Some progress toward Navajo/Chilean prices for our National Parks

by philg

Loyal readers may recall What if our National Parks charged Navajo prices? (2023)

$100 per person per day is the “Navajo rate” for what could reasonably be charged … the Chileans charge foreigners $35 per adult to visit their signature national park for one day. Even at Chilean prices it would seem that the NPS could easily be self-funded.

“Department of the Interior Announces Modernized, More Affordable National Park Access” (yesterday):

Beginning Jan. 1, 2026, the Annual Pass will cost $80 for U.S. residents and $250 for nonresidents, ensuring that American taxpayers who already support the National Park System receive the greatest benefit. Nonresidents without an annual pass will pay a $100 per person fee to enter 11 of the most visited national parks, in addition to the standard entrance fee.

Orwell fans will appreciate the contrast between headline (“more affordable”) and body (“$100 per person extra”). Also, nobody questions that “American taxpayers [SHOULD] already support the National Park System”. Why does a working class American who can’t afford the epic costs of airline tickets, rental car, hotels, etc. have to pay taxes to subsidize rich people from around the world who can afford the $1,200/day cost of a hotel-based family National Parks trip? (I estimated $1,000/day in 2023, but airline ticket, restaurant, and hotel prices have gone up significantly since then.) Separately, if the NPS funds itself via entry fees it won’t have to turn people away during the inevitable government shutdowns.

I can’t understand how the new park entry pricing system will work. Americans aren’t required to carry passports. Tens of millions of residents of the U.S. have no documents at all (22 million as of 2016, according to Yale). How is a gate agent at a National Park supposed to determine if a visitor is a U.S. resident? We’re informed that it is racist to demand ID for voting. Could a National Park demand to see a state-issued driver’s license or other ID before offering the “resident discount” rate? We’re informed by CNN that “Outdoor recreation has historically excluded people of color” and “racist laws and customs kept Black Americans out of these parks”. Surely our government wouldn’t want to intensify the racism inherent in the racist National Parks by demanding ID from visitors of color?

Loosely related, a couple of photos from the Schoodic Peninsula, an often forgotten piece of Acadia National Park. As with the core portion of Acadia, the land was donated to the American People. The Rockefellers donated the island land and Schoodic was donated anonymously in 2015. This reminds me to note the tragedy of Bill Gates giving all of his money to Africa, which doesn’t seem to help average Africans (every year that the Gates Foundation has operated in Africa, the number of needy Africans has increased; maybe some rich people in Africa have gotten richer?). If Gates had to sell the Microsoft stock and pay capital gains before shipping the proceeds to Africa, the tax revenue would easily fund an additional national park. Alternatively, if he spent his money on unspoiled U.S. land he would easily be able to create five new national parks.

Full post, including comments

Open borders don’t lower wages, but sending migrants home will raise wages

by philg

Frontiers of Migranomics from one of our intellectual elites, a New Yorker writer:

We’ve been informed, as a matter of Scientific Fact, that low-skill immigration does not reduce wages for the American working class (contrary to Harvard economists’ analysis). Now the same Scientists are telling us that employers will be forced to pay higher wages, e.g., to apartment cleaners and roofers, if low-skill migrants are sent back to their home countries. More immigrants caused wages to rise (the undocumented built the current American economy) and, also, a reduction in immigrant supply would cause wages to rise.

This reminds me of Immigrants expand our economy, but millions of immigrants exiting the U.S. don’t shrink our economy.

Separately, I’ve refined my Is U.S. immigration policy a form of animal hoarding? post into a more succinct form (without even trying AI!):

The passion for low-skill immigration has the same rational basis as keeping 100 cats in a 2BR apartment: “Animal hoarding is an accumulation of animals that has overwhelmed a person’s ability to provide minimum standards of care. … Rescue hoarders believe they’re the only people that can adequately care for their animals.” The same people who say that the U.S. has a dire shortage of affordable housing and health care then say that the 70 million migrants we’ve welcomed in recent decades aren’t sufficient and we need to bring in more migrants.

