HNHacker News
TopNewBestAskShowJobs

mysterypie

2,965 karma · joined September 19, 2015

submissionscomments
mysterypie··on Do not remain nameless to yourself (1966)
That’s publishing. I’m asking about one to one communication. Like the long personal letter Feynman wrote to his colleague. People seem to do much less of that.
mysterypie··on Do not remain nameless to yourself (1966)
Yeah, Marge Simpson too. It probably refers to a letter that the character wrote. But ignore those. Most of the list are real people.
mysterypie··on Do not remain nameless to yourself (1966)
I’d like to know what the modern substitute is for long personal letters like Feynman’s? It’s not just Feynman; look at this list from the featured website:

> written correspondence includes letters from: Jane Austen, Richard Burton, Helen Keller, Alan Turing, Albus Dumbledore, Eleanor Roosevelt, Henry James, Sylvia Plath, John Lennon, Gerald Durrell, Janis Joplin, Wolfgang Amadeus Mozart, Hunter S. Thompson, C. G. Jung, Katherine Mansfield, Marge Simpson, David Bowie, Dorothy Parker, Buckminster Fuller, Beatrix Potter, Che Guevara, Evelyn Waugh, Charlotte Brontë.

These are some important and busy people. Yet they wrote long personal letters. Whenever I write a long personal letter (by email), the most I get back is 3 or 4 sentences. And forget about having a real discussion by IM or WhatsApp — the replies are two-word sound bites. As far as I can tell, the only place real personal discussion takes place now is in person. Kind of ironic that we have far greater communication power than any of those letter writers did. I’m wondering if my experience is an outlier or if this is pretty much what everyone experiences today?

mysterypie··on It's time to take alternatives to dark matter seriously
> [Milgrom's alternative theory:] But far in the outlying areas of the Milky Way, stars would feel a smaller gravitational force than previously thought

Isn't it the opposite?

Dark matter is extra matter which implies extra gravitational force to hold the galaxies together. If you want an alternative to dark matter, shouldn't the alternative theory predict a bigger gravitational force than previously thought?

mysterypie··on Cryo-electron microscopy breaks a key barrier
So it’s not like what I thought at all. If I understand you correctly, then each one of those bright blue balls is a cloud of electrons at the center of which is a nucleus (but we don’t see the nucleus itself). The tiny blue lines and vertices are just computer generated imagery to define the “surface” of an atom. Those hexagons with six blue balls joined by purple mesh are benzene rings? So each purple area defines a cloud of shared electrons in a benzene ring?
mysterypie··on Cryo-electron microscopy breaks a key barrier
Could someone confirm what we’re looking at in that picture please? In that spider web of blue points joined by blue line segments: Those tiny blue points are atoms? The blue line segments are covalent bonds? And what’s the difference between the blue and purple areas?

If those line segments are probability distributions of shared electrons, that means our high school mental model of a covalent bond (drawing a line between two atoms) is not too bad.

mysterypie··on Want to build a side business? buy a great domain name
I agree with you that I'd prefer GoSojourn myself, and that BikeTours sounds spammy. Having now looked at both sites, they both look legit, and at first glance, roughly equivalent in presentation. At this point, I'd base my decision solely on prices, dates available, type of tour, customer service, etc. -- giving no weight to the domain name.

I can accept that buyers would pay more for BikeTours.com than for GoSojourn.com (and that the author would earn more money for brokering such a name), but is there any empirical evidence that customers prefer the simple descriptive name? And by how much? That's a critical question. If the BikeTours name costs $40,000 and GoSojourn costs $200, could the $39,800 difference be put to better use by GoSojourn such that it beats out BikeTours?

mysterypie··on Want to build a side business? buy a great domain name
The advice is to buy descriptive domain names and build a business basing your decision "solely off the domain name" that you're able to find. His About page claims that he himself does this. But does he actually do that?

Of the examples he gave – ComputerCamp.com, ShippingSupply.com, KobeBeef.com, CameraBag.com, Pacifiers.com, ZoysiaGrass.com, SouthernCookbook.com, Ziplines.com, CannedHam.com, SexyBastard.com – every one of them except ShippingSupply is a parked domain, spam site, or fails to resolve. Since the article is from January 2017, it seems like the average price of $5388 that people paid at auction for the 9 unproductive sites has been wasted for the last 3+ years.

