with URIs of two or three thousand random characters.

I'm quite sure that the folks at Google are either very smart. Or stark, raving mad. Or possibly both.

Trump yanks CHIPS Act cash unless tech giants pony up more of their own dough

Re: Meh

Don't know from personal experience. But back in the 1990s, I was told by several sources that semiconductors for the US military were manufactured in US to Military Specs and individually (as opposed to batch) tested. I doubt that's changed much for combat/operational critical gear. Office equipment? Don't know. Might well be Commercial Off The Shelf (COTS).

LLM providers on the cusp of an 'extinction' phase as capex realities bite

Re: Not going there.

"I'll have what he's smoking."

Probably not a great idea. The High may be great, but I suspect that coming down afterwards might be a truly wretched experience.

Musk's xAI swallows Musk's X in ego-friendly, all-stock deal

Re: A slight case of overpaying

"So Musk is overpaying for Twitter again?"

Quite the opposite. Since Musk is said to own about 50% of xAI, the net effect of this merger would seem to be to shift about half the $44B he "invested" in the twitting turkey to his fellow investors in xAI.

Nvidia GPU roadmap confirms it: Moore’s Law is dead and buried

And furthermpre

You're dead right of course. Moreover, one probably can't dump heat into the moon's bedrock or regolith because dry dirt is a lousy heat conductor and there's (almost certainly) no water or other fluid to improve heat conductivity. The only way to dump heat to space is radiation. Even that is harder than it seems because the radiators would presumably be receiving radiation from the sun and getting quite hot during the two week long lunar days. That's not what you want on the "cold" side of a nuclear powered heat engine. There may be ways to work around all that, but they likely won't be pretty, simple, or cheap.

Re: Heating the air for telling us bad stories

Closer to 11000 kwh per year in the US -- mostly because of the need for air conditioning much of the year in most US climates. But yes, widespread AI (and all cryptocurrency) are resource sinks that an intelligent species would likely consider unacceptable. I must say that the proposal that intelligent life exists in this solar system although once sort of plausible seems increasingly unlikely as more data becomes available.

Re: Not Enough Compute Speed?

Nvidia is a hardware monger. Wouldn't their working on ways to get by with by with less hardware be kind of like a dog food, breakfast cereal or snack food vendor lacing their product with appetite supressants?

Museum digs up Digital Equipment Corporation's dusty digital equipment

"The 1990s were not kind to DEC, and the once-mighty tech giant faded from relevance due to a series of business decisions that, in hindsight, proved detrimental."

The 1990s were indeed unkind to DEC. But it wasn't management that killed them and competitors like SEL. What did the Minicomputer vendors in was that they occupied a market position that was devoured by substantially less expensive "desktop computers" The release of the IBM 5150 PC in 1981 signaled the start of a two decades squeeze on the minis. They couldn't move up very easily because IBM and its cousins had the mainframe market nailed down. They were slowly devoured from below as PCs became ever more powerful. Nothing much management could do about that. They were managers, not magicians.

Ubuntu 25.10 plans to swap GNU coreutils for Rust

Re: So far, the Rust coreutils pass approximately 500 out of 600 GNU tests

Most tests will presumably test a great many things so the failure rate is likely way below 17%. On the other hand, Unix core utilities are almost certainly used by millions of user shell scripts. Shell scripts tend not to be terribly flexible when it comes to dealing with even minor changes in input and output rules/conventions including stuff like spacing, character sets, and capitalization that probably wouldn't bother human users much. The tests that are being used best reflect the need for strict compliance with existing conventions or there will likely be problems for users.

Los Alamos boffins whip up a speedometer for satellites

Re: Relative to what?

The handful of systems I'm familiar with used Cartesian coordinates based on the Earth's Center, the plane of the Ecliptic, and something called the First Point of Aires (basically a line from the center of the Earth to the center of the Sun at the instant when the Sun appears to cross the Earth's equatorial plane in the Spring). But you can use any coordinate system you please although the math is likely to be messy if you were to choose, for example, an egocentric system based on the location of your stomach and a vector to the nearest pub.

Re: is there no GPS in space?

I think you're right that calculating speed would be easy. Except that the density of the plasma is constantly changing. Some of the changes are (e.g. those due to time of day) are predictable. At least in concept. Some (.e.g. those due to variations in the "solar wind") aren't.

Re: Kepler's second law

You're not entirely wrong,but it's a bit more complex than that. See https://en.wikipedia.org/wiki/Ellipse and https://en.wikipedia.org/wiki/Elliptic_orbit for way more information than you really wanted to know about the properties of an ellipse and of elliptical orbits.

Re: is there no GPS in space?

I guess maybe this gizmo might tell the satellite how much drag it is experiencing since the satellite presumably knows its orbital parameters and what it's drag free velocity would be. But I'm having a bit of difficulty seeing how that helps avoid collisions unless the satellite somehow knows the ephemeris (a table of calculated positions overtime) of the objects it is trying to avoid. And if it knows their ephemerides, how come it doesn't know its own?

Interesting technology. And probably its good for something. But I'm not sure what.

OK, Google: Are you killing Assistant and replacing it with Gemini?

It's Psychotic

Bob: It's psychotic! They keep creating new ways to celebrate mediocrity. Bob (Mr Incredible) Parr 2004

Earth's atmosphere is shrinking and thinning, which is bad news for Starlink and other LEO Sats

Re: An increase in space junk...

Surely intelligent aliens would take a quick look at this place and decide in about seven minutes that they have urgent business someplace far, far away.