My new standard response on X, featuring photos from Unlimited Car Wash in Palm Beach Gardens, Florida, November 21, 2025:

Without 70 million immigrants and their children (another 50 million?) who will hand-wash and vacuum my Rolls-Royce for $21?

In case the Jill Filipovic tweet is memory-holed:

Full post, including comments

Hand signals for Cybertruck in Maskachusetts

by philg

From a friend in Maskachusetts who owns a stainless steel monstrosity:

[college-age son] got middle finger in truck yesterday going through Boston tunnel, and today in Hanover. He has been doing Nazi salute back.

Meanwhile, at a strip mall here in Florida, an illustration of the size range of vehicles that Americans typically use to transport a single human:

Finally, my friend provided an update on Tesla’s full self-driving:

I use FSD more and more. [wife] wants me to pay for it for both cars. Thinks it makes [son] a safer driver.

(He has an old Tesla 3 and a new Cybertruck.)

If you thought that the Cybertruck wasn’t wide enough… (photo from a nearby neighborhood here in Jupiter, Florida)

Full post, including comments

How about an AI math tutor that looks at paper and pencil?

by philg

The New York Times, which told us that closing schools for 18 months was the absolute best thing for children (keep them safe from a virus that was killing Americans at a median age of 82), now tells us that screens are bad… “The Screen That Ate Your Child’s Education”:

Many of these devices are provided by schools. You might think that these school-issued devices allow only a limited number of functions, like access to classroom Canvas pages and Google Docs. If you assumed that, you would be wrong.

Sylvie McNamara, a parent of a ninth grader in Washington, D.C., wrote in Washingtonian magazine that her son was spending every class period watching TV shows and playing games on his school-issued laptop. He often had no idea what topics his classes were covering. When she asked school administrators to restrict her son’s use of the laptop, they resisted, saying the device was integral to the curriculum.

In a survey of American teenagers by the nonprofit Common Sense Media, one-fourth admitted they had seen pornographic content during the school day. Almost half of that group saw it on a school-issued device. Students watching porn in class doesn’t just affect the students themselves — picture being a teenager in math class trying to concentrate on sine and cosine while sitting behind that display of flesh. It is disturbing on a number of levels.

(Teenagers are spending 80 percent of their in-class time watching porn and then just wasting the rest of the school day?)

Based on looking over the shoulders of our 4th and 6th graders, electronic math homework is the worst idea that I’ve seen. Each problem is multiple choice. The child can click on A. The software says “Wrong”. The child can then click on B. The software says “Wrong”. The child can then select C. The software says “Right” and proceeds to the next problem. Neither teacher nor parent is notified that the homework was apparently completed via guessing. Then the test comes along and the child who got 100% on homework might receive a grade of 25-40% because the test doesn’t allow for correction of wrong guesses.

What if an AI could work like a human math tutor? My dream is a household with cameras everywhere so that an artificial intelligence can tell me where to find scissors, tape, the coffee cup that I set down 15 minutes ago, etc. (see Why doesn’t ChatGPT tell us where to find items in our houses?). Given those already-installed cameras, an AI can watch a young scholar working on pencil and paper and say “That answer isn’t quite right”, then explain where the child went off track.

Optional enhancement: eye-safe laser pointer on a gimbal so that the robot can point to a place on the physical page while talking about what went wrong and what the learner should do.

The closest existing product to the above is Photomath, I think. You can give it a handwritten math problem and it explains how to solve it. I don’t think that is what most learners need, though. They already sat through the teacher telling them how to solve the same kind of problem (maybe while simultaneously streaming porn?). Also, instead of getting kids off the screen it forces kids onto the screen to use Photomath. The above-described system would be 100 percent audio-based from the learner’s point of view.

