Octokiddie

> The Cambodian Ministry of Health has swabbed 12 of her close contacts, and only her 49-year-old father has tested positive. H5N1 infections typically occur in people who have been in close contact with poultry, and so far, there is no evidence that this strain has spread between people. Investigations into how the girl was exposed to the virus are underway.

I imagine this is one reason for the low-key tone of the article. Had any of those other 12 close contacts, who presumably don't also have contact with the poultry on the farm, tested positive - this would be consistent with human-to-human transmission.

The father is more likely to have been in contact with the poultry, which could explain his positive test.

Still, knowing little about a risk is not the same as knowing that the risk is low:

> Everyone was quite concerned that the girl might have had the strain 2.3.4.4b, which is circulating around the world and causing major problems in Europe, North America and South America right now. 2.3.4.4b is a new viral clade, and we don’t know a lot about it.

Also, check assumptions before concluding that this is the result of encroachment into native habitat:

> We know that, in Cambodia, the pandemic increased the amount of backyard poultry farming. Many people, for example tour guides, couldn’t work and had to supplement their incomes and sources of food for their families. All over the world, people are still struggling, which has resulted in changes in agricultural practices that can increase spillover risk. And changes to people’s health, for example malnutrition or being overweight, can make people more susceptible to getting infected.

What's the simplest way to get this running on hardware without emulation, assuming that one has no RISC-V hardware at all?

> So by violating the first rule of clean code — which is one of its central tenants — we are able to drop from 35 cycles per shape to 24 cycles per shape, impling that code following that rule number is 1.5x slower than code that doesn’t. To put that in in hardware terms, it would be like taking an iPhone 14 Pro Max and reducing it to an iPhone 11 Pro Max. It's three or four years of hardware evolution erased because somebody said to use polymorphism instead of switch statements.

The benchmark is a tight loop where the vtable lookup is a big chunk of the total computation. I don't think one can extrapolate this 1.5x improvement to real code. If anything, it represents an upper bound on the performance improvement you might expect to see.

I also didn't see anything about how the code was compiled. Various optimizations could affect performance in meaningful ways.

> The term "Anglish" was made up by Paul Jennings in 1966, however the desire to remove foreign influence from the English language is a tradition that goes back all the way to 1066 when the Norman invasion brought in the majority of foreign vocabulary to English.

It's worth noting that Paul Jennings is described as a "humourist."

> Paul Francis Jennings (20 June 1918 – 26 December 1989) was an English humourist and author.

https://en.wikipedia.org/wiki/Paul_Jennings_(British_author)

> As a result, 15 types of amino acids were detected, including proteinogenic amino acids such as α-alanine, glycine, and valine, which are common molecules in terrestrial life, and non-proteinogenic amino acids such as β-alanine, aminobutyric acid, and isovaline. The detected optical isomers (Note 5) were racemic, existing almost 1:1 ratios (Figure 4). ...

The 1:1 ratio of optical isomers is not consistent with life as we know it, suggesting an abiotic origin for the amino acids.

Amino acids are actually not very hard to make from simpler components. But quite harder to make enantiopure.

What's really hard to do is turn Acs into anything approaching the complexity we find in abundance in living systems.

It's enlightening to consider what evidence would be necessary to support a claim for the past or present presence of life. Clearly amino acids by themselves are not enough. And even enantiomerically enriched AA's could arise abiotically.

What you might look for is molecular complexity. One of the pioneers in the field, Leroy Cronin, developed a theoretical framework (Assembly Theory) that could be put into observational practice.

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

Cronin makes the analogy that if you found anything as complex as an iPhone on Mars, that would be pretty good evidence of the presence or at least influence of life at some point. It turns out you can do that at the molecular scale, but you need to come up with a way to define "complex."

> Martin states, in this very chapter, that it makes sense to break a function down into smaller functions "if you can extract another function from it with a name that is not merely a restatement of its implementation". But then he gives us:

[function that restates its implementation]

and

[function that restates its implementation]

The article's main beef is that Clean Code sets standards that it itself does not meet. The author makes that point through the above examples and many others. The sheer number of examples suggests that this is a regular pattern in the book.

The reason seems simple enough: it's really easy to give general advice about writing software. But it's quite another thing to put that advice into practice as the thing you're working on turns to goo, which is exactly what appears to have happened with Martin's model project, FitNesse.

> To make long-term presence on the Moon viable, we need abundant electrical power. We can make power systems on the Moon directly from materials that exist everywhere on the surface, without special substances brought from Earth. We have pioneered the technology and demonstrated all the steps. Our approach, Blue Alchemist, can scale indefinitely, eliminating power as a constraint anywhere on the Moon.

