Hacker News new | past | comments | ask | show | jobs | submit | escape_goat's comments login

I think that it was good that it stood out to be honest.

Back here on the lay benches, I think the best starting point in the Wikipedia is probably the the article on inhomogenous cosmology, of which the Timescape Comsology proposed by David Wiltshire (listed as an author on this paper) in 2007 is a notable example; it is discussed in the article.

<https://en.wikipedia.org/wiki/Inhomogeneous_cosmology>


This is a mind-blowing theory!


It's not joining it in a kerning sense, that's just the remarkably serif nature of EB Garamond, which has a little teardrop terminal on the tip of the 'c'. It's possible that you have font smoothing that is tainting the gap, otherwise it's your eyes.


I was curious about this as well, it looks as though he’s using a specific font which creates a ligature between those letters. I think it’s specific because it’s only on the CT and it’s on other pages in his site. I went further to investigate what this might be and it’s a little used print style: https://english.stackexchange.com/questions/591499/what-is-t...


No, the heading font is Lato, not Garamond, and it's definitely some kind of digraph that only shows up with the combination "ct". Compare the letter "c" in these two headings: https://i.imgur.com/Zq53gTd.png


This should be upvoted. Thank you, I hadn't realized that OP was referring to the heading font or scrolled down to see what is yes, quite a remarkable ligature. It appears to be Lato delivered from <https://brick.freetls.fastly.net/fonts/lato/700.woff> The ligature appears due to discretionary ligatures being turned on.

     h1, h2, h3 {
     font-feature-settings: "kern" 1, "liga" 1, "pnum" 1, "tnum" 0, "onum" 1, "lnum" 0, "dlig" 1;
      font-variant-ligatures: discretionary-ligatures;
     }


Actually, EB Garamond has c_t and s_t ligatures.


It does, but those would only be applied if the `font-variant-ligatures: historical-ligatures` property were specified, so they don't appear on this site.


I inspected for a ligature and any evidence of CSS kerning being turned on before commenting, but I didn't test it to see what the page looked like with it turned on, so I didn't have active knowledge of the possibility of a ligature. If I'd know, it would have been better to give wider scope to the possibility that somehow kerning was being activated by OP's browser. I should have known better than to make a remark about a font without absolutely scrupulous precision! I actually appreciate the comments and corrections.


and a very subtle f_f. I don't find those nice though.


I asked ChatGPT-4 mini the previous comment's question (""Which is heavier, a 9.99-pound back of steel ingots or a 10.01 bag of fluffy cotton?") in new incognito tabs until I got an incorrect result. It only took four tries.

1: "The 10.01-pound bag of fluffy cotton would be heavier. Even though cotton is much less dense than steel, the weight of the cotton bag is 10.01 pounds, while the steel ingot bag weighs 9.99 pounds, making the cotton bag the heavier of the two by 0.02 pounds."

2: "The 10.01-pound bag of fluffy cotton is heavier. Even though the steel ingots are denser and more compact, the cotton weighs more at 10.01 pounds compared to the steel ingots' 9.99 pounds. The difference is only 0.02 pounds, but in terms of weight, the cotton bag is heavier."

3: "The 10.01-pound bag of fluffy cotton is heavier. Despite cotton being less dense than steel, the weight is determined by the total mass, not the material's density. So, since the cotton bag weighs 10.01 pounds and the steel ingot bag weighs 9.99 pounds, the cotton bag is heavier by 0.02 pounds."

4: "The 9.99-pound bag of steel ingots is heavier. It weighs 9.99 pounds, while the bag of fluffy cotton weighs 10.01 pounds. Despite the cotton being fluffier and taking up more space, the steel ingots are still lighter by 0.02 pounds."


This sounds like it is de-facto true but it isn't. A private file repository with a closed source license is in-facto different than a private file repository with an open source license. The license is not altered by the access permissions of the repository, and the access permissions of the file repository are not altered by the license. A compiled version of code in a private repository with a closed source license can be released publicly without that code. The same is not true if the private repository contains code with an open source license.


I don't know how you can still be so confused about what de-facto means. But you are.


I'm not confused, I'm pointing out that the license in a private source repository changes the legal rights of the person with respect to distribution of the software defined by the contents of the repository, and than therefore a private repository with open source code inside is not a de facto closed source repository.


> A private file repository with a closed source license is in-facto different than a private file repository with an open source license.

Yes, and "in-facto" is not the same as "de-facto".

De-facto means "describes practices that exist in reality, regardless of whether they are officially recognized by laws or other formal norms". It is not a precise term.

I have some code on my server that I wrote with an MIT license. You have no way of accessing my server, and the code has not been made available elsewhere. You can't even know that the code exists, much less that it has an MIT license. To you, it is for practical purposes the same as closed-source software, in the sense that you cannot obtain the source code of the software, despite it having an open source license. This is an applicable situation for "de facto" by its common meaning.

