Posts by vtcodger 2298 publicly visible posts • joined 13 Sep 2017
"Are you suggesting that we give the AIs access to high powered lasers and fusion reactors?"
Well, maybe. Presumably if we don't give them access, they'll just come up with some other way to exterminate humanity. Perhaps it'd be best not to piss them off.
Ya gotta remember. AI is the **NEXT BIG THING** and, like all **NEXT BIG THINGS** it can do **ANYTHING**
Why not just build a few -- 2 to maybe 6 -- internationally accessible, ITER like, research facilities for AI in order to see what it can and can't do? And ban ANY general use of AI anywhere that doesn't meet reasonable standards of safety, IP protection, power consumption, etc. Yes, that'll require a lot of CEOs/marketeers to find a different Next_Big_Thing. So what? Dealing with problems is presumably what they are paid handsomely for.
Geothermal is a perfectly reasonable source of power. They've been generating electricity at the Geysers complex North of San Francisco for 74 years. A respectable amount -- 900MW currently. There are many other geothermal plants operating in the world.
There are a few problems.
1. There's not all that much really hot rock at reasonable depth, so potential easy sites are pretty limited. You are probably not going to site a geothermal plant near New York City or London. Not any time soon. Probably not any time ever. Naples OTOH.
2. Natural geothermal fields have limits as to how much steam can be generated. Drill more wells than your field will support and your production from existing plants will drop.
3, Natural sites tend to be near volcanoes. There are certain drawbacks to building infrastructure on volcanoes. Parts of Hawaii's Puna plant including the access road were eaten by a lava flow in 2018.
4. I personally wonder about what else is coming out of the ground besides steam/superheated water -- Sulfur very likely. Toxic gases? Dissolved salts probably. Probably no problem as long as proper handling and disposal procedures are in place. But still ...
"There are terrestrial solar arrays ..."
That would be the Ivanpah (California) bird burner. https://en.wikipedia.org/wiki/Ivanpah_Solar_Power_Facility The facility used 13sq km of mirrors to heat 3 boilers to 560C. It also used large amounts of natural gas (4 times the original plan) to start up generation in the morning when sun angles are low. And it fried an estimated 6000 birds every year. the 2.2B USD faciility did generate well over 800Mwh of electricity in 2020 its seventh (and best) year. It had other problems including attempted self-incineration in 2016 (A software bug as I recall). and bankruptcy in 2019. Regulators were threatening to shut it down permanently in 2023. I have been unable to determine if they succeeded.
Overall, apparently not that incredible a success.
I think the FCC rejected Starlink's proposal because if Starlink is going to provide adequate service in cities like New York LA ,Seattle, Boston it'll already have all(?) the infrastructure it needs to provide rural broadband in the lower 48 and Hawaii. Little need for a subsidy. I'm thinking that they might need a few better placed ground stations for some very remote areas, but I don't know enough about their network to say for sure. Maybe they can get money for those.
Actuallly, I pretty much agree with two caveats.
1. Current EVs cost too damn much. Especially if one includes the cost of a home charger. That sort of disqualifies them for a large fraction of the potential users who are probably better off economically with a decrepit used ICE vehicle that barely runs, but is satisfactory few trips a week for shopping, and entertainment.
2. In developing countries the ideal EV for many users is probably not the same as the vehicles those of us in the developed world are looking at. It's probably something like a three wheel tricycle with a maximum speed of 30-40kph, but with enough power to lug 400kg of crops or the wife and three kids 20km to a market town ... and back. I don't expect to see that from Tesla, Ford, VW, or Toyota. Probably they will come from some companies we have never heard of in China, India, Indonesia.
You're probably right. I should have said "Those who don't make long trips thru thinly populated areas"
One problem with long trips in the US is that for many users, they correspond to holiday weekends when most everyone is taking long trips. Thanksgiving and Christmas for example. In populous areas there may be a lot of choice of charging locations, but out in the plains or Great Basin, or the Mojave, one is likely to find themselves at the end of a LONG queue awaiting access to a charger. The holiday weekend queues at the original Tesla supercharger site at Kettleman City half way between LA and San Francisco on I5 are legendary. Tesla, to its credit, has added sites there. But somebody is going to have to pay for all those rarely used chargers. I reckon that in the long run, it'll be the users.
