Caveat: Birds and their brains

I was just reading something that confirmed what many of intuit: birds are quite surprisingly smart relative to their size. Apparently it comes down to neuron count, as opposed to brain size, as such. Thus your average crow has the same number of cortical neurons as your average monkey, and that’s why crows seem as smart as monkeys, despite their much smaller brains. They pack a lot more neurons into that smaller head volume. And it explains why elephants are NOT smarter, too: they have fewer cortical neurons than the crow, despite extremely large brains.

My mother likes the birds that dwell around her house (and make quite a bit of noise, too). Here are some pictures she gave to me of her various neighbors.

A tawny frogmouth.


A pair of bush thickknees.


A king parrot.


Caveat: On Canned Beans and Related Technology

I've been trying to eat more beans or other legumes and vegetable protein.

After my cancer, my weight dropped below 70 kg, but four years later I have completely bounced back to my pre-cancer weight equilibrium, which, frankly, I think is just a bit heavier than my ideal, which I'd put at around 75 kg, maybe. I'm currently about 84 kg.

Back in 2006-2007, when I successfully dropped from around 120 kg down to 80 kg, I did so through three main lifestyle changes: 1) walking everywhere as my primary mode of transport, 2) reducing stress by quitting that horrible job in Long Beach, and 3) eating an almost entirely vegan diet.

So, being vegan is not easy, and especially in Korea. In fact, I have no ideological interest in being vegan – therefore, for example, I have no issues with eating meat when out with coworkers or friends or whatever. Nevertheless, I recognize that less meat is probably healthier, and so I try to balance my daily diet toward vegetable proteins. The hardest thing, always, has been reducing or eliminating cheese intake – despite my lack of taste buds, there are still aspects of cheese that I enjoy, including the satiety it grants, the strong, nostalgic smell of things like mac and cheese or pizza, and whatever 'mouthfeel' is, I still experience that, too.

Anyway, all of that is background to mention I was going eat some beans, today, with my rice. And although I sometimes cook my beans from scratch, I also sometimes get lazy and use canned beans. The Korean market for canned beans doesn't run further than simple "pork and beans" type things, or I guess I've seen the native red beans pre-cooked in cans, but of course that product is painfully sweetened – like the red bean paste that is so popular here – I find such sugary prepared legumes almost unbearable (if you're not familiar with it, imagine some Mexican-style refried beans, with a cup of sugar added for good measure). So mostly if I buy canned beans I prefer to get Anglosphere brands (i.e. US or Australian products in Korean supermarkets). They're hardly expensive and easy to find, and so I buy them frequently.

Now, to talk about what I really wanted to talk about: I wanted to open my can of beans, imported from Australia.

Most canned foods, these days, have those "pop tops" – you pull the tab, the can opens. I don't, therefore, own a can opener.

But this can of beans I'd bought didn't – it had the old style top: just your plain surface tin can.

The convenience store downstairs in my building sells can openers – I've seen them there, in a little display with some other common simple housewares. But I have a different approach: a very "low tech" approach, that might be familiar to my grandfather's generation.

My pocket knife (a "Swiss Army Knife" as they're called) has as a can opener tab. It's quite useful, though entirely old-fashioned. You have to develop the right rhythm of push, tilt, advance, retreat, but you can walk it around an old-style can in about the same amount of time as with a normal manual can-opener.


It occurred to me that despite being fully embedded in the 21st Century, with my computer stuff and my smartphone and my highly urban existence on the edges of the Seoul megalopolis, I still use this antiquated method of opening my canned food. And it's worth observing that that pocket knife is now 30 years old – I received it as a gift in 1988.

I snapped a picture (right). The can that I wanted to open, on the left, and a more typical 21st century can on the right, with my low-tech solution below.

[daily log: walking, 6.5km]

Caveat: Aillucinations (Pseudosystematic significance)

When an AI (artificial intelligence) hallucinates, what shall we call it? I suggest aillucinations.

These AIs are not really that smart, though. Useful, yes, and intriguing, in a science-fictiony sort of way. But they have a long ways to go.

Case-in-point: google translate, which I use quite often, does some strange things, when you give it long strings of garbage. Its neural nets try to make sense of things, and the result is hallucinatory. This has been written about extensively at the Language Log blog – here is the most recent discussion (which includes links to earlier discussions).

I was curious about instances specific to the Korean-English domain (because I'd like to show some of my students, who overly trust online translators). So I set about finding some of my own examples. 

With google translate, I can get some pretty weird stuff in the English -> Korean direction, with the following string:


These give:

도미니카 공화국
상상해 보아라.
상상해 보지 못하다.
도미니카 공화국 인민 공화국
상상해 보지 못하다.
도미니카 공화국 인민 공화국 의회
상상해 보지 못하다.
상상력이 풍부한 사람
상상해 보아라.
상상력이 풍부한 사람

Which, translated, in turn, back to English, gives:

new product
Dominican Republic
Imagine it.
new product
I can not imagine it.
Dominican Republic
I can not imagine it.
new product
Dominican Republic People's Congress
I can not imagine it.
new product
new product
Imaginative person
new product
Imagine it.
new product
Imaginative person

Can you imagine that? The google translate "imagined" it.

Going the reverse direction, Korean -> English, I got a nice result with:


Which gives:

It is the oil that is used in the oil industry and the oil is the oil. The oil is the oil, the oil, the oil, the oil, the oil, the oil, the oil and the oil. In addition to the above, the term "oil" is used to refer to the oil as the oil and the oil as well as the oil and the lubricant. In addition to the above, the term "oil" is used to refer to the oil as the oil and the oil as well as the oil and the lubricant. It is a good idea to have a good night's sleep. It is the oil that is used in the oil industry and the oil is the oil. The oil is the oil, the oil, the oil, the oil, the oil, the oil, the oil and the oil. It is the oil that is used in the oil industry and the oil is the oil. The oil is the oil, the oil, the oil, the oil, the oil, the oil, the oil and the oil. You can not go wrong.

As a matter of curiosity, I decided to experiment with papago, a clone of google translate being created by the Korean web portal Naver. So, English -> Korean, I got a result with:

u oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oeu oee ieoe iou u oe

Which gives:

넌 너의 아내를 보고 있어

Which returned to English is:

You are watching your wife.

I like that. Such a succinct translation! I found papago was easier to "trick" into hallucinating – it will begin getting creative with much shorter strings of nonsense. But it's clear the underlying technology is similar and has the same weaknesses.

Finally, I put in the following in the Korean -> English direction:


And I got:

CAUTION of a Pseudosystem of Pseudosystematic significance of significance of significance of significance of significance of freedom of consciousness of libertarian of liberate of liberality of freedom of libertarise of freedom of freedom of proceedings of freedom of proceedings of proceedings of freedom of will of proceedings

Which is awesome. Pseudosystematic significance, indeed!

Do note that finding strings that produce these kinds of aillucinations is a bit of a hit-or-miss proposition – there are many strings which "don't work" – i.e., they return simple nonsense in return for nonsense. But it can be rather addictively entertaining to keep trying various combinations and seeing what pops out.

Happy aillucinating! I, for one, seem to have found a new, useless hobby.

[daily log: walking, 7km]


Caveat: Sending a Sportscar into Space

Occasionally, I have the thought that I have arrived in the future. Most of the time, I don't feel this. Inevitably, the future arrives more slowly than I expected when I was younger, but it does sometimes nevertheless put in an appearance.

SpaceX corporation's test of their new Falcon Heavy rocket today is one such example. The real innovation is their recovery of the the booster stages for re-use. The recycling of these rocket parts, instead of just dropping them in the Atlantic, in old-school NASA style, will make space flight much, much cheaper over the long run. And the video of the simultaneous landing of two side booster rockets back at Kennedy is a pure science fiction moment, circa 1950s.

That said, Elon Musk, the visionary leader of SpaceX, is also a megalomaniacal plutocrat and basically a living incarnation of a classic James Bond movie villain. Perhaps this is the kind of person who advances humanity – I don't know. Is that just what it takes?

Musk's new rocket test needed a "dummy payload," so, in finest egotistical form, he launched his own sports car (a Tesla Roadster, manufactured by one of his other companies), with a mannequin in a space suit at the wheel. So now, humanity has launched a space-suited dummy at the wheel of a sports car, out into space, and eventually, past the orbit of Mars. Furthermore, he placed a towel and a copy of Douglas Adams' Hitchhiker's Guide in the glove box. Can you imagine the aliens finding that?

Maybe Elon Musk will move to Mars. Somebody should move to Mars, right? Why not him?

[daily log: walking, 8km]


Caveat: scrambling over the wall to the other side

Apparently Ursula Le Guin has died. 

I thought she might not. She seemed a forever type of person. But actually everyone dies.

She was a great writer. And a philosopher, though not in the conventional sense. I don't need to review her life or work – others can do that better than I can. But her writing has influenced me profoundly.

I was 12 years old when I first read the Earthsea series of fantasy books. And I really doubt that I have spent a single day in my life since then when that imagined world hasn't crossed my mind in some way or another. It's a visceral thing – I don't know that the philosophical and psychological ideas there were so impactful – though they're undeniably present in the books. I only mean that I imagined that world quite vividly, in reading those books. and so picturenow I think of it, much as one remembers a memorable trip, perhaps. For example, I think every day of the years I lived in Mexico, or the two months I spent in South America, or my one month studying in Paris, or my six months in Chicago. They were profound and memorable experiences, which shaped who I am. Likewise, the reading of those books, at that time in my life, left a similar type of indelible impression.

Her novel The Dispossessed had a more philosophical impact on me. I consider it a great philosophical novel. The "sci-fi" aspect is nearly irrelevant, except as a way to set the scene – the same story could have been written in a different way, set on Earth in some slightly altered historical context. I would put this book in my Universal Recommended books list.

[daily log: walking, 6.5km]


Caveat: Aesthetica in vivo

What I'm listening to right now.

A Capella Science, "Evo-Devo (Despacito Biology Parody)." This song is truly awesome. It's evolutionary biology. It's poetry. It's music. It's all in a package, like the miracle of life, itself. For the prototype of which this song is a "parody," see here.


B. Mac.
Oh Carroll, Carroll
Gould, Stephen Jay yeah
D-D-D-D-Davidson and Peter

One cell divide and decide on a thousand fates
Did you ever figure how they know?
B. Mac.
Are built of modules combined in a planned out way
Each new piece must be told where to go

Now there's a science helping us to understand
How our cells encode this architectural plan
Signalling each other with genetic tools oh
Oh yeah

Phenotype the interface for mouse and man
Genotype the files and the subprograms
What then are the switches, circuit boards and boot code?

Looking at the logic in the ways that we grow
Every gene directed by a signal key code
Proteins that can activate, enhance or veto
Signals are controlled by other genes that signal
Calculating in a network labyrinthal
Where the heart and liver and the hands and feet go

Signal mapping tells each region what it ought to be yo
With circuits so deeply built upon
They're older than the Paleo
The Paleozoic Era baby
In a crucial pathway changes tend to get torpedoed
Where they go calamity goes
As this cyclopic sheep knows..

See down they cascade like a domino
Like you and I drosophila
The path that makes us optical
Was laid a long long time ago
Back before we blew up the cambrian like a bomb bomb
Now my eye protein can make you see out of your bom bom
And Hedgehog and its relatives like Indian and Sonic
Set up set up in a gradient on segments embryonic
Split forebrains and asymmetric parts depend upon it
Flipping on genetic switches and logic
From devo to evo
Adult and embryo
Mostly don't evolve in the genes of the genome
Safer the mutation aimed at regulation
Keep the building blocks and swap their activation
From devo to evo
Parts have alter egos
Homologs evolved from repeats in the schema
Switch a couple bases in the proper places
You'll be watching flies grow legs out of their faces oh yeah

Stick around for Modern Synthesis the sequel
Only by combining can a new theory grow
Evolution and development amigos
Signals trigger patterns of complexity so
Switching up the switches of a signalling node
Gives a modular and simple way to evolve

Look at how our spinal segments generate a neat row
Built on a molecular clock
One cycle, one vertebra
One vertebra one vertebra baby
Speeding up its rate is snakes' developmental cheat code
That and where a lizard's feet grow
They turn off distal aminos

This is how we go from single cells to people
Every generation and in life primeval
Life in variations endless and beautiful


From devo to evo
Larva to mosquito
Patterns are resolved as the signals proceed yo
Map out a gene with a glow tag
Kill it with a morpholino
Short oligo morpholino baby

From devo to evo
Voyage of the Beagle
Body plans evolve when proteins steer the genome
In this manner life's beauty grows
Aesthetica in vivo


[daily log: walking, 7km]

Caveat: Paperclips

I did something yesterday that I haven't done in a long time: I became immersed in a rather mind-numbingly stupid game. 

In fact, I was led to this game from a philosophical discussion of the AI Paperclip Maximizer problem, in a blog I often read. I suggest you read that, first (it's short).


The game is called, naturally, "Universal Paperclips." It's in the genre of what are called "clicker" games – basically, just webpages with a few clickable controls that allow one to manipulate a kind of limited universe.

The object of the game is to fill the universe with paperclips. You start making one paperclip at a time. Click. Click. Click.

After some time, you develop automation, and then an artificial intelligence to do work for you. And then space exploring-drones, matter-to-paperclip conversion technology, paperclip-to-drone conversion technology. Etcetera. It's entirely text-based. And I spent 10 hours yesterday, filling the universe with paperclips. I believe the specific number of paperclips I produced was on the order of 30,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 (30 septendecillion = 3 x 10^54). Perhaps that's current best guess as to the mass of the universe, in grams (and maybe each paperclip weighs about a gram, right?).

But then the game told me I had run out of matter. So I had to stop. Fortunately, it was bedtime.

It was addictive, but it was mostly a one-shot experience, I think – once you've filled the universe with paperclips, you feel satisfied but there is little incentive to keep repeating the experience. That means I don't feel bad recommending the experience to others.

[daily log: paperclips, 30,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000]


Caveat: sic a principiis ascendit motus

Hoc etiam magis haec animum te advertere par est
corpora quae in solis radiis turbare videntur,
quod tales turbae motus quoque materiai
significant clandestinos caecosque subesse.
multa videbis enim plagis ibi percita caecis
commutare viam retroque repulsa reverti
nunc huc nunc illuc in cunctas undique partis.
scilicet hic a principiis est omnibus error.
prima moventur enim per se primordia rerum,
inde ea quae parvo sunt corpora conciliatu
et quasi proxima sunt ad viris principiorum,
ictibus illorum caecis inpulsa cientur,
ipsaque proporro paulo maiora lacessunt.
sic a principiis ascendit motus et exit
paulatim nostros ad sensus, ut moveantur
illa quoque, in solis quae lumine cernere quimus
nec quibus id faciant plagis apparet aperte.

– Titus Lucretius Carus (Roman poet, 99BC-55BC),

The above are lines from Book II, lines 125-141, in Lucretius' De Rerum Natura.

Here is a prose translation to English, by John Selby Watson, 1851.

For you will see there, among those atoms in the sun-beam, many, struck with imperceptible forces, change their course, and turn back, being repelled sometimes this way, and sometimes that, every where, and in all directions. And doubtless this errant-motion in all these atoms proceeds from the primary elements of matter; for the first primordial-atoms of things are moved of themselves; and then those bodies which are of light texture, and are, as it were, nearest to the nature of the primary elements, are put into motion, and these latter themselves, moreover, agitate others which are somewhat larger. Thus motion ascends from the first principles, and spreads forth by degrees, so as to be apparent to our senses, and so that those atoms are moved before us, which we can see in the light of the sun; though it is not clearly evident by what impulses they are thus moved.