As far as I can tell, his business is buying and selling domain names. (ShippingSupply, although it looks real enough, does not belong to the author; he writes "sell shipping supplies [is] exactly what this buyer did".) I can believe that the author has a profitable business doing what he does (buying & selling domains), but the artilce does not convince me that building a business around nice-sounding but arbitary domain names with no expertise in that business is a good idea. It doesn't seem like he's doing that.

mysterypie··on The unattributable “db8151dd” data breach
I have a large list of unique emails to test, but they are not from a domain I control. It seems that I can test these through the API, but is there any simpler way? I tried obvious things like putting a list of comma-separated email addresses in the search form, but it doesn't work.
mysterypie··on Recovered virus patients retest positive due to 'dead' virus fragments: experts
That’s great news; here’s the link:

https://questdirect.questdiagnostics.com/products/

However, I’m disappointed that it’s not available in the states of AK, HI, OK, AZ, IN, nor outside of the U.S. Further silliness is that in CA and OR you have to wait a week to get the results because California and Oregon state laws require that lab results be held for 7 days before Quest can release them; this hold is in place so that your physician has the opportunity to discuss your lab test results with you prior to you receiving them.

Another annoyance is the little questionnaire you have to go through before they agree to sell it to you. You have to divine their intentions for each question. If you don’t answer “correctly”, this is what you get and they won’t sell it to you:

This test is not right for you. Based on your response, you may still have an active COVID-19 infection and this test may not be right for you at this time. COVID-19 Immune Response testing is specifically used to check for an immune response to the virus which can take time after an active infection.

I understand all that. I just want to buy the test. Maybe I want to keep it for a day when I or someone I know needs it. Or whatever reason. Why force users to reverse engineer the questions and then lie to get it?

Also if you give your date of birth as less than 18, you get the same “This test is not right for you” refusal. I guess they don’t want to sell to minors, though the message is deceptive since it has to do with regulation and not whether the infection is active. So if you’re buying it to test your children, you again have to lie to their questionnaire since the questions are posed for the person taking the test.

The bureaucracy surrounding medical products is just horrible.

mysterypie··on Handbook of Covid-19 Prevention and Treatment from Hospital with 0% fatality [pdf]
I would welcome a handbook like this for individual or family use at home if hospitals become overwelmed and can no longer accommodate even seriously ill patients. The featured handbook looks very useful, but it's written for medical professionals in a hospital setting. I'm thinking about a handbook with instructions on how to do life critical procedures as best as possible with the resources one might have at home. Off the top of my head, one topic could be about the use of a bag valve mask[1] (as a "ventilator"), how to make one at home, when should it be used, things like that.

[1] https://en.wikipedia.org/wiki/Bag_valve_mask

mysterypie··on A chemical reaction in ancient Roman concrete makes it stronger over time (2017)
> We may think we're at the height of human knowledge, but the ancients did possess precious knowledge that has been lost to time.

I would love to hear why a chemical analysis doesn't solve the riddle here. Scientists were able to isolate a strain of the coronavirus and publish the entire genetic sequence in a matter of days, but we can't analyze a sample of concrete?

mysterypie··on Gates Foundation program will issue home testing kits for Covid 19 in Seattle
> somehow they continued paying the wrong provider

How does it happen that the wrong (old) provider still accepts money if they no longer cover you? Can this be used as a legal argument that the old provider is on the hook for coverage since they continued taking the money?

mysterypie··on Covid-19 now exists in two forms, indicating it's already mutated at least once
It is indeed very strange that Korea and Japan should have 400% more hospital beds per capita than the U.S., Canada, or Scandanavia. I tried to dig into the OECD methodology[0] to see if they mentioned anything noteworthy, but nothing grabbed my attention. Their notes about the U.S.[1] and Korea[2] seem pretty similar. I'm wondering if in Japan and Korea, people stay in hospitals for longer periods for the same condition than in the U.S. For example, in the U.S. mothers giving birth would stay in the hospital for a week or more in 1950s from what I heard, but these days they are discharged after a couple days. What's a typical stay for uncomplicated childbirth in Japan and Korea? If it's a week or 10 days, then that would give a possible datapoint for having many more beds.

