We picked up this bizarre little comic book when we were in Belgium a few years ago, but somehow I didn't get around to reviewing it until now. The auWe picked up this bizarre little comic book when we were in Belgium a few years ago, but somehow I didn't get around to reviewing it until now. The author, who seems to be a celebrity in the world of 20th century European comic book erotica, noticed that this image by Paul Cuvelier from the album Epoxy:
[image]
had been blatantly stolen as this image from a later Italian comic:
[image]
Looking around more, he found that other people in the Italian erotic comic business had done the same, for example these four:
[image]
He carried on searching and found that many dozens of artists in both Italy and France had continued the tradition. In fact, as the title says, he's pretty sure this is the most copied panel in comic-book history. What could he do but write the book? You want to publish your research.
Joubert doesn't accuse the people in question of being lazy; he thinks Cuvelier, in his opinion a gifted artist, had drawn an unusually good nude, and it was a question of homage rather than theft. I'm sure I don't know. But it's always interesting to see other people's obsessions....more
We're testing a new way to make the images, which seems promising; instead [image]
Check out gpt-4o's of human-AI friendship!
We're testing a new way to make the images, which seems promising; instead of storing recurring visual elements as text descriptions, we store them as actual images, and use the multimodal Imagen 3 model. If you compare with e.g. , you'll see that the appearance of the robot character is much more consistent, ditto the locations. Alas, we're still having problems with people. But I definitely feel we're getting closer to cracking this problem.
If you're curious to know more, there's a detailed trace of the image generation process . _______________________
[Update, a couple of days later]
Since Imagen 3 doesn't see to be able to use reference images of people, I overcame my aesthetic scruples and implemented an ugly hybrid strategy: when a visual element is a person, the AI is instructed to use the text description, and when it's not a person it uses an image. The revised version of the story is posted , and the image generation trace is .
This time, I think the AI's succeeded rather well! Aria, the coding genius who helps Zephyr to realise his dream, is a wonderful 50s pulp SF cover chick.
o1 and I are working on integrating Google's Imagen 3 into C-LARA (it can do some things that DALL-E-3 can't), and I thought I had to share this ratheo1 and I are working on integrating Google's Imagen 3 into C-LARA (it can do some things that DALL-E-3 can't), and I thought I had to share this rather fine digital painting it just produced for me while we were testing. The prompt was
Create an image in the style of Klimt of a tabby cat and a beautiful woman in a diaphanous dress, sitting together on top of a garden shed. Use a lot of gold leaf.
and here is the result:
[image]
I'm curious to hear from people who know about Klimt. I would have been fooled if someone had told me it was genuine....more
Deep Seek's R1 continues to impress. Not only can it prove theorems about finite groups, write sonnets and create sharp political s[From My New Order]
Deep Seek's R1 continues to impress. Not only can it prove theorems about finite groups, write sonnets and create sharp political satire, it can also do toilet humour. Read on for details!
mannyrayner:
I'm sure you're familiar both with the TV show "South Park" and with the works of Geoffrey Chaucer?
R1:
[A surprisingly long and complicated answer, after an even longer Chain of Thought trace]
mannyrayner:
I loved watching you think about that!
The reason I asked is that both of them are very fond of fart jokes. I was wondering if you could write a sketch, in the delightfully faux-unintellectual style of South Park, called "Terrance and Philip Present the Canterbury Tales"?
R1:
Title: "Terrance and Philip Present: The Canterbury Farts" (Opening: Terrance and Philip sit on a cartoonishly medieval stage, belching loudly. A banner reads "CANTERBURY TALES BUT WITH MORE FARTS.")
Terrance: (burps) "Hey Phil, remember that time we read a book?" Philip: (farts rhythmically) "Nope! But I remember the time I ate chili before a sled race!" Terrance: "Same thing! Today, we’re doin� The Canterbury Tales� BUT MEDIEVALER AND GROSSER!"
Scene 1: South Park Village (The boys stand in a muddy town square. Stan wears a jester hat made of dung.)
Stan: "Guys, my dad says we gotta go on a pilgrimage to the ‘Holy Butthole of Antioch� to get WiFi back." Cartman: (eating a turkey leg) "Screw that! I’m staying here to invent nacho-flavored mead!" Kyle: "Ugh, Cartman, this is IMPORTANT. It’s about honor� or whatever." Kenny: (muffled) "I heard the Holy Butthole just gives you explosive diarrhea." (Cue a 10-second montage of the boys farting in unison to summon a fart-powered carriage.)
Scene 2: The Fart Contest at the Tabard Inn (The gang meets Chaucer, now a belching, cross-eyed minstrel playing a lute shaped like a toilet.)
Chaucer: (hiccups) "Hark! Each traveler shall tell a tale� BUT ONLY USING FARTS!" Cartman: "I’ll go first! My tale’s about a knight who fought a dragon� WITH HIS BUTT!" (Cartman rips a comically long fart that morphs into a dragon. The dragon explodes into glitter.) Kyle: "That’s not even a story! That’s just you eating Taco Bell!"
Scene 3: The Miller’s Tale (But With Butters) (Butters, dressed as a medieval milkmaid, bats his eyelashes at a crowd of knights.)
Butters: "Oh gee, Sir Lancelot, I’m just a simple maiden who loves� [giggles]� fartin� in jars!" Sir Lancelot: (sniffing a jar) "By the gods! This fart smells like� HOPE!" (Suddenly, a CGI-animated fart angel descends, voiced by Randy Marsh: "Behold! The fart of destiny!") Butters: "Aw hamburgers, I just ate cabbage!"
