Octokiddie

From the original paper (linked in the article):

> ... When given a mixture of original and general abstracts, blinded human reviewers correctly identified 68% of generated abstracts as being generated by ChatGPT, but incorrectly identified 14% of original abstracts as being generated. Reviewers indicated that it was surprisingly difficult to differentiate between the two, but that the generated abstracts were vaguer and had a formulaic feel to the writing.

That last part is interesting because "vague" and "formulaic" would be words I'd use to describe ChatGPT's writing style now. This is a big leap forward from the outright gibberish of just a couple of years ago. But using the "smart BSer" heuristic will probably get a lot harder in no time.

Also, it's worth noting that just four human reviewers were used in the study (and are listed as authors). That's a very small sample size to draw conclusions from. The article doesn't mention level of expertise of these reviewers, but I suspect that could also play a role.

Some are focusing on the paper-mill angle. But I think the more interesting angle is ideation.

If researchers can't reliably tell the difference between machine and human generated abstracts, what kinds of novel experiments or even research programs could ChatGPT suggest that might never have been considered?

And Astronauts.

The article quotes this:

> With a native Swift implementation of Foundation, the framework no longer pays conversion costs between C and Swift, resulting in faster performance.

and this:

> A reimplementation of Calendar in Swift is 1.5x to 18x as fast as the C one (calling from Swift in various synthetic benchmarks like creation, date calculation).

First, that range of 1.5x-18x is kind of huge. Why?

Second, why would the intro with C be such a big performance penalty even assuming it's just 1.5x? I know there must be an overhead, but why so large?

Also, why single out the Calendar? Is it somehow representative?

> The problem, in a nutshell, is that for Bitcoin and other digital assets to have any chance of becoming mainstream, banks—with their access to deep sources of liquidity and experience in facilitating payments—will have to be on board. For the time being, they’re running in the opposite direction.

What does Bitcoin "becoming mainstream" look like specifically? The article doesn't offer anything - not even a link.

Let's say it's 1974 and you're writing an article about the desktop computer "becoming mainstream". What does that look like exactly to you? I'm guessing it doesn't look anything like the last 30 years. Not even close.

So how can the author be so confident in asserting that banks must be "on board" for this mainstreaming to happen?

The desktop computer was an attack from below, in the Christenson Innovator's Dilemma sense. It looked and acted exactly like a toy for the first decade of its existence. Most incumbents who saw through that surface illusion were powerless to do anything about it. The one that did (IBM) tried, briefly succeeded, then lost the entire thing.

Banks with their highly touted "experience in facilitating payments" are the very thing Bitcoin has in the crosshairs. Banks losing interest now that the Bitcoin FOMO is gone? Good riddance to them and their AML/KYC bootlicking.

8 deaths due to "Wildlife/Animal" vs 17 deaths due to "Homicide." In other words, you're twice as likely to be killed by your own species as another. But that pales in comparison to other categories such as "Natural Death" (285), "Motor Vehicle Crash" (475), and Drowning (668).

> Meta's position until now was that FTE offers are NOT at risk. Up to even a week ago.

This is the danger of fixating on the economic rearview mirror. And few things are more rearview than hiring and layoffs.

Leading financial indicators such as yield curves and oil futures are painting a very clear picture and it's not good.

Two things to try:

1. Outline the next thing you write, before you write it. It sounds like rewriting as you write is a problem. Outlining gets you out of that habit like few other things can. It forces you to focus on what you're trying to convey and who you want to address.

2. You didn't mention anything about you HN (or other media) use. If you're using that as a way to escape, it's time to stop.

Any ideas on why miniwasm performs better on all the benchmarks except "trap," on which it performs decidedly worse?

> ... Methamphetamine has a chiral center, so you can produce the pure (D) isomer, the pure (L), the exact 50/50 racemic mixture, or anything in between. (L) methamphetamine is a pretty good decongestant, but it is otherwise unenjoyable and has no market as a drug of abuse. The classic P2P reductive amination will of course give you the racemate; there’s nothing chiral about the synthesis at all. At the other end of the scale, reduction of pseudoephedrine starts with a pure chiral material, and you get only (D) methamphetamine as a product.

What he's talking about is molecular mirror images. Our bodies can tell the difference between the (L) and the (D) form. The former drains your sinuses and the latter gets you high.