There's a missing step in there - the production of solar cells on the lunar surface. Having the materials to do so is one thing, but being able to manufacture them to spec on the surface of the moon is another. The article only lightly touches on this.

Still, this is a surprising achievement from an organization I didn't even know did chemistry.

> Getting the song down, I had to fight every impulse to better the song, to make it resolve where it doesn’t otherwise, to massage out the weirdnesses. I wanted to stay as true to its creator’s vision as possible, and at the end, there’s just something missing.

One thing that's glaringly missing is that the song presents itself as a story, but on even casual inspection there is none. Sailors Song entices you with the trappings of a good, melodic story song, and then pulls a Houdini.

Compare this song with The Gambler, the best-known version of which was recorded by Kenny Rogers. Or American Pie. Or Cats in the Cradle.

This seems like a kind of uncanny valley. Real enough to make you take notice, but fake enough to turn you off.

That said, I don't think the author tried very hard to get ChatGPT to add the "something missing." I think that's because s/he hadn't figured out what the missing thing was.

For example, ask ChatGPT to tell the story of the Chinese balloon to the tune of Sailors Song. Or, literally any event that has ever happened and which ChapGPT would know about.

It's actually kind of surprising that ChatGPT came up with the story song idea in the first place because that wasn't part of the prompt. Why not give it a few nudges in the right direction?

Coindesk offers some context:

> Kraken’s staking service offered a 20% APY, promising to send customers staking rewards twice per week, according to its website.

https://www.coindesk.com/policy/2023/02/09/kraken-agreed-to-...

Simple question: how is that guaranteed yield sustainable if not through a Ponzi scheme?

> The C4 attached to the pipelines would be triggered by a sonar buoy dropped by a plane on short notice, but the procedure involved the most advanced signal processing technology. Once in place, the delayed timing devices attached to any of the four pipelines could be accidentally triggered by the complex mix of ocean background noises throughout the heavily trafficked Baltic Sea—from near and distant ships, underwater drilling, seismic events, waves and even sea creatures. To avoid this, the sonar buoy, once in place, would emit a sequence of unique low frequency tonal sounds—much like those emitted by a flute or a piano—that would be recognized by the timing device and, after a pre-set hours of delay, trigger the explosives. (“You want a signal that is robust enough so that no other signal could accidentally send a pulse that detonated the explosives,” I was told by Dr. Theodore Postol, professor emeritus of science, technology and national security policy at MIT. Postol, who has served as the science adviser to the Pentagon’s Chief of Naval Operations, said the issue facing the group in Norway because of Biden’s delay was one of chance: “The longer the explosives are in the water the greater risk there would be of a random signal that would launch the bombs.”)

This is the hardest part to believe in the piece. Making this kind of very technical change (timer to remote detonation) at essentially the last minute seems very hard to pull off. I think the piece could have benefitted from more than one view (besides Postal) on how difficult such a change would be.

That said, that last quote by Postal offers an alternative explanation. Namely, that the explosives were rigged for remote detonation but exploded accidentally.

> Beyond our own products, we think it’s important to make it easy, safe and scalable for others to benefit from these advances by building on top of our best models. Next month, we’ll start onboarding individual developers, creators and enterprises so they can try our Generative Language API, initially powered by LaMDA with a range of models to follow. Over time, we intend to create a suite of tools and APIs that will make it easy for others to build more innovative applications with AI. Having the necessary compute power to build reliable and trustworthy AI systems is also crucial to startups, and we are excited to help scale these efforts through our Google Cloud partnerships with Cohere, C3.ai and Anthropic, which was just announced last week. Stay tuned for more developer details soon.

I'm not sure how any Google representative can make such a statement with a straight face at this point. Google has such an abysmal history of pulling the rug out from under third party developers (and users) that it seems irresponsible to trust Google at this point.

> Use an AI face generator to create a completely believable face to match your new name. Download it once and use it everywhere. Run it through face aging software to use this same persona for the rest of your life.

This approach will need a plan for dealing with Zoom. It has normalized, in a very short time, the video component of what used to be strictly audio - the phone call. If you don't display your face, people get mad.

As far as I can tell, this process doesn't make any electrolysis process more energy-efficient. It merely allows the use of untreated seawater, thus sparing possibly scarce freshwater resources.

It's not nothing, but it's also not a giant leap forward AFAICT.

This sounds like the parental version of swatting someone. Hideous.

> In August 2022, NASA will send a space probe to an asteroid dubbed 16 Psyche that resides in the Main Asteroid Belt between planets Mars and Jupiter.

The article is from last year. The launch was delayed and has not yet taken place:

> In June 2022 NASA found that the late delivery of the testing equipment and flight software for the Psyche spacecraft did not give them enough time to complete the required testing, and decided to delay the launch, with future windows available in 2023 and 2024 to rendezvous with the asteroid in 2029 and 2030, respectively.[18]

> On October 28, 2022, NASA announced that Psyche was targeting a launch period opening on October 10, 2023, which would correspond with an arrival at the asteroid in August 2029.

https://en.wikipedia.org/wiki/Psyche_(spacecraft)

> How to handle this situation best?

You haven't explained what you'd like to see change. What are you after exactly? A raise? Higher social standing within the company?

When you know what you want to change, you'll be in a position to consider how feasible that change will be based on your 3 years of experience at the company.

If your desired changes are not feasible, I'd say find another position immediately.

Actually, I'd suggest doing that regardless of how feasible the changes will be. It's a lot easier to negotiate with another job offer on the table. If you can't find another offer, then you may have to re-evaluate your asks.

The article doesn't actually substantiate the claim that the cuts are being made to maintain the dividend. The article does claim, through unnamed sources, that the pay cuts are real.

But if history is any guide, the dividend cuts are also coming.

The current yield is over 5%, which is higher than any duration of US Treasury. This should be a pretty low bar, but over the last 30 years isn't. When company dividend yields approach this threshold, something often gives.

https://www.nasdaq.com/market-activity/stocks/intc/dividend-...

It's also worth noting that as late as a few weeks ago there were several articles swearing up and down that there's no way Intel will cut its dividend. That's looking more ridiculous by the day.

From what I can tell, this is a very common technical career arc. It's only a problem if you view it as one.

> Salameh told Reuters that commercial banks in the country "will see the part of their equity that is in pound decrease once translated into dollars at 15,000 instead of 1,500."

Why would a bank have any pound-denominated assets at this point?

> ... I don’t want to convince anyone of my existing perspective. I would rather be convinced of theirs. It’s more interesting to assume that they are right.

There's no need to assume anything. Ask good questions instead.

It might not seem like it, but asking questions is one of the most important skills there is. This skill is not taught in K-12. If anything, you're taught to assume that what you read/see is right. In many degree programs, you also won't be taught how to ask questions but rather repeat what you've read or heard, with some level of analysis.

The difference between good questions and bad questions often comes down to leading. The subtext of a leading question is that the asker is trying to push the askee in a particular direction. Those question tend to yield bad answers of the kind that won't convince you of much.

Better questions are interrogative-led. They start with words like "who", "where", "what" and "why", not words like "do" or "are". It takes practice to ask these kinds of questions and the follow-ups that are needed to actually be convinced by someone else's argument.

> There's a reason everyone has a clever idea for a marketplace - they don't require you to actually build a standalone, unique product. And the technical barriers to entry are lower than high tech (like cybersecurity).

This wasn't one of the lessons in the article, but hopefully it was learned: marketplaces are by far the hardest kind of online business to succeed at. The reason is simple: you must satisfy not just one but two entirely different sets of customers: buyers and sellers.

It's not 1 + 1 but more like 2 x 2 thanks to the bootstrapping problem. Buyers will only go where the sellers are, and sellers will only go where the buyers are.

So I kind of disagree with the author about the mote. A successful marketplace has a gigantic moat - its successful bootstrapping phase where the founders accomplish the extremely difficult task of attracting enough buyers and enough sellers to create an actual marketplace.

Have a look at Story Grid:

https://storygrid.com

There's a book. There's a free podcast with an extensive catalog. There are paid courses.

I suggest reading the book first to get the gist. The team behind Story Grid has really taken the original concept into some highly technical and fascinating directions. You can go as deep as you want, but they take it very deep.

Most importantly, there's a system for thinking about stories. What makes them work (or flop). What makes people pay attention (or tune out). The underlying conventions and structure all good stories share. How to learn from master works.

These lessons work not just for fiction, but non-fiction as well. It's a powerful mix of theory and practice that hash changed how I look at movies, TV series, fiction, and even news articles.

A lot of the complaints in the article have to do with the near-synchronous nature of the communication. It undermines focus and prevents deep thought.

What happens in a company that has normalized this kind of always-availability when an employee decides to regularly not respond within 5 minutes?

These hucksters come and go with boring regularity, following the Bitcoin hype cycle. What they fail to understand is that Bitcoin doesn't need them. It doesn't need their money, and it doesn't need their ra-ra. I imagine this comes as a shock, and explains in part why these folks are so frequently embedded with obvious scams:

> These crypto companies include the exchange FTX—Thiel and Mr. Wonderful have that in common—and BlockFi, a Thiel-backed service that tried to persuade people to take money out of their savings accounts and put it in crypto instead. Both companies promoted crypto as a no-lose proposition for normal savers, and both are now bankrupt, leaving the investors who trusted them with huge potential losses. Thiel will also almost certainly lose money on these bankruptcies, too, but he’ll be OK. Thanks in part to his well-timed Bitcoin sale, Founders Fund returned $13 billion to investors over a two-year period, according to the FT.

> Someone fed my essays into GPT to make something that could answer questions based on them, then asked it where good ideas come from. The answer was ok, but not what I would have said. So I asked: What would I have said?

It's not entirely clear whether or not the essay was written by GPT because it's not clear who PG asked. Himself or GPT?

I'm leaning toward the former, but can't rule out the latter. Maybe that's the point.

It's worth noting that the general class of reaction described here ("Haloform reaction") was discovered in 1822, and rediscovered after that:

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

Organic chemistry is a very old field whose older literature can still be used and remains relevant today.

Also, inhaling anything you produce in a lab, even if you did distill it, is idiotic.

> The initial focus will be on how a new currency, which Brazil suggests calling the “sur” (south), could boost regional trade and reduce reliance on the US dollar, officials told the Financial Times. It would at first run in parallel with the Brazilian real and Argentine peso.

That last part is pretty important. Both countries have a history of severe inflation. Circulating a new currency alongside the old seems like a pretty baldfaced attempt at inflation without having to admit it. It beggars belief that either country would retire its national currency given the deflationary shock that would create.

> I don't know when things started to go wrong, but the work eventually began to stagnate. Tasks were taking longer than anticipated. I began dreading team meetings because I had to address that things weren't getting done. As we approached the end of the semester, students needed time to study for finals, which meant even less work getting done. The project turned from something that I woke up excited to work on, to something I hated even thinking about. I avoided the topic as much as I could when talking to my professor.

This is the turning point. Before this point, there was progress and enthusiasm. After, stagnation and revulsion. Something happened. The author knows it, but can't elaborate.

Some have offered advice on the writing aspect ("vomit draft"). Writing doesn't appear to be this author's problem - it's working on the project itself. The author "hates" even thinking about it. No research progress means nothing to write about. Writing more effectively won't solve the problem because the problem isn't writing.

The problem is motivation. The author started with a project that was going to be awesome. Then little things started to go wrong and didn't get fixed. Those little things snowballed into big things. The disappointment and shock of watching an awesome project with buy-in from authority figures and lavish resources reverse so decisively has undermined the author's ability to move the project forward.

I'm speculating, but these things tend to follow a similar pattern.

I'd recommend that the author try to understand the turning point. What actually happened? What was the first indication that something was not right with the project? What did the author do about it? What pattern of behavior can the author pick out in responses to unexpected problems?

> Over the past two years we’ve seen periods of dramatic growth. To match and fuel that growth, we hired for a different economic reality than the one we face today.

Translation: we fell for the economic mirage.

I'm not buying it. Google sees something on the economic horizon and is now bracing for impact. And that something is not a mean reversion. It's an overcorrection.

Google's main product, advertising, puts it in a unique position to see into the economic future. Google is a bellwether. If the coming recession were truly going to be "mild," then Google wouldn't care.

It's time for everyone in tech to admit that the last two years were a mirage driven in part by corporate FOMO. Many do, but I don't think they're prepared for what comes next.

Tesla is in the process of discovering that the electric car was in fact never a disruptive innovation in the Christensen sense. It's a sustaining innovation. Under the sustaining innovation model, the upstart is not likely to succeed because the innovation plays the the incumbents' strengths.

It's just taken time for the industry to respond.

I say this because I interpret Tesla'a nonstop obsession with self driving as an admission they're in a race they will lose. The feature is a gizmo that appeals to techies and can, for a time, prop up sagging unit margins. But it's a nice to have bobble at best compared to the real prize of dominating the car market.

Christensen book, The Innovator's Dilemma even specifically identifies the electric car as a potentially disruptive innovation, mapping the path an upstart could take to eventually dominate the car market. Tesla never took that path, and instead did the exact opposite. It sold social status to wealthy customers rather than basic transportation to an abandoned market.