[0] https://stats.oecd.org/index.aspx?DataSetCode=HEALTH_REAC

[1] Notes about the United States: Coverage: AHA-registered hospitals in the United States. U.S. hospitals located outside the United States are excluded. Includes all the AHA registered hospital beds for all types of hospitals. Estimates are for all AHA registered hospitals. AHA-registered hospitals include facilities such as short-term general, psychiatric hospitals, wards, rehabilitation institutes, maternity homes, tuberculosis hospitals, leprosariums and alcoholic treatment institutions. Estimates exclude U.S. associated areas such as Puerto Rico and AHA non-registered hospitals.

[1] Notes about the Korea: Coverage: Hospitals: all sorts of medical institutions equipped with wards of at least 30 beds. Excludes day-care beds, emergency beds, surgical tables, recovery trolleys, delivery beds and cots for normal neonates.

mysterypie··on Covid-19 now exists in two forms, indicating it's already mutated at least once
> L type is the more lethal type and the S type is the more silent type

Well that's easy to remember, L=lethal and S=silent. But I imagine there is a prosaic explanation for the haplotype naming. Anyone know what the L and S stand for? Googling hasn't turned up anything for me.

mysterypie··on Mozilla’s DNS over HTTPs
What are some reasons why someone would prefer to or need to disable it. Just curious.
mysterypie··on RCMP arrest 'money mules' tied to scam calls from India
How about putting a small tariff on every phone call originating from countries that do not shut down these fraudulent call centers? The tariff perhaps could be as small as 2 or 10 cents per call to make these operations unprofitable. I'm sure that the telecom operators in the foreign country could easily track down (or already know precisely where to find) the call centers. Once the fraud falls below a threshold the tariff can be removed.

Punitive tariffs, quotas, bans get applied to every kind of product -- counterfeits, contaminated foods, polluting vehicles. I never heard any government agency suggest applying it to countries that are the overwhelming source of scam calls.

mysterypie··on Dashlane's Super Bowl ad proves password managers have arrived
> Once you’re locked into one company, there’s not much incentive to switch. In fact, doing so can be a real hassle, since it requires resetting all those passwords all over again.

Why would it be a hassle? It should be trivial to switch password managers. Do some not allow you to export or import data?

This brings up another question: How can VCs justify $200 million in funding for businesses with essentially no customer lock-in?

mysterypie··on Verlan: French slang that inverses words
That seems similar to the way people say PIN number even though PIN itself means personal identification number. I don't know if au jour d'aujourd'hui sounds horribly redundant in French, but I find that PIN number is perfectly acceptable in English, along with SSN number, UPC code, and many others. It takes mental energy to decompose PIN into its components and realize it contains the word "number", so I welcome the redundancy.

When someone asks, "Do you remember your PIN?", I'm thinking about sewing needles or medallions for a second -- if they didn't add the word "number". On the other hand, the acronym ATM is so common that I don't feel the need to say "ATM machine" anymore; I'll say "Where's the nearest ATM?" and expect that most people will immediately understand it without having to think about what it stands for.

mysterypie··on Clinical features of patients infected with 2019 novel coronavirus
A 15% mortality rate is huge, but considering that this is a newly identified disease, the actual mortaility rate might be much less. Inventing some numbers, let's say that 3000 people were already infected when the initial 41 patients were identified -- those 41 being the sickest. The other 2959 people either didn't have symptoms or were assumed to have a cold or flu, and they all recovered. Had there been a reliable way to identify the hypothetical 3000 infected people, the mortaility rate would be only 0.2% if 6 had died.
mysterypie··on The Essential Guide to Electronics in Shenzhen (2016) [pdf]
> The US has been doing that for about 15 years

Except for US citizens and Canadians. Interestingly, Canada is the only country exempt from the US fingerprinting requirement, at least for tourism or business entry.