This is a tangential plot point in Breaking Bad. In the episode "Boxcutter," Walter White asks: "If our reduction is not stereospecific, then how can our product be enantiomerically pure?"

There's a lot to unpack there, but it implies that Walt is isolating (D) methamphetamine exclusively. It also implies that this is occurring through a selective reductive procedure.

The first few seasons established that the synthesis relies on methylamine through a reductive procedure called P2P. But that procedure produces (D) and (L) forms in equal quantities.

What Walt is revealing in that one sentence is that he's developed a way to do the reductive amination step in the P2P cook such that just the (D) form is isolated. His use of the term "stereospecific," implies that the selectivity is close to 100% in the bond-forming step itself. The process therefore doesn't just produce highly pure product, but does so very efficiently.

At the time period of the show, such a reaction would have been very close to cutting edge research.

I can't tell if this is parody or not. Either way, the proposal is stupid and amounts to check-and-guess. Zero analysis of probability of success. Zero estimate of resources required.

But this makes me think the article is not actually parody:

> If you want to send some cryptocurrency to someone else, you encrypt the transaction using their public key which is as its name suggests, public, and your private key which is known only to you.

Encryption does occur on some chains, but none of the widely-used ones. So, not parody and just another ignorant take.

> There was a time when you have to install a dedicated sound card in order to have audio coming out from your PC. You have a number of options but there was one particular name that dominated the scene for so many years and that name is Sound Blaster, for which you have Sim Wong Hoo to thank.

Not only that, but doing so put you on an upgrade path. There were so many models released over the years. Just a tad like graphics cards today.

And it was an ordeal to get Linux working with them. Every. Single. Time. Hearing that test sound: "Hello, this is Linus Torvalds and I pronounce Linux as Linux" was grounds for a major celebration.

Kind of shocking to recall that there was a time when you had to plunk one of these cards into a computer before you could hear much of anything useful.

From another thread, the actual application appears to be something called a "hike and fly race":

> The Global Rescue XRedRocks is a premiere hike and fly race in North America, organized in a similar way to the Eigertour, Vercofly and DolomitiSuperfly- multi-day hike and fly events that take participants into magnificent mountains to see what they’re made of when we pair back free-flight to it’s most raw and exciting form. Travel is only allowed by wing or on foot. There are no supporters. ...

https://xredrocks.com

The devices appear to be internet hotspots, possibly powered by the Helium network:

https://gristleking.com/helium-deployed-the-network-in-actio...

This could explain why you'd want to put a hotspot at the top of a mountain: to provide real-time tracking of paragliders during the competition.

The hotspots in the local news clip could be put there either for real-time telemetry during the competition or for training.

The actual application appears to be tracking paragliders. Helium just provides the internet hotspot:

> Over the course of a week, supported by Tommy and Ryan at Lonestar Tracking, Matthew at Digital Matter, Travis at Helium, and Jeremy C (@jerm on Discord), I deployed 2 off-grid Helium Hotspots high in the mountains of Utah (one at over 8,000′ and one above 11,000′) to track 30+ paragliders as they flew during the annual Red Rocks Fly In as well as raced during the inaugural X Red Rocks Hike & Fly race.

This explains why somebody would want a hotspot in such a remote, high location.

The race itself is kind of wild:

https://xredrocks.com/

From the Rules page:

> The task is to reach the control and turn-points defined by the Race Committee every day for three days as quickly as possible traveling only by paraglider or on foot.

I imagine you'd want good tracking data to get an idea of where you're going.

> Four separate entities have laid claim to the approximately 56 million shares, worth about $460 million. FTX’s new management, which is trying to claw back funds for investors and customers of the bankrupt platform, have sought to wrest control of the shares from an Antigua-based holding company 90% owned by Bankman-Fried.

The article mentions:

1. SBF himself

2. FTX

Who are the other two? This seems important because those other two entities may have recorded the shares as assets or even themselves pledged them as collateral.

Do you feel the urge to share these opinions in person? If not, why do you feel the urge to do so online? If so, then you can use the same approach you use in meatspace: discretion.

> The main point is that your function can only return either a success or a failure. Failure should be handled using the throw statement so as to throw an exception, while success is what leads to another function, which can be of any type.

Rust bakes this into the language with the Result type. You signal railway oriented programming by returning a Result, which contains either the thing you want or an error. If you try to get the thing, you must deal with the possibility of an error in some way. There is no half-way.