"Open source" is not even a precisely defined term, although people who are persnickety about the definition refer to the OSI definition, which includes "Where some form of a product is not distributed with source code, there must be a well-publicized means of obtaining the source code". By this definition, code with the MIT license held secretly in a private repo is not open source. It became open source when Roman made it public, as was his right to do by the license. The license is one component of what makes software open source, but there are other components as well.


In your view, what would be an example of code that is de jure open source but de facto closed, if not a case where people have the legal ability but not the technical means to access it?


I'm rejecting the idea that software distributed in any way that has an MIT license can be "de facto" closed source. I think that this is a conceptual misapplication of the term "de facto," which implies an argument of equivalence in a situation where that equivalence does not exist. The question is not whether or not we can access the source code; we can now access the source code. The question is what the license was. To say that the license was "de facto" closed source before the software was distributed without the repository controller's knowledge or intent is to make a legal argument about that license. That argument was false. It just just as false as the argument that an electrical fire in the wall of your house that you could not see occurred 'before' your house was on file.


Calling it defacto closed while it was physically inaccessible, is exactly not making any legal argument. That's the whole point of the term is to say "it's not really closed legally, it just has the same effect as being closed, because for some other reason besides it's license, no one can have it.

There are these books that explain the meanings of terms. We don't have to guess and have random individual ideas about what words mean, and then wonder why no one can communicate.


Just to clarify again, this is a dry bulk / Panamax vessel. It is part of the shipping industry. At scale, it is analogous to a railroad car. In 2015 it was operating as the Avra under the flag of Greece. The foreknowledge of the Chinese government that a Russian officer would conduct hybrid operations from the vessel cannot be inferred from the circumstance. It is like thinking that someone with an American passport is an American spy.


It is quite the opposite from what you are arguing. China is responsible for the conduct of the vessels they allowed to fly their flag.

They can later claim that the crew and captain acted on their own will, without orders from the Chinese leadership. They can duly punish the captain and crew or disavow the vessel and declare them renegade, disallow them to fly their flag. But without such a declaration, a nation such as China is responsible for the conduct of their fleets, be they civilian or military. And any vessel they allow to fly their flag is part of their (in this case civilian) fleet.


Well sure we can both make unsourced assertions all day but as far as I can see the flag state is responsible for illegal conduct of commercial vessels only insofar it has failed to meet its obligations for regulatory and legal oversight.


Well, yes. But I'd claim that having a vessel intentionally damage foreign property and then ignoring the issue and not exercising legal oversight by at least investigating what happened is such a failure in obligations.


Is the US responsible for any crime committed by members of ships that fly the star-spangled banner?


To some extent, yes. US law applies on US-flagged ships.

https://en.wikipedia.org/wiki/Flag_of_convenience

> A ship's flag state exercises regulatory control over the vessel and is required to inspect it regularly, certify the ship's equipment and crew, and issue safety and pollution prevention documents.

Because US law is strong in this regard, the US military is by far the largest contributor to the count. Less than 200 civilian vessels are flagged in the US; https://www.statista.com/statistics/652126/us-flag-oceangoin....


There was a degree of proof of work involved. Text took human effort to create, and this roughly constrained the quantity and quality of misinforming text to the number of humans with motive to expend sufficient effort to misinform. Now superficially indistinguishable text can be created by an investment in flops, which are fungible. This means that the constraint on the amount of misinforming text instead scales with whatever money is resourced to the task of generating misinforming text. If misinforming text can generate value for someone that can be translated back into money, the generation of misinforming text can be scaled to saturation and full extraction of that value.


I do not think it true that it goes "both ways." Patients do not have a collective responsibility to doctors. Patients should not go untreated because other patients lie about their addictions. Being a doctor means exactly watching dozens of horses go in and out the door each day and saying "that's a zebra" before the zebra makes it to the exit. There's a fundamental responsibility to perform differential rather than normative diagnosis.


Patients do not have a collective responsibility to doctors, but I dont think doctors have the responsibility to provide a diagnosis at all.

I think people have strange and exaggerated expectations.

You don't go to a car mechanic and expect them to have all the answers and perfect accuracy. The expectation is that they will take a look and provide their fallible opinion.


Doctors should be held to higher standards than auto mechanics.


What standard of care are you proposing? Please be specific.

Auto mechanics frequently misdiagnose problems, especially those caused by electrical or software faults. But in the worst case they can usually just keep following the manufacturer's service manual and replacing parts in a trial-and-error process until the vehicle works again.

The human body is orders of magnitude more complex and there is no service manual. We have a few evidence-based medicine clinical practice guidelines but those cover only the simplest of cases. For anything more complex, physicians have to fall back on theory, intuition, and experience. It's not surprising that they sometimes get it wrong. And sometimes there's just no way to make a definitive diagnosis for the root cause of a patient's complaints and so treatment is necessarily symptomatic; this can be tough for patients to accept. I'm not trying to defend clinicians who make preventable errors or dismiss legitimate patient concerns but we need to be realistic about what is achievable given the current limited state of medical knowledge.


The standard here was the doctor didn’t listen to the woman. He could easily have found she was lying or not with a simple overnight visit to the hospital. Put her in there for one night, see she’s not drinking but still drunk, and that’s it. Instead, she suffered for ages.