The other potential problem is the chargers. Vermont seems on paper to have a lot of publicly available EV chargers. But reading the labels on the (non Tesla) devices in my local supermarket parking lot, it looks like they will give one, at best, 10kwh (40 miles or so on a nice June day) charge in an hour. There's not all that much to do in a rural strip mallish sort of place for hours while one "refuels." Especially at night. In Winter. Queuing probably won't be a problem though. Nobody much in Vermont owns an EV. I think I've seen two cars in the past 3-4 years actually use the things
I once had an elderly (10 year old) lead-acid battery actually freeze on me during a Northern New England cold snap. Took the fill caps off, and there was ice where sulfuric acid belonged. Surprisingly perhaps, after being thawed out in the bathtub,it managed to start the car.
Yes, oil can be a problem. My impression is that modern multi-viscosity oils have much reduced the problem. But a chemist once told me that the viscosity extenders are the first thing to fail if intervals between oil changes are too long.
Anyone who has actually done "due diligence" on pure EVs should know that the current crop are best suited to users who:
. Live in mild climates
. Don't make many (preferably no) long trips.
. Are driven frequently. (parasitic drains of various sorts will likely eventually kill the battery if you let them sit without charging for many months -- and yes people do that with ICE cars sometimes).
. Have a place to charge them (and a power grid capable of charging a large number of EVs if one's neighbors are also buying EVs)
. Don't drive in very mountainous areas.
One doesn't have to fully satisfy ALL those criteria in order to be happy with an EV. But satisfying most would be a good idea.
Hybrid EVs should be a satisfactory alternative for many folks. They still burn fossil fuels, but often significantly less than pure ICEs.
Future EVs will be better. Probably on all counts. But it'll be a long time (maybe never) before EVs are the right vehicle for every use case.
It's possible that AI will actually be good for something. But it looks to me like projections of AI future capability are dramatically in excess of likely actual utility. And my guess is that MS management might well end up forcing Clippy_on mind-altering_drugs on one and all.
Those who haven't long since jumped ship might want to start thinking about an exit strategy from Windows. Hopefully, they won't have to use it. But if they do, it's likely not going to be anywhere near as easy as just buying a bunch of Apple gear or simply downloading some Linux release. Being prepared to bail will probably require some effort. But if one does need to switch OSes, one does not want to do it in a state of panic.
Wouldn't we be better off with a small attack surface and a smaller bag of goodies for the bad guys to steal?
Exactly. You're advocating what I'm calling Plan-Z. Plan-Z -- reduce attack surfaces and store as little data as possible -- is where we'll likely end up in a couple of decades after Plans-A thru Y have been proposed, hyped, forced on long-suffering users because they're good for us, and eventually failed.
Meanwhile, my question after reading the article is -- What the heck is "SMART" about the RubeGoldbergish collection of weird stuff that allows this vulnerability to exist and be exploited?
"Personally, I'd much rather have fewer batteries and an on-board ICE generator to charge them. The combustion engine could keep running at peak efficiency, there'd be no heavy transmission (and that engine would likely be air-cooled, so no fluid cooling system either)"
That's called a hybrid and they are available today from most major manufacturers. Often at reasonable prices because their (expensive) drive batteries are generally not all that large. I think that most hybrids have transmissions. I'll leave it to you to suss out why because I don't know and don't much care. And yes, they get excellent range,
Battery swapping is perfectly feasible. IIRC Elon the cagefighter once demonstrated it in order to capture a California EV subsidy. The problem with battery swapping is economic, not technical. EV batteries are outlandishly expensive. How many owners of a new EV are going to be anxious to swap their almost brand new (but charge depleted) battery module for whatever battery pack of unknown provenance the "charging station" happens to have around and charged?
Also current EV batteries are heavy. The internet tells me that a 6 to 12 kwh battery (maybe 18 to 40 miles of juice depending on capacity and the vehicle) will weigh 100-150kg (220-330lb). Very few folks are capable of moving that sort of mass around. Unless you want to deal with dozens of battery packs at every refueling, battery swapping will require machinery and standardized fittings.