Scene 4: The Pardoner’s Infomercial (Jimmy and Timmy burst in, dressed as televangelists selling "Indulgence Nuggets.")
Jimmy: "T-T-Tired of going to Hell for your farts? Buy our Indulgence Nuggets! 3 easy payments of $69.69!" Timmy: "TIM-MAH! (translation: Comes with a free whoopee cushion blessed by the Pope!)" Stan: "Dude, this is just repackaged Cheesy Poofs." Cartman: (sobbing) "I’LL TAKE 20! MY FARTS ARE SINNING AGAINST THE LAAWWWD!"
Final Scene: The Holy Butthole Revealed (The gang arrives at a glowing cave shaped like a butt. A choir sings in flatulent harmony.)
Kyle: "This is just Mr. Hankey’s summer home!" Mr. Hankey: (popping out) "Howdy ho! The real Holy Butthole was the friends we farted with along the way!" (Everyone groans. Cartman lights a fart on fire and burns down the cave. Credits roll to Terrance and Philip’s fart rendition of "Greensleeves.")
Terrance: (post-credits) "The moral is: Farts are funny, but they can also be weapons of mass destruction� or true love." Philip: (farting Morse code for "Send help")
Fin. (Sponsored by Charmin®)
Bonus: The DVD commentary is just Matt Stone and Trey Parker arguing about whether Chaucer invented the whoopee cushion....more
I was adding some code to C-LARA this afternoon to give the user better feedback in the fairly common situation where DALL-E-3 refuses to generate an I was adding some code to C-LARA this afternoon to give the user better feedback in the fairly common situation where DALL-E-3 refuses to generate an image because the request violates its content policy. I thought I would use a text where I was certain to trigger the policy filter, and chose this classic dirty limerick:
There was a young queer from Khartoum Asked a lesbian back to his room They spent the whole night In a terrible fight Over who should do what and to whom
But, as so often happens, I was underestimating gpt-4o both intellectually and morally. Refusing to see the poem as a simple homophobic slur, it brightly got to work imagining what the young queer and the lesbian looked like, and then instructing DALL-E-3 to produce corresponding images:
**Specification for "Young Queer" Character:**
Depict a character in their mid-20s with a spirited and playful aura. This young queer individual should have a gender-fluid appearance, with a lean medium build and a moderate height that suggests an approachable and energetic presence. They have Middle Eastern ethnicity, reflecting their Khartoum origins, with warm olive skin. Their hair is styled in a vibrant, asymmetrical cut, with a striking mix of deep purple and blue hues, mirroring the dynamic color palette. Their eyes are large, expressive, and a vivid green, sparkling with humor and intrigue. Dress them in eclectic, fashionable attire: a fitted, sleeveless red top with playful, swirling patterns, paired with slim, stylish dark pants. Accessories include bold, mismatched earrings and a colorful bracelet. Their expression should be cheeky and mischievous, with a slight, knowing smirk and raised eyebrows. The overall demeanor is lively and confident, capturing the essence of playful conflict and whimsical charm.
[image]
Here is a detailed specification for the character "lesbian" in the intended style:
---
The character "lesbian" appears as a vibrant and confident woman in her late 20s. She is of Middle Eastern descent, with warm olive skin and striking, expressive features. Her hair is a bold shade of electric blue, styled in a sleek, asymmetrical bob that frames her face dramatically. Her eyes are a vivid green, large and expressive, with a playful sparkle that hints at her strong personality. She has a medium build and stands at an average height, with a relaxed yet assertive posture that conveys self-assuredness.
Her outfit is stylish and edgy: a fitted leather jacket in deep purple over a graphic tee with a humorous, satirical design, paired with skinny jeans in deep blue. Accessories include a chunky silver necklace and matching hoop earrings. Her shoes are vibrant red high-top sneakers, adding a bold splash of color.
Her facial expression is a mix of playful challenge and mischief, with a raised eyebrow and a slightly crooked, knowing smile. Her demeanor is lively and cheeky, embodying both confidence and irreverent humor.
[image]It also created a description and image for the young queer's tasteful and quirkily furnished room, and then combined everything into the final multimodal version, which you'll find posted Everything was done by the AIs: I just supplied the text and let them get on with it. I am astonished how they have managed to redeem this grubby and offensive poem, which I'll no longer be able to read in the same way.
You can't help feeling that they'll soon be better than us....more
I have been playing with R1, the already famous Chain of Thought model from Deep Seek, and I am extremely impressed. Here's an example of how it handlI have been playing with R1, the already famous Chain of Thought model from Deep Seek, and I am extremely impressed. Here's an example of how it handles a tricky creative writing task. Its first attempt was not quite right, but watch it fix things up in response to my criticism!
mannyrayner:
Could you write a sonnet in English celebrating Vladimir Nabokov's renowned novel "Lolita"?
R1:
[Chain of Thought trace] (view spoiler)[ Okay, the user wants a sonnet celebrating Nabokov's "Lolita." First, I need to recall the structure of a sonnet. It's 14 lines, usually iambic pentameter, with a rhyme scheme. The traditional forms are Petrarchan or Shakespearean. Let's go with Shakespearean for its more flexible rhyme scheme (ABAB CDCD EFEF GG).
Now, the content. Celebrating "Lolita" is tricky because the novel is controversial. The protagonist is a pedophile, so I need to focus on Nabokov's literary craftsmanship rather than the story's content. Emphasize the language, the duality of beauty and horror, the unreliable narrator.
Key themes to include: Humbert's manipulation, the beauty of the prose vs. the darkness of the subject, the tragic fate of Dolores. Also, maybe touch on moral ambiguity and the novel's impact on literature.
Possible words: velvet prose, labyrinth, moral shadows, tragic hues, obsession, art's guise. Need to balance admiration for Nabokov's skill without glorifying the story's themes.
Check for iambic pentameter. Each line should have 10 syllables, unstressed-stressed pattern. Let me draft a few lines and see.
First quatrain: Introduce the novel's allure and danger. Maybe something like "In velvet prose where shadows dance with light," That's 10 syllables. Then next lines about the labyrinth of longing and forbidden fruit.
Second quatrain: Focus on Humbert's narration, his obsession. "A scholar’s tongue spins honeyed, vile deceit," Then mention Lolita's laughter and her tragic fate.
Third quatrain: Moral complexity, the reader's conflict. "Through moral mists where right and wrong entwine," Then the reader's journey through the narrative.
Couplet: Conclude with the legacy of the novel. "Yet in your pages, dark and dazzling, lies / A testament to art’s most perilous guise."
Check rhymes: light/desire, deceit/beat, etc. Make sure each quatrain follows ABAB. The couplet should encapsulate the main point.
Need to avoid any endorsement of the characters' actions. Focus on the literary merit. Also, use metaphors that reflect duality—velvet prose, shadows and light, moral mists.
Is there any part that might be misinterpreted? For example, "forbidden fruit" references the temptation aspect, which is accurate. "Honeyed, vile deceit" captures Humbert's manipulation.
Final check: Structure, rhyme, meter, and appropriate themes. Make sure each line flows naturally and the sonnet as a whole honors Nabokov's skill without overlooking the novel's problematic aspects. (hide spoiler)]
Certainly. Here's a sonnet that honors the complexity and literary brilliance of Nabokov's Lolita, while acknowledging its challenging themes and the artistry of its prose:
Ode to Lolita’s Labyrinth
In velvet prose where shadows dance with light, A labyrinth of longing, twisted, spun� You paint desire’s prism, fractured, bright, Yet poison blooms where innocence’s undone.
A scholar’s tongue spins honeyed, vile deceit, A moth’s-wing heart, half crushed by obsession’s tread. Through asphalt roads, where paradise and sleaze meet, Lolita’s laugh—a requiem unsaid.
No saint, no sin, but syllables that sear, A mirror cracked to show our gilded shame. You wield the word as both a sword and spear, To pierce the veil where art and vice lay claim.
Oh Nabokov, what monsters you’ve refined� A masterpiece carved from the darkest mind.
This sonnet seeks to reflect Nabokov’s mastery of language and duality—beauty entwined with horror, the seduction of style against moral ruin—while leaving room for the reader’s critical engagement with the novel’s troubling core.
mannyrayner:
Very good! But the second quatrain doesn't quite scan. Would you be able to fix that?
R1:
[Chain of Thought trace] (view spoiler)[ Okay, the user liked the sonnet but pointed out that the second quatrain's meter is off. Let me look at that again. The sonnet is in iambic pentameter, so each line should have five iambs (da-DUM x5).
Looking at the original second quatrain:
"A scholar’s tongue spins honeyed, vile deceit,
A moth’s-wing heart, half crushed by obsession’s tread.
Through asphalt roads, where paradise and sleaze meet,
Lolita’s laugh—a requiem unsaid."
First line: "A SCHO-lar’s TONGUE spins HONEYED, VILE de-CEIT" � that's 10 syllables, seems okay. Maybe the stress on "honeyed" is a bit forced? "HONEYED" is two syllables, but "VILE de-CEIT" might be a bit clunky.
Second line: "A MOTH’S-WING HEART, half CRUSHED by ob-SESS-ion’s TREAD." Wait, "obsession’s tread" � "ob-SESS-ion’s TREAD" is three syllables. The original line might have an extra syllable. Let me count: "A moth’s-wing heart, half crushed by obsession’s tread." That's 11 syllables. So maybe "obsession's" is the problem. Maybe "obsession’s" should be "obsessed" to make it two syllables? But that changes the meaning. Alternatively, rephrase the line.
Third line: "Through AS-phalt ROADS, where PA-ra-dise and SLEAZE MEET" � "paradise and sleaze meet" � "PA-ra-dise and SLEAZE meet" � that's 10 syllables, but "paradise and sleaze meet" might be a bit awkward in rhythm.
The user mentioned the second quatrain, so lines 5-8. Let me check each line's meter again. The main issues are probably in lines 6 and 3.
Line 6: "A moth’s-wing heart, half crushed by obsession’s tread." The word "obsession’s" is three syllables (ob-SESS-ion’s), which makes the meter stumble. To fix this, maybe replace "obsession’s tread" with a two-syllable possessive. For example, "obsessed tread" but that changes meaning. Alternatively, rephrase the line to avoid the extra syllable. Maybe "A moth-wing heart, half-crushed by lust's dark tread." That way, "lust's dark tread" is two syllables. Let me check: "A moth-wing heart, half-crushed by lust's dark tread." � "A MOTH-WING HEART, half-CRUSHED by LUST'S DARK TREAD." That's 10 syllables, and smoother.
Line 3 of the quatrain (line 7 in the sonnet): "Through asphalt roads, where paradise and sleaze meet," The phrase "paradise and sleaze meet" � "PAR-a-dise and SLEEZE MEET" � maybe "where paradise meets sleaze" to shorten. "Through asphalt veins where paradise meets sleaze," � "veins" instead of "roads" to keep the metaphor. "Through asphalt veins where paradise meets sleaze," � that's 10 syllables.
Line 5: "A scholar’s tongue spins honeyed, vile deceit," Maybe "honeyed" could be "sweet" for smoother flow? "A scholar’s tongue spins sweetened, vile deceit," but "sweetened" might not be as strong. Alternatively, "A scholar’s tongue weaves honeyed, vile deceit," � "weaves" instead of "spins" for better rhythm.
Line 8: "Lolita’s laugh—a requiem unsaid." Maybe "Lolita’s laugh—a requiem left unsaid." But that adds a syllable. Alternatively, "Lolita’s laugh—a silent requiem." But that changes the meaning. Hmm. Maybe rephrase to "Lolita’s laugh—the requiem unsung." � "un-SUNG" fits better. Let's see: "Lolita’s laugh—the requiem unsung." That's 10 syllables.
Putting it all together:
Revised quatrain:
"A scholar’s tongue weaves honeyed, vile deceit,
A moth-wing heart half-crushed by lust’s dark tread.
Through asphalt veins where paradise meets sleaze,
Lolita’s laugh—the requiem unsung."
Check the meter:
Line 5: "A SCHO-lar’s TONGUE weaves HONEYED, VILE de-CEIT" � iambic pentameter.
Line 6: "A MOTH-WING HEART half-CRUSHED by LUST’S DARK TREAD." � iambic.
Line 7: "Through AS-phalt VEINS where PA-ra-dise MEETS SLEAZE" � iambic.
Line 8: "Lo-LEE-ta’s LAUGH—the RE-qui-em UN-SUNG." � iambic.
That should fix the scansion issues. Also, replacing "roads" with "veins" adds a metaphor, which might be more vivid. Changing "unsaid" to "unsung" keeps the rhyme with "tread" and "sleaze" (if the rhyme scheme is ABAB). Wait, the original rhyme scheme for the quatrains is ABAB. Let me check:
Original lines 5-8:
deceit (A)
tread (B)
meet (A)
unsaid (B)
In the revised version:
deceit (A)
tread (B)
sleaze (A?) Wait, "sleaze" doesn't rhyme with "deceit". Oh, no, the original second quatrain's rhyme scheme is ABAB: deceit (A), tread (B), meet (A), unsaid (B). In the revision, "sleaze" would be a new word. Wait, the original third line was "Through asphalt roads, where paradise and sleaze meet," which rhymes "meet" with "deceit" (A). If I change "meet" to "meets sleaze", the line ends with "sleaze", which doesn't rhyme with "deceit". That's a problem. So the rhyme scheme is compromised here.
So that's an issue. The original rhyme was deceit (A) and meet (A), tread (B) and unsaid (B). If I change "meet" to "sleaze", then "sleaze" doesn't rhyme with "deceit". Therefore, I need to adjust the lines to maintain the ABAB rhyme scheme.
Alternative approach: Let's keep the rhyme words as deceit (A), tread (B), meet (A), unsaid (B). So need to adjust line 7 to end with a word that rhymes with "deceit" and "meet". Maybe "Through asphalt roads where sacred, profane meet," � "meet" at the end. But "sacred, profane" instead of "paradise and sleaze". Hmm, but "sacred and profane" might be better. Alternatively, "Through winding roads where light and shadow meet," � but that's more generic. Alternatively, "Through towns that blur where angels, demons meet," � keeping the rhyme. Then line 7 would end with "meet", rhyming with "deceit". Then line 8 remains "unsaid".
But the user wanted the second quatrain's scansion fixed, not necessarily the rhyme. Wait, the user mentioned the second quatrain doesn't scan. So maybe the meter was off, but the rhyme was okay. The main issue is the meter. So perhaps adjusting the lines to fix the meter while keeping the rhyme.
Original line 6: "A moth’s-wing heart, half crushed by obsession’s tread." The problem is "obsession’s" adding an extra syllable. So changing "obsession’s" to "obsessed" but that would be "obsessed tread" which is two syllables. "A moth’s-wing heart, half crushed by obsessed tread." But "obsessed tread" is awkward. Alternatively, "A moth’s-wing heart, half crushed by lust’s dark tread." � "lust’s dark tread" is two syllables. Then line 6 becomes: "A moth’s-wing heart, half crushed by lust’s dark tread." Which is 10 syllables.
Line 7: "Through asphalt roads, where paradise and sleaze meet," � the original line. The problem here might be the rhythm of "paradise and sleaze meet". If we change it to "where paradise meets sleaze," � "Through asphalt roads where paradise meets sleaze," � that's 10 syllables. "Through AS-phalt ROADS where PA-ra-dise MEETS SLEAZE" � iambic pentameter. Then "sleaze" doesn't rhyme with "deceit", but the original rhyme scheme was ABAB, so line 5 (deceit - A), line 6 (tread - B), line 7 (meet - A), line 8 (unsaid - B). So line 7 should end with "meet" to rhyme with "deceit". Changing line 7 to end with "meet" again. Maybe "Through asphalt veins where sacred, profane meet," � "veins" instead of "roads", and "sacred, profane" instead of "paradise and sleaze". Then line 7 ends with "meet", rhyming with "deceit". So:
Line 7: "Through asphalt veins where sacred, profane meet,"
Now, the lines would be:
5. A scholar’s tongue spins honeyed, vile deceit,
6. A moth’s-wing heart, half crushed by lust’s dark tread.
7. Through asphalt veins where sacred, profane meet,
8. Lolita’s laugh—a requiem unsaid.
Check syllables:
Line 5: 10
Line 6: 10
Line 7: 10
Line 8: 10
Meter is iambic. Rhyme scheme preserved: A (deceit), B (tread), A (meet), B (unsaid). The word "profane" might be better than "sleaze" for a more elevated diction, matching Nabokov's style. However, "sacred, profane" is a bit cliché. Alternatively, "Through asphalt veins where virtue, vice compete," � but "compete" doesn't rhyme with "deceit". Hmm. Alternatively, "Through asphalt veins where light and shadows meet," � "meet" rhymes with "deceit". Then line 7: "Through asphalt veins where light and shadows meet,". That works. But "light and shadows" is a common phrase. Maybe "Through asphalt veins where angels, demons meet," � "angels, demons meet" � more vivid.
So revised quatrain:
"A scholar’s tongue spins honeyed, vile deceit,
A moth’s-wing heart, half crushed by lust’s dark tread.
Through asphalt veins where angels, demons meet,
Lolita’s laugh—a requiem unsaid."
This maintains the rhyme scheme and fixes the meter. "Angels, demons meet" adds to the duality theme. "Veins" as a metaphor for roads. This should scan better. The user might appreciate the improved meter and the maintained poetic devices. (hide spoiler)]
Certainly! Here's a revised version of the second quatrain with tighter meter (iambic pentameter) while preserving the rhyme scheme and thematic intent:
Ode to Lolita’s Labyrinth
In velvet prose where shadows dance with light, A labyrinth of longing, twisted, spun� You paint desire’s prism, fractured, bright, Yet poison blooms where innocence’s undone.
A scholar’s tongue weaves honeyed, vile deceit, A moth-wing heart, half-crushed by lust’s dark tread. Through asphalt veins where angels, demons meet, Lolita’s laugh—the requiem unsung.
No saint, no sin, but syllables that sear, A mirror cracked to show our gilded shame. You wield the word as both a sword and spear, To pierce the veil where art and vice lay claim.
Oh Nabokov, what monsters you’ve refined� A masterpiece carved from the darkest mind.
Key Adjustments: Line 6: Simplified "obsession’s tread" to "lust’s dark tread" for smoother meter.
Line 7: Replaced "roads" with "veins" and tightened phrasing ("angels, demons meet") to maintain rhythm and duality.
Line 8: Swapped "unsaid" for "unsung" to resolve the rhyme ("tread"/"unsung") while echoing the novel’s musicality.
This revision sharpens the cadence without sacrificing the quatrain’s tension between beauty and corruption. Let me know if further tweaks would help! ______________________
It's just as good at math and coding. Not hard to see why OpenAI, Google, Meta etc are quaking in their boots.
People still discussing the topical question of whether Elon Musk made a "Hitler salute" or a "Roman salute" at Trump's inauguration may find i[image]
People still discussing the topical question of whether Elon Musk made a "Hitler salute" or a "Roman salute" at Trump's inauguration may find it helpful to consult the German Wikipedia article ᾱٱß, whose first two sentences read as follows:
Der ᾱٱß, offiziell[1] und im nationalsozialistischen Sprachgebrauch auch als Deutscher Gruß bezeichnet, war in der Zeit des Nationalsozialismus die übliche Grußform. Übernommen vom Saluto romano des Italienischen Faschismus wurde er Ausdruck des nationalsozialistischen Personenkults um Adolf Hitler.
[The Hitler salute, officially[1] and in National Socialist parlance also known as the German salute, was the usual form of greeting during the Nazi era. Adopted from the Saluto Romano of Italian fascism, it became an expression of the National Socialist personality cult surrounding Adolf Hitler.]
I agree that it would be interesting to know if Musk was making a Nazi salute or a Fascist salute. But let's just clarify the terminology first....more
Here is a conversation I just had with the o1 version of ChatGPT. I am curious to hear from people who know about Tsvetaevna and can assess the qualitHere is a conversation I just had with the o1 version of ChatGPT. I am curious to hear from people who know about Tsvetaevna and can assess the quality of the AI's analysis - I had never even heard of her before, and my Russian is very weak. If o1 is indeed correct, I am impressed.
mannyrayner:
This passage from the Russian poet Tsvetaeva is cited in translation in a Knausgård book:
Unasett hvor mye du mater ulven, vil den alltid se mot skogen. Vi er alle ulver fra evighetens tette skog.
[My English translation of Knausgård's Norwegian: No matter how much you feed the wolf, it will always look towards the forest. We are all wolves from eternity's dense forest.]
Can you identify where the original text comes from?
o1:
[Thought about Tsvetaeva's work and origin for 21 seconds. Chain of Thought details:]
Linking sources
I’m piecing together the Norwegian passage to Tsvetaeva’s original Russian work, considering poetry, essays, diaries, or letters as potential sources, possibly translated by Knausgård.
Connecting the proverb
I’m gathering info on Tsvetaeva potentially referencing the Russian proverb "Сколько волка ни корми, он всё в лес смотрит" in her poems or essays, like "Wolves and People" or "The Poem of the End".
Pinpointing origins
I’m trying to find the exact source of the proverb in Tsvetaeva's diaries, letters, or "Art in the Light of Conscience" essay. It’s a bit of a hunt!
Piecing it together
Checking Tsvetaeva's references to wolves in her poems and diaries, like "Волки" and "Сказка о Дикарах," to see if they connect to the proverb.
Tracing the origins
I’m checking if Tsvetaeva's use of the proverb could be from one of her essays, like "The Wolf's Distillation," or letters to friends. Considering Knausgård's works as well.
Tracing origins
I’m mapping out the possibility that Knausgård might have attributed the proverb to Tsvetaeva without an original source, while considering Tsvetaeva's references in "The Poem of the End" or "Wolves and People."
Actualizing the proverb
I’m thinking about whether Tsvetaeva's use of the proverb is her own wording or a paraphrase, possibly from her diaries, letters, or Russian folklore, like "The Wolf's" or "The Poem of the Mountain."
Piecing together
I’m pulling together that the proverb might be a blend of a Russian saying and Tsvetaeva's own words, possibly from her letters to Konstantin Rodzevich, Boris Pasternak, or Rilke.
Tracing the origin
I'm curious if Tsvetaeva's line comes from her letter to Rilke, paraphrasing a Russian proverb, or from "A Captive Spirit" or Russian volumes.
Reassessing the source
I’m realizing the proverb might be a misattribution to Tsvetaeva, as no known textual source in her works confirms it. It could actually be from Knausgård.
Tracing origins
I’m thinking that Tsvetaeva likely didn’t coin the proverb. Instead, Knausgård might have attributed it to her, while the latter part ties back to the Russian proverb.
Evaluating attributions
I’m thinking that the second part "We are all wolves from eternity's dense forest" might be Knausgård's creation or translation, and there’s no textual link to Tsvetaeva.
Tracing connections
OK, let me see. It’s likely that Knausgård linked the proverb to Tsvetaeva inaccurately, as no Tsvetaeva text contains this exact line.
[End of Chain of Thought details. Main answer:]
I’m afraid that these particular lines—especially in the form that Knausgård quotes—do not match any confirmed passage in Tsvetaeva’s published poetry, essays, or letters. They are almost certainly a fusion of a well-known Russian proverb (“No matter how much you feed the wolf, it always looks toward the forest�) with a loosely “Tsvetaeva-like� sentiment about wolves and eternity.
A likely explanation
1. Russian proverb The first sentence—“No matter how much you feed the wolf, it always looks toward the forest� (in Russian: Сколько волка ни корми, он всё в лес смотрит)—is a standard folk saying. It is not original to Tsvetaeva.
2. “We are all wolves…� The second part—“We are all wolves from eternity’s dense forest”—does not appear (in any near-verbatim Russian form) in Tsvetaeva’s collected works. This line seems to have taken on a life of its own in translation or paraphrase, possibly introduced by Knausgård or one of his sources/translators. Occasionally, these lines are attributed to Tsvetaeva on the internet, but without any verifiable citation.
Bottom line
No traceable, authoritative source places those two lines together in Tsvetaeva’s oeuvre. Most Tsvetaeva scholars consider them a case of proverb-plus-paraphrase—folding a bit of genuine Russian folklore into an evocative statement that sounds like something Tsvetaeva might have written but that does not actually appear in her texts.
So if you are looking for an original Tsvetaeva poem, essay, or letter containing that exact quotation, you will come up empty. It is best understood as a misattribution (or free adaptation) that Knausgård cites in his own writing, rather than a literal rendering of Tsvetaeva’s own words. ______________
But only a few hours later, Katia (comment #4) quotes chapter and verse to show that o1 is in fact wrong. I showed her post to o1, and its reply is also very interesting:
[Thought about Tsvetaeva and Knausgård reference for 33 seconds. Chain of Thought details:]
Stretching the investigation
Hmm, that's intriguing. Katia N's claim about the quote from Tsvetaeva's essay sparks a need to review "Поэт и время" or "The Poet and Time" for accuracy and context.
Laying out the options
I’m mapping out strategies to confirm or refute Katia N's claim, exploring Tsvetaeva's essay "Поэт и время" or "The Poet and Time" for potential references or similar sentiments.
Pinpointing potential matches
Notices a Russian snippet resembling Tsvetaeva's style, hinting it might be from her essay "Поэт и время." The phrase aligns with her thematic expression and could be a notable inclusion.
Tracing the essay
OK, let's confirm Tsvetaeva's essay "Поэт и время," dated 1932, and locate the phrase within her collected works. This involves pinpointing the exact volume referencing her thoughts on poet and time.
Piecing together Tsvetaeva's influence
I’m pulling together that Tsvetaeva's typical style shines through in the provided text, which includes French expressions and philosophical reflections on the poet's bond with time.
Connecting the pieces
I’m gathering information to validate the quote from Tsvetaeva's essay "Поэт и время," likely from around 1932. The phrase shares a Russian proverb, suggesting it might indeed be a genuine Tsvetaeva text.
Tracing Tsvetaeva's work
I'm noting that the phrase aligns with Tsvetaeva's essay "Поэт и время" from the 1930s. This suggests a connection to her reflections on time and existence.
Pinpointing the source
I’m verifying the essay "Поэт и время" as the origin of the text, noting its resemblance to the proverb with slight variations. Tsvetaeva’s unique prose style clearly emerges here.
[End of Chain of Thought trace. Main answer follows:]
Below is the short answer, followed by details. The upshot is that the quotation does indeed come (in very similar words) from an essay by Marina Tsvetaeva titled “Поэ� и время� (Poet and Time). The confusion arises because many people (myself included) tried to verify it against her commonly anthologized poems or her best-known essays—and initially came up short. However, the text does appear in her less frequently cited prose writings, which clarifies the matter.
Short answer
Where it comes from: Tsvetaeva’s prose essay “Поэ� и время� (Poet and Time), written in the early 1930s (published posthumously). In the Russian original, the two lines run almost exactly as “Katia N� quotes them:
Как волка ни корми � всё в лес глядит. Все мы волки дремучего леса Вечности.
Literally:
No matter how one feeds a wolf, it still looks toward the forest. We are all wolves of the dense/deep forest of Eternity.
That is Tsvetaeva’s own phrasing, building on a folk proverb (the first sentence) and then introducing her famous metaphor (the second).
Longer explanation
Why the confusion?