Should this be called “MathGPT”? Of course that name is already taken. The product seems to be a way to get learners to spend more time on screens:

Full post, including comments

Replacement theory is false: Maskachusetts public schools edition

by philg

Boston Globe:

Massachusetts public school enrollment has long been stagnant, with international arrivals balancing out a low birth rate and high numbers of residents moving out of state.

Who benefits when public schools are jammed with undocumented migrants?

The drop in the number of students means less funding for the districts. State aid formulas provide every district with at least $150 per student, but some districts receive more than $17,000 per child on average because they serve a high number of children who require additional resources for being low-income and English learners, and do not have as much local revenue. Any cost savings from the reduced number of students tends to be far less than the amount of lost funding for those districts.

(i.e., the government workers are able to turn a profit on the migrants)

My question to ChatGPT:

Here’s a Boston Globe article that says “Massachusetts public school enrollment has long been stagnant, with international arrivals balancing out a low birth rate and high numbers of residents moving out of state.” Is this an example of native-born Americans being replaced by immigrants? https://archive.is/zbSbM

Answer:

That’s a thoughtful question. The short answer is: no, the Boston Globe article’s statement alone does not support the claim that native-born Americans are being “replaced” by immigrants. But it does reflect important demographic dynamics worth unpacking. Let’s go through what the article says, what it doesn’t, and how to interpret it.

What about Grok?

No, the quote from the Boston Globe article does not describe an example of native-born Americans being “replaced” by immigrants. Instead, it highlights a demographic balancing act in Massachusetts public school enrollment

The Great Replacement conspiracy theory remains #Scientifically proven false.

Related, a DHS.gov story about a 13-year-old detained after school:

On October 9, 2025, the Everett[, MA] Police Department arrested a 13-year-old alien from Brazil on dangerous weapons charges for allegedly possessing a firearm and a 5-7-inch knife.

The teen is mentioned 11 prior police complaints filed by Everett PD for a laundry list of criminal behavior, including ‘flash mob’ style shoplifting, consuming alcohol underage, breaking and entering, vandalism, theft, fighting and more.

The same student is featured as mostly peaceful by CNN… “Her 13-year-old son was arrested, then taken by ICE to a detention facility”:

“They didn’t give me any information,” said [Josiele] Berto[, speaking in Portuguese], who is from Brazil and along with her family have had a pending asylum application since arriving in the United States in 2021. “I asked where [Arthur Berto] was being taken, and they said they weren’t allowed to say.”

Everett Mayor Carlo DeMaria said in a news conference a teenage boy – whom he declined to name because he is a juvenile – was arrested last week after Everett Police received a “credible tip” accusing him of making “a violent threat against another boy within our public school.”

Here’s the mischievous tyke who won’t be attending Maskachusetts public schools at least for the next few days:

Full post, including comments

Audioengine HD4 desktop Bluetooth speaker review

by philg

I called in an airstrike on my own audio position by “upgrading” from a 10.5-year Windows 10 PC to a brand-new Windows 11 machine with 100% pimp ASUS ProArt Creator motherboard. If I’d spent half as much on a motherboard from ASR the machine would have had an optical S/PDIF audio output compatible with my old Nuforce Dia amp (mighty 18 watts) and Audioengine P4 passive speakers (both purchased in 2012 and worked without failures for 13 years). The cheap ASUS motherboards seem to have a header for S/PDIF even if there is no connector.

I decided to give the P4 speakers a vacation and purchased an Audioengine HD4 Bluetooth speaker system. They’re about the same size as the P4 speakers so I put them on the same stands. The result is less desktop clutter because the Nuforce Dia is gone. The Nuforce Dia’s power supply is gone (the HD4’s power supply is internal). One of two speaker wires is gone (the powered HD4 on the left still needs a speaker wire, included (with banana plugs!), to send the output of its power amp to its passive brother/sister/binary-resister on the right). The cable connecting the PC to the amp is gone. (Note that if you’re a serious audio nerd you might nonetheless need to reintroduce a USC-C cable from the PC to the Audioengine HD4; the digital-to-analog converter in the HD4 is capable of handling 96 kHZ/24 bits, but Bluetooth aptX HD is limited to 44 kHz/24 bits. One thing that is painful about my ASUS motherboard is that it doesn’t have any standard connector for a Bluetooth antenna. It has a proprietary pair of connectors for a combined WiFi/Bluetooth antenna that is huge and connected by a long ugly cable to the back of the PC. Given that my PC is hard-wired to the switch via a Cat 5 wire that the 2003 builder of this house thoughtfully included, I just need a small Bluetooth antenna that will live on the back of the motherboard. This apparently does not exist in the ASUS universe.