Probably things will change in the future. But not quickly I think.
Copyright and IP in general looks to be potentially a bottomless well of grief for anyone who uses it on a profitable project. I'm told that there is an old saying in the music industry -- "Where there's a hit, there's a writ." If your song makes significant money, someone is probably going to sue you because your song purportedly resembles their song from 1983. Now AI is probably going to extend that to artwork and maybe prose and possibly computer code.
Popcorn time.
Mostly what you say seems reasonable although there are large arid and semiarid regions where I suspect there is more than enough sunshine and plant growth is limited not by sunlight, but by availability of water. Spend some time in the Mojave or Texas west of the Balcones escarpment, and I think you'll conclude that too much sunshine is quite possible.
But what I really want to point out is that the only experience humanity has with deliberately tinkering with the atmosphere is cloud seeding to increase precipitation. The results of that seem not especially encouraging. It works fine on cocktail napkins. In practice, often not so much. Overall, it does seem to work sometimes. Probably. It often isn't all that clear that it wouldn't have rained or snowed anyway. It's rather expensive. And it can be argued that forcing rain/snow in one place likely diminishes rain/snow in downwind areas that might need the water as much or more.
What, you think the US doesn't have this technology?
Why yes, I think exactly that. Why would the Chinese bother to ban export of the technology otherwise?
BTW, if you read the Wikipedia article on the Mountain Pass mine, you'll find that the rare earth oxide concentration there is around 7% which is actually quite a bit higher than the typical commercial ore concentrations of minerals like Copper, Silver, or Gold. I suspect, but don't know for sure that the technology in question might have to do not with extraction, but with separating the various rare earth elements from each other. My hazy recollection is that the chemistry o the 17 elements is virtually identical -- which makes separating them from each other and purifying them quite difficult. I should think that someone else around here would know a lot more about this than I do.
Quite likely the best model we have for the capability and limitations of AI is IBM's Watson. Watson, it will be remembered, was an AI agent built by IBM a little over a decade ago to play a game -- Jeopardy. And it did that very,very well. IBM then spent many millions of dollars trying to adopt Watson to more serious (and more profitable) uses. None of those efforts succeeded. The full story can be read at https://www.nytimes.com/2021/07/16/technology/what-happened-ibm-watson.html
This seems to me a clear warning that the road ahead for AI is likely to be -- as Waylon Jennings would have it -- rocky, dusty and hard. A truly intelligent race would probably approach AI with extreme caution. Humanity however .... Best fasten your seatbelts folks. it's probably going to be a bumpy ride.
"Scripting has its place"
Ehrrr ... no, in the very long run I think it probably does not -- at least not in the way you mean.
I think a much restricted form of scripting directed solely at controlling content layout might be OK. But scripting today is synonymous with Javascript. And Javascript as it stands now looks to be far too capable to be compatible with computing security. If you let third party material that loads Javascript code into your computing ecosystem, I doubt any amount of protective superstructure can reliably keep the bad guys out of your machinery. And they only need to get in once to leave you with an incredible mess that will be very time consuming and very costly to fix. If it can be fixed at all.
I fear that in the very long run, I think we have have two choices -- secure computer intercommunication with the outside world or scripting. Pick one.
I'd like to be wrong about that because an awful lot of stuff including useful stuff and many serious businesses is built on scripting capability. But I don't think I am.
"Meanwhile NT has evolved into a monster that looks as if it wants to spend 30 years displaying the message "Preparing to Configure Windows. Don't turn off your computer"."
Somewhat reminds me of this:
Vasa or Wasa[a] (Swedish pronunciation: [²vɑːsa] ⓘ) is a Swedish warship built between 1626 and 1628. The ship sank after sailing roughly 1,300 m (1,400 yd) into her maiden voyage on 10 August 1628. ... Richly decorated as a symbol of the king's ambitions for Sweden and himself, upon completion she was one of the most powerfully armed vessels in the world. However, Vasa was dangerously unstable, with too much weight in the upper structure of the hull. Despite this lack of stability, she was ordered to sea and foundered only a few minutes after encountering a wind stronger than a breeze.
https://en.wikipedia.org/wiki/Vasa_(ship)
Seems to me that many of the problems with modern computing are due not to lack of innovation, but to marketing folk insisting on the appearance of innovation. Potemkin innovation as it were.
Come to that, why Word at all? In the late 1980s, we were running offices with no huge problems using text based Word Perfect and Lotus123. For most people most of the time, they worked perfectly well. They may well have been not only as good at actually getting work done -- which is, I think, most likely the point of PCs -- as their modern GUI equivalents. Maybe better.
"I for one will welcome our new AI overlords!"
Ahem ... How do you plan to do that? Looks to me like one very likely result of the overenthusiastic embrace of AI will be that effective communication. with or via computers will become next to impossible.
For example your "Hail Overlords" message will likely be massaged and come out on the overlord end as "What is the hail threat today in Overland Park, Kansas?" Or maybe as a challenge to take them on in a cage fight.
"Yet, in interviews, Elon will show just that by removing his hands from the steering wheel ..."
Not any more apparently. The OTA update will reportedly require hands on the wheel at all times.
BTW, how does one toggle turn signals with both hands on the wheel? Toes? Tounge? Brain implant? (Can that last be installed OTA)?
Am I the only one who thinks that just maybe a company whose driver assist software can't identify a bunch of brightly flashing lights as probably something that should not be driven through at speed, should NOT be doing over the air software updates? At least not without regulatory approval of the results of a well monitored test program.
Seems to me that there's a big difference between gratuitously breaking millions of printers (again) and messing with control of vehicles with a mass of two tonnes or more travelling at highway speeds. In the first case, people will eventually tire of having their workflows randomly broken, and they will find another vendor. In the second, the customer likely won't be around to complain. And neither will some of the folks who those lights were intended to protect.
Elon and other auto makers may not like the idea of being tighly regulated. But speaking as potential collateral damage, I think regulating these folks tightly is a just dandy idea.
This is hardly an every day issue. One problem one might expect that probably 97% of the members of the House have no idea what any of this is about. More important, the US House of Representatives is currently in a state of extreme dysfunction. A week ago,the majority party had a majority of 4 in a body with 455 members. But one of their members -- facing 23 federal felony charges -- was expelled. And the majority's deposed former leader just resigned effective at month end. (We won't have Kevin McCarthy to kick around any more). One might expect that a legislative body so evenly split would be a model of moderation. Au contraire, both parties are firmly controlled by idealogical extremists. The future? Pretty much anything might happen. Nothing whatsoever is a strong possibility
One doubts that this issue will be resolved on its merits. Indeed, it may not be resolved at all unless and until some debacle or other somehow forces action.
In the unlikely event the Supremes elect to hear the case, and if they find a prior restraint, what's the problem. Musk then gets to tweet whatever he wishes. And the SEC presumably is then free to whack him with appropriate fines if they judge his tweets to be attempts to manipulate the financial markets.
Sounds to me like Elon is fighting tooth and nail for the opportunity to screw himself. Have I got this wrong somehow?
We have a little used inkjet printer because despite its many flaws, the print quality is pretty good. And installing a color laser printer in linux seems invariably to be a nightmare. The inkjet seems, incredible as it may seem, the least bad choice if we wanted to print a few pages in color every now and then. What I've been doing for several years is printing a CUPS printer test page which includes a small multicolor circle every two weeks. That really does seem to keep the print heads from clogging.
If you can buy a "proper laptop" for the same money, by all means do so. But I doubt you can. Chromebooks -- especially used ones or older models --are really, really cheap. The internet tells me that Walmart has one new for under $50 US.
Are they good for anything? Out of the box, they seem to be OK for email, web surfing, and probably for most schoolwork. I assume that many of the apps in the Chrome Web Store actually work although I've never tried one.
BTW, Linux will run after a fashion on top of ChromeOS via Crouton or Crostini. There are purportedly some Linux distributions that can replace ChromeOS. Don't know how well they work.