Re: Adjust the orbit

Yes, the paper actually discusses that. The downside is that dead satellites and miscellaneous junk in higher orbits will take even longer to burn up on their own if the thermosphere contracts or if future solar activity is lower than expected. There was a long period called the Maunder Minimum in the late 1600s and early 1700s when solar activity seems to have been very low.

Keep in mind that there are also factors other than collision possibility (actually rather low at present) affecting orbit choice. Coverage area, transmission delays, and such.

Re: Starlink?

FWIW -- Starlink satellites are mostly in 550km (give or take) circular orbits which suggests that they should last 10 to 20 years (depending on future solar activity) before re-entering on their own. However, the plan seems to be to deorbit them intentionally after about 5 years of service.

Judge says Meta must defend claim it stripped copyright info from Llama's training fodder

Re: What if ...

"a student uses an AI to generate ..."

And if that selfsame student goes to the library and generates the same product without the help of AI, is he/she guilty of doing anything other than practicing research? And isn't research exactly what the student is supposed to do?

Perhaps we're looking at fundamental problem with the concept of "intellectual property". With the exception of Trademark, IP seems to me a quite nebulous concept with an enormous gray area between clear "theft/misuse/abuse" and equally clear "fair/proper use". I have a lot of doubts about the ability of legal systems -- current or future -- .to resolve IP issues equitably.

On the other hand, I suppose it keeps lawyers employed. I reckon that's a good thing. Lord knows what mischief they'd be up to if they weren't arguing about IP issues.

'Cybertruck ownership comes with ... interesting fan mail'

Re: spoon-fed by media nonsense designed to make them hate Elon Musk and EVERY technology

Ah yes. But you haven't seen our block-chain enabled, AI capable spoonmwith regular OTA updates. We'll be rolling it out to a few select customers in a few weeks. To order one, just send an email with full financial details to [e-mail].

Here's the ugliest global-warming chart you'll ever need to see

Re: Denialists - move 'm to the beach

Current Sea Level Rise is about 3 cm a decade. Not a lot actually. But you need to add/subtract local tectonics to that. In any case buying a house close to sea level isn't especially bright because with or without climate change it'll almost certainly be within the reach of the strongest storm that might come along, and insurance companies have finally noticed that and are adjusting their rates accordingly.

Elon Musk calls for International Space Station to be deorbited by 2027

Re: Follow the money

"I would say they have a contract to supply X number of missions until 2030. With no exclusion of the ISS not actually being there."

Plausible. But not likely I think. Typically US government contracts don't work that way. Amongst all the hundreds of paragraphs forbidding the hassling of endangered species, requiring the contractor to fill in any holes they dig in the landscape, forbidding discrimination against midgets, etc,etc,etc. there are clauses that allow the government to cancel the contract at will. When they do that, the contractor is reimbursed for any expenses already incurred in anticipation of future missions, and the government picks up a lot of the costs of relocating contractor personnel, cancelling facility leases, etc. At least that used to be the case in the 20th century. I doubt it has changed.

HP ditches 15-minute wait time policy due to 'feedback'

Re: Re-parse that response

"and we'll quietly reintroduce ..."

Of course not. That'd be dumb. Do you think all big tech managers are idiots? What they'll do is introduce a random delay with average delay of 15 minutes. Or they'll introduce a ficticious queue counter that will inform the caller every two minutes that their call is very important to HP and they are now 171 from the top of the queue. Then slowly count it down. Or they'll just turn the whole support thing over to AI agents that make Clippy look helpful by comparison.

Time to make C the COBOL of this century

Re: C is the new COBOL

I suspect that most folks who want to replace COBOL with some "modern" language have never looked at a real COBOL program. I am not, never have been, and I'm quite sure would not want to be, a COBOL programmer. But I have had to peruse COBOL code a few times and found it to be remarkably readable -- more so than any other language I have encountered in the 65 or so years I've been dealing with computers. I can't say the same for C

I really can't see much of a case for abandoning COBOL. I can see a case for replacing C (and assembler) with a memory safe language where security is an issue and where such memory safe language has adequate performance and will fit into the machine memory.

Mobile operators brace for bigger, faster headaches with 6G

Re: Tera-hurts

Of course they will. But how are marketing folks to peddle their wares without constant progress? Or at least the illusion of constant progress. Thinner devices and rounded corners have already been done. And it's not like the manufacturers can add bigger tail fins or more cupholders. Have you no sympathy for the (potentially) starving families of industry leaders without 6G ... And 7G .... And 8G ... And .... ?

Re: 4g is fine

But how will you get future bigger, more complex, higher resolution advertising if you only have 4G?

Methinks that perhaps the inmates have taken over the internet asylum.

The Doom-in-a-PDF dev is back – this time with Linux

The next BIG thing

And QUANTUM, BLOCKCHAIN, AI will rule all. Well, just as soon as we fix the last couple of bugs.

Lawyers face judge's wrath after AI cites made-up cases in fiery hoverboard lawsuit

Re: what are you paying for

"so I'm really surprised so many US law firms are doing this."

Why are you surprised? A massive hype machine led by blatant scam artists is pushing AI to the public in hopes of recovering huge investments quickly. Lawyers are clever. Probably considerably smarter than the average human. (But not necessarily smarter than the average dog. After all, how many dogs voted for Donald Trump?) Why would it be a surprise that some lawyers actually believe the hyperbole?

I would be surprised if lawyers don't learn very quickly that they need to actually check any citations found by AI and make sure the cases exist and that the cases actually say whatever AI agents claim they say.

BTW, can a lawyer sanctioned by the court for citing ficticious cases turn around and sue his or her AI provider for damages?