/s ?

Some are suggesting that the author is against the use of tagged unions.

I don't read it that way. What the author is against is using parameters for the specific purpose of controlling the flow of a function. Instead, the author recommends breaking the function up into as many functions as there are variants of the type key.

This also applies to boolean flags (where only two variants are possible). There are, of course, exceptions as the author points out.

The small win is that the type key no longer needs to be maintained. The big win is that each function that gets split out has fewer reasons to change. The individual functions are therefore easier to maintain and understand.

What's he on about? I can't see the tweet he's replying to.

> I think the large caps in web3 (BTC and ETH mainly) will start to attract more interest from investors and should do well in 2023. I am more bullish on ETH personally because it has the best underlying economic model of any web3 asset.

It's bizarre to lump Bitcoin and Ethereum together at this point because the value propositions are so different:

- Bitcoin: censorship-resistant money

- Ethereum: tokens (i.e., unregistered securities)

Reading between the lines, I think what the author means by "best economic model" is that censorship-resistant money has a smaller market than tokens.

The problem is that a token economy without censorship resistance will face constant regulatory pressure. Proof-of-stake gurantees that increasing legal pressure applied to exchanges will find an outlet at the protocol level.

All of the people in denial about this are going to wake up to a new reality in which those dreams of decentralized organizations and free token economies are shut down in spades by regulators who see the obvious and have the power to clamp down on it.

Without censorship-resistance, any financial system you try to build ends up looking exactly like the old. And that growing lack of differentiation offers little incentive or reward for the risk involved.

This is going to be an extremely hard lesson for a lot of people, including the author I suspect.

Unless the peg breaks, the currency doesn't "slide."

> Then, in 2021, a team of archaeologists dropped a bombshell into this debate: they’d found footprints – unquestionable evidence of a human presence – at White Sands National Park in New Mexico, dating to between 23,000-21,000 years ago.

The article places that date long before the Last Glacial Maximum (LGM):

> It [the Clovis point] appeared widely across the continent beginning about 13,400 years ago, near the end of the LGM.

Which raises the question of what people were doing in the Americas before the LGM?

USGS has this to say about what might have been going on with the climate at the time the White Sands footprints were laid down:

> During the late Pleistocene (~129,000 to ~11,700 years ago), full glacial conditions were interrupted repeatedly by abrupt warming events called Dansgaard-Oeschger, or D-O events. In the southwestern U.S., D-O events caused centuries-long megadroughts, which devastated spring ecosystems, lowered lake levels, affected sea-surface temperatures and circulation patterns, and even triggered seismic activity. One of these events, called D-O 2, dates to ~23,500 years ago and was especially impactful to spring ecosystems, as water tables never returned to the same levels after the event that they had maintained for thousands of years prior.

https://www.usgs.gov/programs/climate-research-and-developme...

Wikipedia notes that D-O events flared up extremely rapidly on a geological timescale:

> In the Northern Hemisphere, they [D-O events] take the form of rapid warming episodes, typically in a matter of decades, each followed by gradual cooling over a longer period. ...

https://en.wikipedia.org/wiki/Dansgaard–Oeschger_event

> Experts say the adoption of the euro will help protect Croatia’s economy at a time when inflation has been soaring globally since Russia’s invasion of Ukraine in February led to heightened fuel and food prices.

Wikipedia notes that the kuna has been pegged to the euro for some time:

> The modern kuna was introduced on May 30, 1994, starting a period of transition from the Croatian dinar, introduced in 1991, which ended on December 31, 1994.[10] One kuna was equivalent to 1,000 dinars at a fixed exchange rate. The kuna was pegged to the German mark from the start. With the replacement of the mark by the euro, the kuna's peg effectively switched to the euro.

So I doubt there will be much effect on the rate of inflation after the switch.

> The new calendar was adopted in some European countries where 4 Oct 1582 was followed by 15 Oct 1582 thereby skipping 10 days in between. However, it took as long as September 1752 for the new calendar to be adopted by Britain. In Great Britain and the British Empire, 2 Sep 1752 was followed by 14 Sep 1752 and that is the gap of 11 days we see in the cal 9 1752 output.