The drawbacks -- the OS is nowhere near as easy to customize as Linux or even Windows. And Google is presumably spying on every action you take -- which may well be worse than Microsoft spying on your Windows PC.
For those unfamiliar with diabetes, the symptoms of of severe hypoglycemia are confusion and anxiety followed by coma and possibly death. The treatment -- Glucose. 15 or 20 grams. Perhaps followed up by more Glucose or some starchy food. The starch breaks down after a while into more Glucose.
From Honolulu to Vienna and all points in between lawyers are sharpening their pencils. And their knives. They are smiling. 2024 may be a very good year for lawyers.
2024 may not be such a good year for diabetics and for Insulet shareholders.
Google maps used to fuzz out critical locations. I discovered that a few years ago when I tried to check whether Jimmy Carter's home in Plains, GA had solar panels installed. (I'll be surprised if it doesn't). Couldn't tell. Looks like they've stopped doing that although the image of the White House on Google Maps (1600 Pennsylvania Ave NW in Washington DC) looks perhaps slightly fuzzy to me. Try it yourself.
For comparison, here's a link to an (eventually) declassified intelligence satellite photo posted on Twitter in 2019 courtesy of one Donald J Trump. https://www.npr.org/2022/11/18/1137474748/trump-tweeted-an-image-from-a-spy-satellite-declassified-document-shows
I should think that the current NK satellite is just a proof of concept. Not really a serious intelligence collection vehicle. Those will presumably come later.
I'm inclined to agree. A human would presumably have felt a bump and stopped. But maybe not that much sooner than the Cruise vehicle did. The article says 20 feet (about 6m). That's really not all that far.
And the internet tells me that a Chevy Bolt has a ground clearance of 5.6in (14cm). I don't think there's any way to get a car that low off the unfortunate pedestrian other than jacking the vehicle up and then dragging them out. I think that doing that probably is best left to professional emergency service folk. So maybe parking on the pedestrian is the best of a selection of bad choices.
Not that I think that Cruise or any other automated vehicle is currently smart enough to know that. BTW, How did the Cruise vehicle know to stop? It's probably not like it has a "human under the car" sensor.
"In fairness, if a human driver dragged a human body under their car the human driver would be prosecuted."
As I understand it, the Cruise vehicle didn't run down the pedestrian in the fashion you are probably visualizing. The pedestrian was struck by a car in an adjacent lane and thrown in front of the Cruise vehicle. Why do you think either the Cruise vehicle or a human driver would have enough information to understand the situation? I doubt that a human driver would be prosecuted in most cases like that unless they were obviously grossly negligent.
BTW, front proximity sensors might prevent/minimize some accidents like that. But most cars don't have them. And I get the impression that some cars that have them only activate them in "Parking Mode"
In fairness, I'm not sure that most human drivers would notice that they were dragging a human body under their car. And if they thought something seemed wrong without knowing exactly what, would they not stop (well, OK, not in NYC. But most places) -- likely on top of said human?
That said, if Cruise was not transparent about what happened, substantial fines and maybe other penalties would seem quite appropriate.
"Human artists are so much more creative and efficient than the giant energy sucking sound made by gen-AI ... why even bother![?]"
Because humans are not the NEXT BIG THING. We humans may be energy efficient, but we are yesterday's news. This is not an issue of efficiency, it seems to be an issue of venture capitalists and other parasitic lifeforms seeking big bucks without having to work too hard at acquiring them.
"WE are doing exactly that. ... Distill it. Presto: Fuel from the atmosphere that virtually any petrol/gasoline engine can be run on."
Sorry, I guess I should have addressed that in my post. But it would have made a rather long post. The major problem with your somewhat utopian scenario is distillation. Making fermented corn mash into 190 proof whiskey -- which is what corn ethanol pretty much is -- takes a lot of energy. In the US, that's mostly done with natural gas which is abundant and pretty cheap. Cheap in North America. Not necessarily elsewhere. There have been a number of studies that conclude that the energy budget for corn ethanol is slightly positive in the best growing areas and actually negative in less favored areas. Bottom line -- after one takes into account fertilization, mechanized field work, harvesting, and conversion to fuel, corn ethanol is a relatively inexpensive way to convert natural gas to a liquid fuel. That's not a bad thing. But a replacement for fossil fuel it is not. (BTW petroleum from Alberta Oil shales is pretty much the same story. How to the get the hydrocarbons out of the shale? Steam heated by Natural gas.)
Can ethanol be distilled more cheaply? Maybe. Solar distillation MIGHT work. But I suspect only in lower latitudes. Almost certainly not here in Vermont where corn is grown, but making a crop is problematic some years because of the short growing season. By the time you got your mash fermented, you'd be trying to use solar heat with an eight hour day, a maximum sun angle of about 25 degrees and solar collectors likely covered with ice or snow. The only study I've seen says solar distillation of ethanol is not economic even in the tropics, but I'm not sure that's the final word.
How about ethanol from sugar cane? Maybe. I've never seen an energy budget for that. Maybe it'll work. But Sugar cane is a tropical crop and it won't grow well even in some parts of the tropics. It needs a LOT of water.
How about biofuel crops that don't require distillation? Maybe. I tried once to figure out if it was possible to grow enough oil palms to support modern air travel. My conclusion. Maybe, probably not. Too many unknowns to be sure. BTW, environmentalists loathe palm oil. I'm not sure I disagree. Endless rows of oil palm trees are not my idea of living in harmony with nature.
My point remains. We do not currently have the technologies to feed 7 or 8 o 10 million people and fuel a modern civilization. It might be a good idea to have those technologies in hand before we run out of fossil fuels.
"We could make unnatural petrol with the CO2 from the atmosphere."
WE can do no such thing. At least not at any scale beyond that required to run a demo. Our great, great, ever so great grandkids might be able to do it. BUT ONLY IF "we" plug away at developing the technology and have it on the shelf when we start to run short of fossil fuels. My guess is that we will start to run short of fossil fuels sooner than anyone expects. Yes, the supply of stored hydrocarbons is vast. No one really knows how vast. But don't underestimate the vastness of the demand as billions of goat herders and subsistence farmers/workers in the developing world get richer and demand stuff like heat in winter, refrigeration, tractors, game consoles, vacations in distant lands, and other essentials of modern life.
"Even if they aren't currently buying ads, both companies still have Twitter / X accounts. Why do you suppose that is?"
1) In the case of IBM, they are likely cruising for an opportunity to sue X some reason. Sooner or later they'll find a patent violation. It's only a matter of patience and time.
2) Disney? Who knows? Maybe The Mouse has an untreated Twitter addiction.
"For me, any touchscreen is too much touchscreen."
I'm inclined to agree. However, I bought a Garmin SatNav device many years ago, that I still use from time to time. I usually mount it in the upper left corner of the windshield. It needs a tap to switch between displaying time_to_go and arrival time. My experience has been that it is far less annoying to view/use its touch controls while driving than touch controls on the center entertainment console. Maybe "heads-up" is really works better than down and right.
"Once EVs get old enough we will likely see more of this."
Possibly. Or not. Gasoline fires are relatively easy to start. Albeit nowhere near as much so as Hollywood would have you believe. They are also relatively easy to extinguish. Cut off access to oxygen and they will go out.
OTOH, Lion battery fires require heating at least one cell beyond 450C. Not so easy to do. Once started, they are likely to keep on burning. Cutting off Oxygen has little affect since the "fire" is fueled not by oxygen but by spontaneous discharge of the cell(s). The only(?) ways to put them out are to either break up the battery, isolate the burning cells and let them burn out. Or to somehow cool the whole mass of the battery below 450C.
My GUESS is that EV battery fires are/will be caused mostly by battery manufacturing defects or physical damage from collisions. And that they will decrease over time as battery designs improve.
In the meantime, I possibly wouldn't park an EV next to anything I valued. And I'd expect to pay a LOT for collision insurance because repair shops will likely refuse to repair any damage that might have damaged the battery. And to charge a lot for any work they do. Because THEIR insurance costs will likely go through the roof if they choose to work on EVs.
(NOTE: Not alll EVs are a LIon fire hazard. Older Prius hybrids used NiMH batteries that don't spontaneously "burn")
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