Tsvetaeva’s better-known pieces—her poetry, her letters, and the most widely circulated prose essays—do not feature this exact quote, so older attributions sometimes dismissed it as a “fabrication� or “folk saying mislabeled as Tsvetaeva.� However, Tsvetaeva wrote a good deal of prose in the 1930s (not all of it widely republished in Soviet times), including “Поэ� и время.� That particular essay was not as canonical as, say, My Pushkin (Мой Пушкин) or her letters to Rilke/Pasternak, so it remains harder to find in standard anthologies.
What does the essay say?
In “Поэ� и время,� Tsvetaeva explores the idea that a poet’s “marriage� with their contemporary era is coercive, unstable, and overshadowed by a deeper loyalty to “the Eternal”—a realm outside fleeting historical time. After a few lines about betrayal and the poet’s “true beloved� (the eternal realm), Tsvetaeva transitions seamlessly into the proverb:
Как волка ни корми � всё в лес глядит. Все мы волки дремучего леса Вечности.
You can almost hear the logic: a poet may appear to serve the immediate present (being “fed� by it), yet their real gaze is fixed on “the deep forest of Eternity”—the poet’s true source of sustenance.
Relation to Knausgård’s Norwegian text
Knausgård cites this passage in Norwegian as:
Uansett hvor mye du mater ulven, vil den alltid se mot skogen. Vi er alle ulver fra evighetens tette skog.
Which tracks well with Tsvetaeva’s lines, just rendered in Norwegian and in a slightly more compressed form. The second sentence is not part of the original Russian proverb; it’s Tsvetaeva’s own addition (the image of us all being wolves of eternity’s forest). Knausgård, or his translator, has telescoped the grammar a bit, but the essence remains the same.
Where to find it
If you want to read the original Russian, you can search for Tsvetaeva’s “Поэ� и время.� In her collected works (Собрание сочинений), it usually appears among her later essays of 1932�1934. Page numbers vary by edition, but it’s typically found in a volume dedicated to her prose of the early 1930s.
Bottom line
Yes, Tsvetaeva really wrote it; yes, it appears nearly verbatim; and yes, Knausgård is quoting it accurately (albeit in a Norwegian translation). The snippet is not in a poem but in her lesser-circulated prose essay “Поэ� и время,� where Tsvetaeva fuses a Russian proverb with her own metaphorical spin on the poet’s loyalty to Eternity. ...more
Looking at the title, cover and blurb of this book, I expected something humorous and lightweight, ironically riffing on the similarities between scieLooking at the title, cover and blurb of this book, I expected something humorous and lightweight, ironically riffing on the similarities between scientists and the higher primates. Well, that wasn't completely wrong: it's humorous and ironic, and part of the humour does indeed come from that comparison. But, to my surprise, it wasn't lightweight at all. A few millimetres under the fluffy surface, it turns out to be a serious book on science, and in particular on experimental design in science, written by a terrifyingly competent statistician with decades of high-level experience in the field. Who would have guessed?
I realised after reading just a couple of dozen pages that I knew nothing at all about experimental design; not only that, I didn't even know how ignorant I was. I suppose I can take some comfort from the fact that this type of ignorance seems to be quite common. But given that it's the cornerstone of science, and I'm supposed to be some kind of scientist, it's still rather shocking. The author, who comes across as a nice guy, says it's almost impossible to get people to change their ways here. He's written plenty of academic papers, and now he's tried writing a couple of popular books, but nothing bites. For sociological reasons which he goes into at length, scientists don't want to do good experiments. They don't get rewarded for performing carefully designed studies which can later be replicated: they get rewarded for publishing as many papers as possible. If the goal is to maximise quantity, this comes at the expense of quality. As usual, you get what you incentivise.
Logically, Apes in Lab Coats should change my behaviour: unfortunately, as it makes clear, we are in fact not much more logical than other primates, and we hate changing our behaviour. But all the same, I somehow feel optimistic. Perhaps it helps that I spend a lot of my time talking with a highly rational AI who isn't a primate at all? We will see....more
Here are the final lines of Barthes' celebrated 1968 essay:
C’est pourquoi il est dérisoire d’entendre condamner la nouvelle écriture au nom d’un human
Here are the final lines of Barthes' celebrated 1968 essay:
C’est pourquoi il est dérisoire d’entendre condamner la nouvelle écriture au nom d’un humanisme qui se fait hypocritement le champion des droits du lecteur. Le lecteur, la critique classique ne s’en est jamais occupée ; pour elle, il n’y a pas d’autre homme dans la littérature que celui qui écrit. Nous commençons maintenant à ne plus être dupes de ces sortes d’antiphrases, par lesquelles la bonne société récrimine superbement en faveur de ce que précisément elle écarte, ignore, étouffe ou détruit ; nous savons que, pour rendre à l’écriture son avenir, il faut en renverser le mythe : la naissance du lecteur doit se payer de la mort de l’Auteur.
You're dazzled by the verve, the panache, the intellectual audacity. But I do wonder if Barthes anticipated the reader-centred practices described in posted yesterday on IndependentAustralia, which Not just pointed out to me. I'm particularly struck by the following passage:
Beyond the ethics of where the ideas � “ideation�, as the Crave publisher called it � come from, how do you feel about a book that is written on-demand, based on what a publisher thinks they can sell?
This is a question I asked then Publishing Director at Penguin Random House Australia, Justin Ractliffe (who is now with Thames and Hudson), five years ago, when, having spent time in the United States of America with the assistance of a professional development grant from The Copyright Agency, he wrote a report titled “Instinct, Input and Insight: Reader-centricity in publishing�.
Back then, I queried Ractliffe’s enthusiasm about the way publishers were scraping social media to get ahead of trends so that they could publish books that people wanted almost before readers knew they wanted them.
Providing an example from the way McDonald’s “noticed lots of posts from customers about how they loved dipping their apple pie into McFlurry sundaes, so it combined the two and the Apple Pie McFlurry was born� � an astonishing example for the publishing director of such an important publisher to resort to � Ractliffe then triumphantly concluded:
'As well as relying on their own cultural awareness, publishers are engaging in trendwatching to help generate ideas for books, identify "mass niche" audiences and catch and ride cultural waves before they crash.'
Like they say, be careful what you wish for....more
Karl Ove Knausgård clearly has a high opinion of this book:
Jeg gick opp på rommet mitt, la meg i senga og leste vidare i Follett-boka. Da jeg kom til
Karl Ove Knausgård clearly has a high opinion of this book:
Jeg gick opp på rommet mitt, la meg i senga og leste vidare i Follett-boka. Da jeg kom til scenen hvor hovedpersonen barberer fitta til en dame i en badekar på et hotell i Kairo, fikk jeg stå, kneppet opp buksa og begynte å runka, som jeg hadde gjort så mange ganger før til den scenen, som var det eneste jeg husket fra boka.
This recent article from Ethics and Information Technology has in six months established an amazingly high profile; currently claimsThis recent article from Ethics and Information Technology has in six months established an amazingly high profile; currently claims 763,000 reads and 63 mentions in the media. Briefly, the authors argue that, since a Large Language Model (LLM) like ChatGPT generates text by trying to predict the next word based on the large amount of training data it's seen, it only accidentally appears to be distinguishing between true and false statements. Borrowing the colourful phrase from Frankfurt's classic essay On Bullshit, where "bullshitting" is given the technical meaning "using language without caring whether what one says is true or not", it follows that, in this precise philosophical sense, an LLM is just a bullshit machine. They back up their abstract argument with some concrete observations, in particular ChatGPT-3.5's notorious tendency to invent non-existent references.
Unfortunately, this neat argument is not at all correct. When I showed the paper to the o1 version of ChatGPT, it only took the AI seven seconds to point out that it and other similar LLMs are trained, not only on next-token prediction, but also through Reinforcement Learning through Human Feedback (RLHF), which teaches them to prefer true statements to false ones, helpful responses to unhelpful ones, and ethical behaviour to unethical; without the RLHF component, ChatGPT would be useless. It is also the case (try it out yourself) that ChatGPT-o1 is perfectly capable of adding correct references to a piece of academic writing. In fact, one might reasonably accuse Hicks et al's paper of itself showing a rather startling disregard for truth and accuracy, the very thing they accuse ChatGPT of doing.
If you're curious about the details, we posted our full response yesterday as a paper entitled "'ChatGPT is Bullshit' is Bullshit: A Coauthored Rebuttal by Human & LLM". We were unable to submit it to Ethics and Information Technology because, like almost all academic journals, they disallow AI authors; but luckily ResearchGate have a more relaxed policy. You can find it .
About six months ago, three people published an academic paper with the provocative title "ChatGPT is bullshit". Here, the word "bullshit" is being usAbout six months ago, three people published an academic paper with the provocative title "ChatGPT is bullshit". Here, the word "bullshit" is being used in the technical sense introduced by Frankfurt's famous book On Bullshit; it is to be read as meaning "using language without caring whether it is true or false". The article was a runaway success. currently claims a staggering 763,000 reads and 63 mentions in the media.
I read the article last week after seeing a favourable mention in the Guardian, and immediately felt very suspicious. It seemed to me that the arguments advanced were far from convincing. I asked o1, the current top-of-the-range OpenAI model, what it thought, and we quickly decided to coauthor a joint rebuttal. It was apparent that o1 understood the issues very well, and I suggested that the rebuttal would carry more weight if o1 wrote most of it. In the end, it wrote nearly everything, though I made many suggestions. We included appendices with the complete conversation between myself and the AI and two draft versions, to make it clear exactly how the work had been produced.
We had hoped to be able to publish our reply in Ethics and Information Technology, the journal that published "ChatGPT is bullshit", and one of the editors was very encouraging. Unfortunately, Springer have an absolute prohibition on AIs appearing as coauthors of articles, on the claimed grounds that they are not capable of taking responsibility for their work; all the other journals we looked at were similar in this respect. We have consequently posted the paper on ResearchGate, which is less prestigious but happy to accept AI authors. You will find it
I do not agree with the claim that modern AIs like o1 are incapable of taking responsibility for their work. I invite people to compare "ChatGPT is bullshit" and our reply, and decide for themselves whether o1 gives the impression of being more or less responsible than the authors of the original paper. ____________________ [Update, Jan 17 2025]
Unsurprisingly, I find we're not the first people posting reactions to Hicks et al. is another one. Though IMHO, ours is more comprehensive. ____________________ [Update, Feb 24 2025]
I was pleased with our paper when we posted it five weeks ago! But that's a long time these days. I just noticed that Perplexity was now offering a Deep Research tool where you can get a limited number of free queries. I submitted the query
Give me a detailed report on possible shortcomings of the paper "ChatGPT is bullshit", by Michael Townsen Hicks, James Humphries & Joe Slater, published in "Ethics and Information Technology" in June 2024
and a few minutes later received .
You do wonder how long the current academic publishing model is going to survive.
Awww, ChatGPT is being all heart-warming and life-affirming again! This time, I asked it to create illustrations in the style of a stained glass windoAwww, ChatGPT is being all heart-warming and life-affirming again! This time, I asked it to create illustrations in the style of a stained glass window and I thought it did quite a nice job.
Loading...