Setup took about 2 minutes. I powered the HD4 off and then on after 5 seconds to simulate a brief power failure. The Windows 11 machine reconnected automatically. Sound quality seems similar to what I was enjoying before. So… my stupidity in assuming that every modern motherboard would have an S/PDIF optical audio output resulted in the recovery of a bit of desktop space at a $329 cost (on sale from the usual $429 price).

Unlike Sonos, Texas-based Audioengine suggests via its photos that white people may purchase and use its products. Here’s a person at serious risk of “tech neck” unless the AI revolution renders the job obsolete.

The one thing that I don’t love about the speakers aesthetically, compared to the P4, is the metal strip across the front. I guess it would be pretty tough to design a wooden volume knob and a wooden headphone jack!

This photo shows the speakers with the Bluetooth antenna pointing up, which was completely unnecessary in my setup. It also shows the old-school RCA inputs and outputs. The RCA output can be used for a subwoofer. I don’t think that the HD4 has a crossover network and, therefore, the HD4 would keep getting driven at full range even with a subwoofer hooked up. Audioengine seems to include a low-pass filter in their subwoofers so that maybe it all works out, but I’m not gaming in the home office nor watching Hollywood action movies so I don’t think I will be trying out the subwoofer config.

Conclusion: this thing works, but it probably would have been smarter to buy a motherboard with S/PDIF optical out! Also probably smarter to buy a motherboard with a standard antenna connector to which a short Bluetooth antenna could be attached.

Full post, including comments

Use normally dead/black televisions as virtual windows into interesting places via webcams?

by philg

Readers may recall my passion for doing something with the dead/black/huge televisions that are on many walls of our houses (the house that we bought in Florida actually came with six flat-screen TVs at no extra charge because the previous owners didn’t feel like demounting and moving them), e.g.,

I’m surprised that nobody has implemented a business idea that I proposed to entrepreneurial friends about 10 years ago: a streaming service that turns any television into a “virtual window”, but not a window onto the boring street where one actually lives. A subscriber could choose to be looking out at the Champs Elysée, at the crazy intersection in Shibuya (Tokyo), a lake/mountain view from a famous resort hotel, etc. Since all of my ideas are terrible, from a business point of view, the original concept was a cable TV channel. Cable companies offer roughly 50 music channels for ambient use. Why not 50 virtual windows as well?

High quality webcams have only gotten cheaper in the decade since I proposed this idea. Internet has become faster and more reliable. Why hasn’t this idea caught on?

There is a construction documentation company that branched out into this market a little and offers earthcamtv.com, which seems to be supported by low-rent ads rather than subscription. They have an Android TV app so I guess it would work on a Sony, TCL, or HiSense . I don’t think anyone would want this running continuously in his/her/zir/their house.

Here’s a newer twist on the idea: Immigration TV. This could have virtual windows into the countries that enrich us, e.g., Venezuela, Haiti, Colombia, India, Pakistan, etc. It could be sponsored by both the Democratic Party (channels that show how great life is in places that migrants claim are too dangerous to inhabit) and the Republicans (channels that show the crowded, dirty, and disorganized conditions that people in source countries have created for themselves).

As far as I know, all current TVs lack the interface required to be programmed to “wake up at 0900 and start up the Virtual Windows app” so it would be somewhat tedious to go around to every TV in the house every morning and configure this.