This is about Brownian motion. It was written a bit less than 2100 years ago.

[daily log: walking, 1km]

Caveat: 17776

I ran across a very weird bit of avant-garde science fiction that has been created on a sports news website (SBNation), of all places. This seems unexpected. Anyway, it's a very strange thing – it's not a straightforward sci-fi story, but rather a kind of multimedia "text" in the postmodern sense. Nevertheless, it has a narrative, and the genre is definitely sci-fi.

If you don't like unexpected animations and fiddling with your mouse to make things happen, I don't recommend it, but if you don't mind those things, give it a try: link.

[daily log: walking, 7km]

Caveat: Jumping Out of the Sky

In 1960, a guy name Joe Kittinger jumped out of a high altitude balloon and fell to Earth. I don't think I could ever have the nerve to skydive, but this activity has always fascinated me. I suppose this is the ultimate in skydiving.

[daily log: walking, 6.5km]

Caveat: Smashie O’Smasherson Jr is possibly the most powerful goldfish in all of history

A conceptual artist named Neil Mendoza has created a combination of gadgetry and software that allows his pet goldfish, named Smashie O'Smasherson Jr, to interact with his surroundings with a robotic hammer. I'm not sure the fish is really in on the joke, but some stuff definitely gets smashed. Here is a link. I've embedded the video below.

Perhaps if several generations of goldfish were allowed to grow up in this environment, they'd evolve some interesting behaviors – I could imagine fish going on smashing rampages when hungry, for example.

Surely with technology like this, our future is bright.

[daily log: walking, 7km]

Caveat: Exploring the Sky of Stone

Sometimes some strange new germ of a story idea occurs to me, and I feel fairly certain I won't actually write that story. In such events, I think maybe the best thing to do is to publish the idea on this here blog thingy and maybe someday, someone else might decide it's an interesting idea.

I was thinking about the interior of the Earth. The Earth's core has a solid inner part, and liquid outer part. The boundary is a kind of surface of crystallization, expanding gradually outward at a rate of a millimeter a year or some such tiny amount, as the Earth's core cools. Not that it's cool, in there. The liquid is mostly iron and nickel, with dissolved lighter elements: sulfur, calcium, oxygen. The idea that oxygen is included got me to wondering: could some type of chemo/thermophilic lifeform emerge in such an environment?

It wouldn't be carbon-based, or even silicon-based. Iron-based, maybe? Is that chemically plausible? I don't know enough about it. But I also thought back to a book, Dragon's Egg, by physicist Robert L. Forward. It's science fiction, but it's quite "hard" science fiction, in that he's worked out the physics of the emergence of intelligent life on the surface of neutron star. It's a rather interesting book.

Anyway, couldn't a similar treatment be applied to some core-dwelling lifeform, evolving intelligence over a billion years or so down there in the deeps, in a soup of liquid metal. And maybe their main sensory systems are based on magnetism (which makes sense in an iron-based environment, maybe). And these creatures start exploring upwards… building rivers of "breathable" molten iron upwards through their sky of stone. Until they arrive on our surface and meet us – dwellers of the outermost atmosphere, frozen beings made of puffs of something less than air, from their perspective.

What kind of close encounter might that be? 

[daily log: walking, 7.5km]

Caveat: Oklo-Boom!

I'd always wondered about this idea, which I recently ran across: a natural nuclear reactor.

Of course, if you understand the principle for how uranium fission works, you know that if enough uranium gets together in one spot, you get what's called 'critical mass' and you will get a fission reaction. So there's no reason, in principle, why it couldn't happen entirely by accident, in the natural world – some uranium deposit getting too dense, by natural geological forces, and, bingo, fission! I just assumed, based on my quite limited understanding, that the odds were too low for it ever actually happen. Yet apparently, it has, and it has been scientifically confirmed. I found some links that led me to this article, at an architecture-related blog I read sometimes. It all seems quite remarkable.

[daily log: walking, 7km]