This is also the reason that some Christian groups celebrate Christmas on Jan. 7 rather than Dec. 25:

> “Christmas on Jan. 7 is also known as Old Christmas Day. Eleven days were dropped (from the Gregorian calendar) to make up for the calendar discrepancy that accumulated with the Julian calendar when England and Scotland switched in 1752,” Fr. Sammour explained.

https://www.arabnews.com/node/1999461/middle-east

> From the airlines, to healthcare, to education, to restaurants, every industry that requires workers to be in-person, face-to-face with customers, has seen drastic issues with staffing since COVID began. Those staffing issues worsened when masks came off. This is precisely the outcome public health experts warned of when “the urgency of brunch” was pursued as national pandemic strategy.

This seems to be the thesis. What I don't see is a link to anything demonstrating that masks played a material role in controlling the pandemic.

By now, it's very common for people to know of others who have contracted covid despite multiple vaccinations.

Where are the clinical data showing efficacy of current vaccines against contracting and spreading covid?

The article that Snowden cited:

https://www.cbc.ca/news/canada/manitoba/sunil-tulsiani-speak...

is freely available. Tulsiani's reputation must be known to everyone involved in this organization, which certainly smells like a scam. What did Snowden hope to accomplish by first playing along, and then bringing this old case up again?

The format of this summary doesn't do the book justice. The book has a big idea (seek "No", don't try to avoid it), but the format hides that pretty well. There's a lot to unpack before that idea makes any sense, but this ordered list doesn't do the job.

Also, the pronouns "I" and "my" imply that the summary was written by the author (Jim Camp), which it was not.

The article doesn't fulfill the promise of the title. Not only does it not explain how to identify a scientific fact (who samples the 95% "consensus" and who decides what to ask, for example?), it doesn't even define the term, despite multiple uses.

The reason is, of course, that there are no scientific facts. There are hypotheses, theories, and observations. There is a scientific method, which attempts to connect them together.

This is a deeply unsettling perspective to a lot of people who demand something science can't deliver but which religion can, and I'd count the author among them.

Whenever an author starts tooting about "scientific facts" I sense an agenda. And it's often followed by demands that I submit to it.

Elsewhere, the author pitches the book like so:

> My new book Identifying Future-Proof Science argues that we can confidently identify many scientific claims that are future-proof: they will last forever (so long as science continues). Examples include the evolution of human beings from fish, and the fact that the Milky Way is a spiral galaxy. ...

https://petervickersphilosophy.com/publications/

The first claim is laughable. Without a path for abiogenesis (there is none yet), claims this specific about evolution are anything but "fact." And the tautology of classifying the Milky Way galaxy as a galaxy makes me wonder if the author has problems with basic logic.

> ... Padding oracle vulnerabilities, use of ECB mode (leaks information about password length and which passwords in the vault are similar/the same. recently switched to unauthenticated CBC, which isn't much better, plus old entries will still be encrypted with ECB mode), vault key uses AES256 but key is derived from only 128 bits of entropy, encryption key leaked through webui, silent KDF downgrade, KDF hash leaked in log files, they even roll their own version of AES - they essentially commit every "crypto 101" sin. All of these are trivial to identify (and fix!) by anyone with even basic familiarity with cryptography, and it's frankly appalling that an alleged security company whose product hinges on cryptography would have such glaring errors. ...

Seems bad.

I took the author's advice and did the google search "jeremi gosney" + "lastpass". This 2015 article turned up:

> A longtime LastPass user himself, Gosney says he doesn’t plan to change a thing following the breach – not even his master password. “I am confident in the strength of my master password. As a password cracker, I know it’s impossible to crack,” he says.

https://www.secureidnews.com/news-item/consultant-dont-sweat...

When the same author now recommends Bitwardenen and 1Password, it's hard to view it as anything other than more of the same.

The author might want to tone down the unqualified language around computer security to avoid history repeating itself.

> And as I said, there’s a parallel here with Bitcoin. Despite years of effort, nobody has yet managed to find any serious use for cryptocurrency other than money laundering. ...

I suspect Krugman is being deliberately dishonest here. The use case is censorship-resistant money. Not just money laundering, but any use of money outside national jurisdiction. This includes routing around capital controls, civil asset forfeiture threats, and increasingly draconian financial surveillance. Not just in developing countries and in dictatorships, but in western-style democracies.

It's understandable that the privileged like Krugman hate this idea. It's not unlike thinking that the Internet's effect on the economy will be no greater than the fax machine. Krugman wasn't the target user then and he isn't now.