mysterypie··on The Essential Guide to Electronics in Shenzhen (2016) [pdf]
Is there any way to visit Shenzhen without submitting your fingerprints to the Chinese government? I've heard that everyone gets their fingers scanned at entry these days even if you never had to submit fingerprints with your visa application. Is that correct?
mysterypie··on Medieval Price List
Yes, I would love to see this in modern prices, no matter how roughly it gets converted. The use of pre-decimalized pounds makes it even more inscrutable. It would make that list a hundred times more interesting if we had some basis to compare with.
mysterypie··on The MLM Epidemic in 13 Maps
If that were the law, then the MLM's simple workaround would be to pay you a recruiting fee based on the success of the people you recruit. If the people you recruit sell a lot, you get a bigger fee because you did a better job at recruiting.

I imagine it'll be tough to craft a law to target MLMs because if it's too broad it'll hit many kinds of legitimate business and if too narrow they'll creatively bypass the rules.

mysterypie··on PayPal stops payouts to models on Pornhub
> In Germany every bank is obliged to offer everyone a so-called base account

That's very commendable; I wish more countries had such a regulation. Does it extend to non-citizens living in Germany? What about to visitors to Germany? Could someone open an account (by right) while there on a holiday?

mysterypie··on Peloton Is Spinning Faster Than Ever
> Peloton wound up opening at $27 a share and closed at $25.76 — not a bonanza day

From the company's perspective it's good that it didn't shoot up to $50 a share, isn't it? It means it was valued optimally; they squeezed the maximum value from the investors and didn't leave any money on the table.

mysterypie··on Someone is stealing Christopher Columbus letters
> Perhaps these artworks in their store have unknown owners

For unknown owners, fine. But they know where the Caravaggio came from. The article says, "Giovanni Prisco: Yes, and it is a paint stolen at the end of 1990 and was from a private house. It was discovered in the north of Italy."

I'm reading that as saying that it was stolen from someone's private house and that it was recovered somewhere else in the north of Italy.

mysterypie··on Someone is stealing Christopher Columbus letters
> One of the most valuable art collections in Europe isn't housed in a museum visited by millions of tourists, but in this warehouse in the back of his police station. It's basically an evidence locker for stolen artifacts. This is a Caravaggio. What does a Caravaggio go for these days? It is millions of dollars. It is a paint stolen at the end of 1990 and was from a private house.

What's not explained is why are police keeping all this artwork? If you saw the 60 Minutes piece on this you'd see that it's an enormous evidence room full of artworks they are hoarding. They know where the multi-million dollar Caravaggio was stolen from in 1990. Why hasn't at least this particular piece been returned to its owner?

I also wonder why such an obvious question isn't answered in the story. I'm afraid that the answer might be that it doesn't fit the agenda of the story about sophisticated art thieves and super rich collectors. The reason might be that the Italian police declare that your art is a national treasure and they simply keep it. Or they impose an impossible burden of proof that you acquired it legally (similar to U.S. asset forfeiture laws) and require you to show a chain of receipts back to Caravaggio in 1602.

mysterypie··on Dutch family 'waiting for end of time' discovered in basement
On a lark, I went to try and find this house on Google's satellite view. The article says that the "farm is outside the village of Ruinerwold and is accessible by a bridge over a canal." I discovered that there are lots of bridges, lots of canals, thousands of farms within a 20km radius of the town. We need an aerial image search tool that figures out where you took that snapshot.
mysterypie··on “I was told to buy a software or lose my computer: I ignored it”
> With a U.S. population of ~200M, let’s call it 4M infections/year. We know that 4% of respondents paid the ransom, so that’s 160,000 ransom payments. If we say $200 per paid ransom as a guess, that means the ransomware ‘industry’ is making about $32M a year.

The U.S. population is actually ~327M, but since we're talking ballpark figures, let's just say that ~200M represents the adult computer-using population. It strikes that ransomware is both much bigger and much smaller than I thought. The 160,000 ransom payments is much bigger than I would have guessed. It's not like being struck by lightning; it's more like a serious car accident. But the actual revenue generated, $32M as he's estimated, is pitiful. There are all sorts of illegal, quasi-legal, and legal ripoffs that generates billions per year. For the amount of attention ransomware gets, it's just noise.

← PreviousPage 4 of 16Next →