Samsung is still trying to sell people on its absurdly deficient The Frame system (requires an external box that nobody has a place to put except maybe if a house was built from scratch with The Frame in mind; they make a wireless version of the box, but of course everyone says that it doesn’t stay connected). Most humans are much more drawn to moving pictures than to still images, even still images of great art (art museums that are free still struggle to attract a wide audience). Why wouldn’t LG introduce The Window in which the television comes preloaded with the ability to show streams from curated webcams around the world?

Partial personal list of desired virtual windows:

  • One for each of the nicest Japanese gardens in Japan (that would be around 50 window choices?)
  • One for the bonsai collection with pond behind at Morikami Japanese garden in Palm Beach County, Florida (good for the Japanese winter months)
  • Ngorongoro Crater (Tanzania)
  • Churchill, Manitoba (polar bars)
  • Piazza San Marco, Venice (from a second-story window since nobody needs to see the pigeons up close)
  • Miradouro de São Pedro de Alcântara, Lisbon
  • Portofino South condo rooftop in West Palm Beach (looks out towards Mar-a-Lago so we can keep our envy levels appropriately high)
  • Miami waterfront skyscraper (any) looking out toward Biscayne Bay and Miami Beach (watch the cruise ships come and go)
  • Looking out on the main square of Santiago de Compostela to watch pilgrims who’ve completed their walks
Full post, including comments

Moen Flo Artificial Intelligence Water Overlord

by philg

Everyone in Florida prepares for the Category 5 hurricane that, generally, never shows up (e.g., Tampa hasn’t been hit badly by a hurricane for 100 years). Hardly anyone prepares for an internal water leak and quite a few of our neighbors have suffered severe home damage from, e.g., burst sink faucet supply lines. All of the houses in our neighborhood are about 23 years old and that’s apparently a great age for a massive water escape.

I debated and dithered between the Moen Flo and the Phyn Plus. The vulnerability of the Moen Flo is a mechanical impeller that reportedly fails after 1-3 years, but my friend decided that the Phyn Plus would impose more restriction on water pressure (maybe due to 3/4″ internal pipe where our supply is 1″?). Our plumber has a Flo so we decided to go with Moen (the evil Kohler empire rebrands Phyn Plus). The Phyn Plus makes some stronger claims for intelligence.

About one month after the Flo went in we had our first leak, a steadily dripping Roman tub faucet. Here’s an excerpt of the Flo’s recorded usage:

What’s remarkable about the above is that the AI Water Overlord decided that it was a perfectly normal usage pattern for a human to stand at a faucet for several days and pull out 0-0.01 gallons per minute. No alerts were issued. I did an AI support session with the app, which told me to go onto the web site and see if I could do a “MicroLeak” test there. The app wouldn’t let me do it maybe because it thought that a faucet was legitimately in use. In any case, the history showed a few MicroLeak tests passed with flying colors during an obvious and steady leak. Through all of this, the app displayed a “fat/dumb/happy” screen:

I eventually did call the Moen support number. The phone line was quickly answered by an American with no accent. He had no explanation for why the device hadn’t raised an alert, at least, and decided to escalate the question to Tier 2. Meanwhile, although I’d purchased a bundle of the device and three years of extended warranty and other support (assuming that the impeller will fail for us as it has for everyone else), the Web application showed me as not subscribed to “FloProtect” and kept trying to sell me “FloProtect”. The customer support agent confirmed that I actually WAS subscribed.

The leak was bad enough that if had happened while we were away for a few weeks it could have caused $50,000 of mold and other damage. Fortunately for us, however, it was leaking into a tub with a drain and we observed it with our own eyes.

The second fun part of this story was that I had previously asked a contractor to sort out the access panel situation for the Roman tub faucet and the shutoff valves. These were mostly tiled in so a plumber couldn’t service them. He converted two of the tiles to a magnetic mount. “Just cut the grout with a utility knife,” said the contractor, “and then maybe use a suction cup to get the tile out.” Of course, this was a huge challenge for me and I didn’t succeed without a trip to Home Depot to get a pro-grade suction cup grabber.

When I did get my unskilled paws on the 23-year-old shutoff valves I was able to turn them all the way to the right. This shut off the cold water to the tub, but the hot water still flowed out at a substantial rate and continued to drip (apparently, it was the hot water that was dripping). I torqued it down some more and finally got the pressure low enough that the dripping stopped with the faucet tap turned off. ChatGPT says that failure of old shutoff valves is common:

Most angle stops or multi-turn shutoff valves use a rubber washer or packing that seals against a brass seat. Over time, the rubber hardens, cracks, or disintegrates. When you turn the handle clockwise, the valve stem no longer presses tightly enough to fully stop flow — so you get a partial seal and reduced flow instead of a full shutoff.

So, as part of the joy of homeownership, in one day we had

  • a leak detector that wouldn’t detect leaks
  • an access panel that couldn’t be accessed
  • a shutoff valve that wouldn’t shut off

I’ve been wanting to replace the tub filler for a couple of years but I can never wrap my head around how to find something that will fit the hole pattern that we have. Nobody has ever been able to figure out what kind of faucet this is. There is no manufacturer’s name or logo above or below the tub deck. It seems to all be high quality stuff, but there wouldn’t be any way to replace a cartridge or the trim (finish badly marred maybe by some previous owner’s cleaning attempt?) since we can’t figure out where it might have come from.

Maybe the goal should be to have it all done by National Fix a Leak Week (March 16-22, 2026 unless there is another government shutdown).

The actual leak:

Did the Flo ever shut off the water for any reason? Yes! The app had been told that we have a pool and that the pool lacks an auto-filler and it had seen multiple episodes of the pool being topped off by hose before. After I turned up the sensitivity following the failure to detect the above drip, the Flo shut off the water after about one hour/1,000 gallons. So I guess if we ever do have a leak in the house, the Flo will prevent more than 1,000 gallons from covering the floors?

Update: Mere hours after this post went live, the Tier 2 folks at Moen called me back. I learned that it does the MicroLeak test by shutting off the water in the middle of the night when, in theory, nobody would want to use it. Then it spends 1-4 minutes watching for a pressure drop that could be caused by a leak. This effect can be masked however, by traditional hot water heaters and their expansion tanks. It’s possible that the layout of our house and the expansion tank would prevent any Flo device from noticing a leak on the hot side (which I think is what we had). The Tier 2 folks watched the above video and said, however, that our Flo should have noticed that level of leakage. So they’re sending me new hardware, but with the caveat that it probably won’t fix the issue. I’m planning to test it by introducing a fake leak. At the same time, I learned that our plumber should probably have put in a new “consumer water shutoff” valve before the Flo. That would enable us to easily replace the Flo without having to go to the street shutoff.

Full post, including comments

Getting a 20-year head start on ridiculing Larry Summers

by philg

Larry Summers is in the news lately for his association with Emmanuel Goldstein: “Larry Summers steps back from public commitments, ‘deeply ashamed’ by Epstein revelations” (Politico).

Was there anyone ridiculing him 20 years ago? Yes! See “Women in Science” from this very server:

Summers was deservedly castigated [fired from Harvard for musings regarding why there weren’t more mathematicians and physicists identifying as “women”] , but not for the right reasons. He claimed to be giving a comprehensive list of reasons why there weren’t more women reaching the top jobs in the sciences. Yet Summers, an economist, left one out: Adjusted for IQ, quantitative skills, and working hours, jobs in science are the lowest paid in the United States. This article explores this fourth possible explanation for the dearth of women in science: They found better jobs.

Is “early ridiculer” credit a thing?

Separately, remember that Larry Summers and Claudine Gay have nothing left of their careers other than lifetime jobs and monthly paychecks that will keep flowing regardless of how senile and incompetent they become.

Full post, including comments