Listening is a simple standard. Doctors don’t listen because they don’t care further than getting more patients through the door faster.

If I tell my mechanic there is a problem with the steering, he’s not going to change the oil and send my car out, he’ll check the fucking steering.


Did the woman come in and say "I have auto-brewery syndrome, and I want you to test me for it"?

If so, then the comparison is off. It isnt that the doctors "didn't listen", it is that they didn't correctly deduce a 1 in a million cause, based on the information they had.


She came in, the doctor asked if she had drunk alcohol, she says no, what more is there to understand? This doctor didn’t listen. If he had, even if he didn’t know about this specific disease, he could have started tests and brought in other doctors. Instead, suffering. Not hard to understand. Doctors don’t listen.


You keep saying they don't listen, but there's no evidence they didn't hear what she said. It seems your problem is more with how much weight they put on that information, and how much effort they put into getting to the bottom of things.

Most doctors aren't interested in playing Detective for the extremely rare cause. They treat the most likely cause given the information that they have on hand


So you’re admitting they don’t listen, they just try whatever’s most common.


I think you are hung up on the word listen, and I am saying there are a lot of things that happen after they hear what a patient says.

patients aren't saying "I have auto brewery and would like you to confirm it".

Patients are instead reporting symptoms which the doctor then has to interpret and find a likely cause. Even if they 100% believe the patients, the diagnosis may not be obvious. IF they dont 100% trust the patient, or think they may be confused, then it is even harder. Patients ARE very unreliable.


So if patients are so unreliable, why not send her to inpatient for one night to find out? There was a really simple way to solve this.


To find out what? What are they looking for in this hypothetical?

That's my whole point.

Why not make her do hand stands? Why not make make her wear pink?

You are picking a test because you know the disease.


They think she’s a drunk, she says she isn’t, the easiest way to find out is ensure she can’t have alcohol. Not that hard to do at a hospital! Seems like a simple idea to me.

Remember, it took several visits to figure this out, and no one even tried this simple thing to suss out if she was truly an alcoholic.


Why do you say they never tried this? were you there?

If they kept her at the hospital and took multiple BAC measurement, her BAC would be going down, making her look like a drunk.

The only way they would see BAC spike is if they loaded her up with carbs, which only makes sense to do if they already suspected auto-brewery.


You can make up whatever standard you want, but it will fail when an impossible expectation collides with reality.

Reality is messy, and optimal care has a non-zero failure rate.


My standard is that doctors listen to patients, which they don’t do because they don’t respect them. They think they’re all knowing, incapable of mistakes, and let their personal biases rule. Every doctor I’ve been to has struggled to listen.

Listening is not an impossible standard.


I have never expereinced a doctor that doesnt listen to their patient. However most dont take everything the patient says at face value.

Processing, weighing, and interpreting what patients say is a fundamental part of their job. Sifting through crappy data and figuring out what is relevant.

You are paying them for their personal biases.


“I haven’t drank alcohol.”

The doctors interpretation:

“She’s clearly a drunk.”

Like I said elsewhere, a night in the hospital would have solved this immediately, but the doctor was unable to get past their own bias.


I don't disagree, but what you are missing is that those biases are correct 99.9999% of time.

It's unclear if the doctors she saw even knew that Auto Brewery syndrome existed.


Six nines? Really? Sounds like you need to show evidence of that claim.


Severe auto-brewery has been documented on the order of 20 times in the history of the western world. Thats why it is worth publishing a paper about, and why doctors dont suspect it, and might not even know about it.

How many people have have gone to the doctor and lied about their alcohol use? 6 nines is 1 a million patients. There are probably more nines than that.


Now you’re just mixing up sub groups and definitions to meet your criteria. Mine are simple: doctors should listen more.


Do you think it might sound similar to someone clueless about cars discussing an engine issue with their mechanic?


No, cars are not able to talk to the mechanic to tell them where it hurts.


If a particular doctor is clueless, that's what referrals are for. You don't just shrug and move on.


When there are single digits for the number of people experiencing this, and many digits for those who are alcoholics, you cannot expect anything but Occam’s razor.


Exactly, that is where patient responsibility for their own care comes into play.


Generally speaking, I defer to your expertise point of view in the matter, and I agree that it will be far easier to generate meaningless research that passes the test of appearing meaningful to reviewers than it will be to generate meaningful research that passes the test of appearing meaningful to reviewers.

However, it is an open secret that this is already true, is the thing. Meaningful peer review is already confined to islands within a system that has devolved into generating content. The automation of the process doesn't represent a tipping point, and I don't think that the ethically disclosed production of 'research' by large language models is going to represent a significant part of the problem. The errors of the current system will be reduced to absurdity by the existent ethical norms.


There's an interesting story inside this interesting story. In the counterfactual, the company continues to manufacture and sell the product as a cordless telephone for a few years before the rest of the vendors catch up with it in terms of quality. That's not what happened. Why not?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: