I have decided to work with this image, with a ceaned-up street. The film below
makes the traffic move, but the image size is wrong. Need to look into that...
https://labs.google/fx/tools/flow/shared/video/82558439-ffef-44c7-8930-6cc0a292d72b
Sunday, May 31, 2026
Saturday, May 30, 2026
Anthropic_Go
Anthropic is crushing it with Claude4.8: it can recode an app to the user's language choice
in a breeze...
Intelligence artificielle | Le secret du succès d’Anthropic
* * *
Some progress. There is a new set-up, with the bike already facing the street.
Below, someone starting a motorbike;
I'm having an ongoing problem with continuity. I'm going to need help from Claude AI
to keep my character the same, and maintain a friendly shopping area...
Friday, May 29, 2026
The Project
Anthropic's Claude asked me, at one point, what my video was about. In itself,
it is very simple. A young man on a city street puts on his helmet, gets on a motorbike,
and drives off merging with the traffic.
But our young man is not the works-out-on-protein-powder posterboy of out time.
He is thin, somewhat haggard; his clothes are chic but worn; he has known hunger ans struggle.
In effect, the seed description which made him come alive is from Dostoevsky's
Crime and Punishment(1866) the tortured student who murders an old woman for her money
he desperately needs, but then he is horrified by what he did and leaves the money behind as he
flees. In short, he is one of the most famous figures of World Literature!!
And this is how a Large Language Model would have him appear, today.(Chat GPT)
Dostoevsky himself studied to be an engineer, but ended up writing. Clearly, a gifted and
somewhat tortured person in his own right.
So if we add the subtext to the video's narrative, we end up with something more interesting.
The work is about how to use the software, but also about ctrating template descriptions that
are structured and sequenced to avoid weirdness and hallucinations. Still not sure what the
subject is driving from, or toward. Claude seemed to have views on that itself!!
Rider_Project
Hard at work on my rider project. The stakes are high!!
https://claude.ai/share/8bcccfe5-780b-44c7-92ae-2a524b1eedca
* * *
* * *
🎬 TAKE 1 — @rider puts on his helmet and approaches the bike
Character:
The same male rider as in the reference image — identical face, body proportions, hairstyle, clothing, leather jacket, jeans, boots, and overall silhouette. No changes between takes.
Motorcycle:
The same motorcycle as in the reference image — identical model, color, shape, accessories, and details. No changes between takes.
Action:
@rider stands beside the parked motorcycle on the cobblestone sidewalk.
He lifts the black helmet with both hands, brings it up to his head, and smoothly puts it on.
He adjusts the chin strap with a small, natural motion.
After securing the helmet, and leaving it on throughout, he takes two or three calm steps toward the motorcycle, approaching it from the left side.
He grabs the motorbike by the handles and turns the front toward the street.
He stops in position beside the seat, ready to swing his right leg over in the next take.
Camera:
Medium shot, waist‑up during the helmet action, then widening to a medium‑full shot as he steps toward the bike.
Stable, cinematic, naturalistic camera movement.
Continuity instruction:
The rider and the motorcycle must remain identical in appearance, clothing, proportions, and details across all takes.
No changes to lighting, environment, or props unless explicitly stated in later takes.
https://labs.google/fx/tools/flow/shared/video/156430b6-bd69-49a2-a6f3-f32e2dee8e63
Needs more work, but that will have to wait till tomorrow!!
Thursday, May 28, 2026
Graphite
Copilot:
Canada; PM Carney recently made an announcement about graphite mining for export. Where is the graphite mine located?
The graphite mine referenced in Prime Minister Carney’s announcement is the Nouveau Monde Graphite (NMG) Matawinie Mine, located in Saint‑Michel‑des‑Saints, Québec, about 120 km north of Montréal.
Carney says construction underway on G7’s largest graphite project - Canadian Mining Journal
source: Wikipedia
Cinematic
Created a motorbike rider, from a Dostoevsky description: “He was a young man of average
height, slim, with dark eyes and a pale, haggard face.” From Crime and Punishment...
Produced a handsome young man, although I found him a little too sickly. Asked for the
circles around his yes be removed. It is a minor thing, but makes a big difference:
* * *
Met with various difficulties. I wanted @Rider to get on the bike and ride off in the traffic
but it never worked out on one description. Asked Gemini for help.
* * *
Will need to write clear descriptions, and remove frames that are impossible. Just waiting
for more credits to my account, at this point. I have a sense this might all work out easily
enough. Interestingly, I will need to become more 'cinematic' in my thinking, and less literary!
* * *
* * *
Advice form Anthropic's Claude model:
https://claude.ai/share/274549a0-224d-498d-8ef2-fc0233fee914
Wednesday, May 27, 2026
Temperature_LLM
Copilot:
If I ask a large language model - such as yourself - the same question over and over, I will not get the same precise answer every time ( other than for a math problem). Is this because the Internet is always changing? Are there other factors, making for this? Has anyone written on this issue. I am trying to build scenarios in Google Flow, and keep running into this issue...
You’re noticing something very real — and it has nothing to do with the internet changing under your feet.
Even if I were completely cut off from the web, you would still get different answers to the same question.
Let me give you the clear, structured explanation you deserve, especially since you’re trying to build consistent scenarios in Google Flow.
⭐ Short answer
Large language models (LLMs) don’t give identical answers because they are probabilistic, not deterministic.
They generate text by sampling from many possible next‑word options.
Even with the same question, the model can choose a different path each time.
This is true even if:
the internet hasn’t changed
the model hasn’t updated
the question is identical
🧠 Why LLMs vary their answers
Here are the actual factors, in order of importance.
1. Sampling randomness (the biggest factor)
LLMs don’t “retrieve” an answer — they generate one word at a time.
At each step, there are many plausible next words.
The model picks among them using probabilities.
This means:
small randomness → slightly different phrasing
larger randomness → noticeably different content
This is controlled by parameters like:
temperature
top‑p
top‑k
Most consumer interfaces (including this one) use a non‑zero temperature, so variation is expected.
2. Hidden state sensitivity
Even tiny differences in:
punctuation
whitespace
timing
internal tokenization
can nudge the model down a different path.
LLMs are like chaotic systems:
small input differences → large output differences.
3. Context window effects
If you ask the same question in a long conversation, the model’s internal context shifts.
Even if you repeat the question verbatim, the surrounding conversation influences the answer.
4. Model updates (rare, but real)
Sometimes the underlying model is updated.
But this is not the main reason for everyday variation.
5. Your phrasing triggers different “latent patterns”
LLMs store patterns, not facts.
A single question can map to multiple internal patterns, each producing a different style of answer.
📚 Has anyone written about this?
Yes — and the literature is surprisingly rich.
Here are the most relevant categories (in plain language):
Academic papers
“Language Models are Few-Shot Learners” (Brown et al., 2020) — explains probabilistic sampling.
“On the Dangers of Stochastic Parrots” (Bender et al., 2021) — discusses variability and unpredictability.
“Temperature and Diversity in Text Generation” — many papers analyze how randomness affects output.
Industry explanations
OpenAI, Anthropic, Google DeepMind, and Microsoft have all published blog posts explaining:
sampling
randomness
prompt sensitivity
determinism vs. creativity
Developer documentation
Google Flow documentation
OpenAI API docs
Anthropic Claude API docs
All of them explicitly note that LLMs are nondeterministic by default.
If you want, I can pull together a curated reading list with summaries.
🎬 Why this matters for Google Flow scenarios
You’re running into the same issue every creator hits:
Flow’s video generation is also nondeterministic.
Even if you:
reuse the same prompt
reuse the same reference image
reuse the same seed
Flow still introduces variation because:
the model is generative
the diffusion process has randomness
the “interpretation” of your prompt can shift slightly
This is why creators use:
reference images
identity blocks
style anchors
shot templates
These reduce variation but never eliminate it.
If you want, I can help you build a Flow‑optimized prompt template that minimizes drift.
🧩 Practical takeaway for your scenario building
If you need repeatable, stable outputs, you must:
Fix the randomness
Some systems allow setting temperature = 0 (Flow does not expose this yet).
Use a stable “identity block”
For your cat characters, this is essential.
Use consistent structure
Same order
Same phrasing
Same camera language
Same style description
Avoid open‑ended phrasing
“A cozy scene” → too many interpretations
“A cozy scene with warm tungsten lighting, 35mm lens, static camera” → much more stable
Use reference images every time
This is the strongest stabilizer Flow currently offers.
If you want, I can help you with:
A Flow scenario template that minimizes variation
A stable character identity block for your two cats
A shot‑by‑shot structure that keeps consistency across scenes
A prompt debugging workflow to reduce drift
Just tell me what kind of scenario you’re building, and I’ll tailor it to your style.
What do you mean by temperature?
Temperature is one of the core knobs that controls how creative or stable a language model is.
Think of it as the “randomness dial” in the model’s brain.
Here’s the clean, intuitive version:
⭐ The essence
Temperature = how adventurous the model is when choosing the next word.
Low temperature (0–0.2) → predictable, stable, almost identical answers
Medium temperature (0.5–0.7) → balanced, natural variation
High temperature (0.8–1.5) → creative, surprising, sometimes chaotic
Most consumer interfaces use a medium temperature, which is why you see variation even when you repeat the same question.
🧠 What temperature actually does
When generating text, the model looks at a probability distribution for the next word.
Example (simplified):
With temperature = 0, the model always picks the highest probability word (“is”).
With temperature = 1, the model samples from the whole distribution.
With temperature > 1, the distribution gets flattened, making rare words more likely.
This is why answers drift.
🎨 A metaphor you’ll appreciate
Since you’re building scenarios in Google Flow:
Temperature = how much the actor improvises.
Low temperature → the actor sticks to the script.
High temperature → the actor riffs, adds lines, changes tone.
Flow’s video generation has a similar concept internally, even if it doesn’t expose the setting.
📚 Has this been written about?
Yes — extensively.
The concept comes from statistical physics and was adopted in machine learning decades ago.
Key sources include:
Research on softmax sampling
Papers on temperature scaling in neural networks
OpenAI and DeepMind documentation on sampling parameters
Academic work on diversity in text generation
If you want, I can give you a curated list with short summaries.
🎬 Why this matters for your Google Flow scenarios
If you want consistent characters, consistent tone, consistent shots, you need to reduce the model’s “freedom to wander.”
Since Flow doesn’t let you set temperature directly, you control consistency through:
rigid prompt structure
fixed identity blocks
reference images
explicit camera and lighting instructions
avoiding open‑ended adjectives (“cozy”, “dreamy”, “magical”) unless you define them
I can help you build a Flow prompt template that behaves like a low‑temperature model — even though Flow itself is stochastic.
If you want, tell me what kind of scenario you’re building, and I’ll shape a stable template around it.
What are Top-p and Top_k?
Top‑p and top‑k are the other two big “randomness controls” used in language models.
If temperature is the heat, then top‑p and top‑k are the shape of the menu the model is allowed to choose from.
Let me give you the clean, intuitive version — the one that actually helps when you’re building scenarios in Google Flow.
⭐ The essence
Top‑k = limit how many possible next words the model is allowed to consider.
Top‑p = limit the model to only the most probable words whose combined probability reaches a threshold.
Both are ways of restricting the model’s freedom, which reduces variation.
🧠 Top‑k (the “shortlist” method)
Top‑k says:
“Only consider the top k most likely next words. Ignore the rest.”
Example:
If k = 5, the model only chooses among the 5 most probable next words.
Effects:
Small k (1–20) → very stable, repetitive, predictable
Medium k (40–100) → natural variation
Large k (200–1000) → wild, creative, sometimes incoherent
Metaphor:
Top‑k is like telling an actor:
“You may only choose from the first 5 lines on the script page.”
🧠 Top‑p (the “probability mass” method)
Top‑p (also called nucleus sampling) says:
“Only consider the smallest set of words whose combined probability is at least p.”
Example:
If p = 0.9, the model includes words until their probabilities add up to 90%.
This means:
If the distribution is sharp → maybe only 2–3 words are included
If the distribution is flat → maybe 20–30 words are included
Effects:
Small p (0.1–0.3) → extremely deterministic
Medium p (0.5–0.9) → balanced, natural
High p (0.95–1.0) → creative, surprising
Metaphor:
Top‑p is like telling an actor:
“Choose from the lines that cover 90% of the director’s preferred options.”
🎨 How these interact with temperature
Temperature changes how bold the model is.
Top‑p and top‑k change how many options the model sees.
Together, they shape:
creativity
consistency
drift
stability
Most consumer interfaces (including this one) use:
temperature > 0
top‑p around 0.9
top‑k around 40–100
This is why answers vary even when you repeat the same question.
🎬 Why this matters for your Google Flow scenarios
Flow does not expose temperature, top‑p, or top‑k.
But the underlying video model still uses them internally.
This is why:
characters drift
lighting changes
style shifts
motion varies
the same prompt produces different shots
To simulate “low temperature / low top‑p / low top‑k” behavior in Flow, you must:
use rigid structure
use reference images
define identity blocks
avoid open‑ended adjectives
specify camera, lighting, and composition
keep prompts consistent across shots
I can help you build a Flow prompt template that behaves like a low‑temperature model — meaning stable, repeatable, consistent.
If you want, I can do one of these next:
Build a stable identity block for your two cats
Create a shot‑by‑shot Flow template that minimizes drift
Show you a side‑by‑side comparison of temperature vs top‑p vs top‑k
Help you debug a specific Flow prompt that keeps changing
Which direction do you want to go?
Test
Learning to work with Flow, the hard way. Created a character with a seed description from
Anne of Green Gables: “She was a slim girl, with eager, luminous eyes and a face as alive as a spring
morning.” And put her in an urban setting. I then asked that she be made ten years older, in an
office setting and glamorous. Checked if she could be made to be seen in profile, and that
worked.
Uploaded that girl picture as a resource, and tried to make a video of her turning in profile.
Ended up with a totally different girl and background.
Ooops, used up 30 credits!
Will try again tomorrow, following the advice Gemini just gave me. We'll see!!
https://gemini.google.com/share/ae78a7276fac
* * *
Tuesday, May 26, 2026
Permanent_Base
Big announcement from NASA today: Blue Origin will furnish the
Lunar lander, ast hree missions are planned to set up what amounts to
a large city area ner the South pole of the Moon.
https://www.thesun.co.uk/news/39226010/nasa-reveals-plans-permanent-moon-base/
Moving_On
Not sure to what extent building tiny appartments would make sense in Canada. Our
living spaces have to protect us against extreme cold and appreciable hot. But what
I did appreciate about this video is the
notion that housing needs change over a person's
life, and that's just fine.
Been running into, increasingly, older couples who seem on the verge of meltdown, and the
husband arguing he needs a younger spouse.
Maybe, just maybe, if the couple sold the house and moved into an age-appropriate appartment,
things would play out much differently.
Mowing the lawn can become a formidable task, and earlier
than one expects. Whereas living in a small space with communal amenities might bring
back a great deal of joie de vivre.
Just saying...
Monday, May 25, 2026
Complicated
No sooner up with my cup of Joe, I just spent 40 credits adding a male ginger cat
to my Flow project.
https://labs.google/fx/tools/flow/shared/video/3f2c13c1-3058-4d61-adf2-b4c321257f34
Asked Gemini about how to create a new scene for tomorrow:
https://gemini.google.com/share/23501781fa61
This could get complicated...
* * *
Some resources:
From the three dots...
* * *
The male cat looks like a stalker. I should not have specified 'taller'!!
Maybe tomorrow I will revert back to the image without the male, and have him appear
as a friendly drive in the car sequence.
Sunday, May 24, 2026
Hottie
Looks like I have credits at my disposal.
* * *
Asked that the kitten's mouth be closed, and that we add pink lipstick. That cost me 40 credits!
Saturday, May 23, 2026
Fifty_Credits
Look like there was some confusion on my part about Google Flow credits. one gets
50 daily credits on the Free Tier, with renewal at the time of day of one's first 'generate'.
And putting something in the trash does not regenerate credits. So I should be at 50 later today:
https://gemini.google.com/share/8947f2566e53
* * *
It took a bit of work, bit I now have an 8 second long walking kitten in a bikini. Need
to find a musical track to upload as background music...
Could have kept going to add lipstick and make-up!!
* * *
Friday, May 22, 2026
FLOW
Been thinking this might be a good time to learn to create videos with AI.
And, by all accounts, the tools provided by Gemini are top tier.:
20+ Google AI Products (March 2026) Latest Breakthroughs & Updates
* * *
Below, a how-to on Flow (which is the interface). The person here wants
quality and is opting for one month free with 'Advanced' which then turns into
a subscription;
* * *
The FREE route:
Copilot:
What is Flow; is there a charge to use it?
Flow is Google’s new creative interface for generating videos with Veo 3 — think of it as the workspace where you write prompts, upload images, and assemble scenes. It’s not a model itself; it’s the platform that lets you use the model.
Let me give you the clearest, most practical explanation so you immediately understand what it is and whether it costs anything.
🌊 What Flow is
Flow is:
Google’s official interface for Veo 3
A web‑based creative canvas where you can:
write prompts
generate videos
chain scenes together
add audio
storyboard
remix clips
extend shots
A place where you can combine text → video, image → video, and video → video tools in one timeline‑like environment.
It’s basically Google’s version of Runway or Pika — but built around Veo.
💰 Is Flow free?
Yes — Flow currently has a free tier.
You can sign in with a Google account and generate Veo 3 videos at no cost.
But there are important nuances:
✔ What’s free
Access to Veo 3.1 Fast
Short video generation
Image → video
Video → video
Storyboard mode
Audio generation
Editing tools
⚠ What is not guaranteed to stay free
Google has not committed to Flow being permanently free.
Historically, Google:
launches creative tools with free access
then introduces quotas or paid tiers once usage grows
So Flow is free right now, but it is almost certainly a freemium product in early rollout.
🧠 Why is it free right now?
Because Google is:
competing with Runway, Pika, Luma, and OpenAI
trying to get creators to adopt Veo
gathering usage data to improve the model
building an ecosystem before monetizing
This is the same pattern as:
Google Colab → free → paid tiers
Bard → free → Gemini Advanced
Imagen → free → limited → paid
Flow will likely follow the same trajectory.
🎬 What Flow looks like when you use it
You get:
a left panel for prompts
a center panel for video previews
a timeline‑like strip for scenes
buttons for “Extend”, “Remix”, “Regenerate”, “Add Audio”
a library of your generated clips
It’s extremely simple — more like a creative notebook than a full editor.
If you want, I can show you:
how to write Flow‑optimized prompts
how to chain scenes into a mini‑film
how to get Veo to produce elegant, soft, cozy, Québec‑style visuals
how to avoid Veo’s uncanny valley issues with humans
Just tell me the vibe you want to explore next.
But Veo 3 itself is only free for one month!!??
Ah — I see exactly where the confusion comes from, Louise.
Let me untangle it cleanly, because Veo 3 itself is not “only free for one month.” What is limited is Gemini Advanced, not Veo.
Here’s the real situation, in plain terms.
🌟 Veo 3 is free. Gemini Advanced is not.
Google gives you two different ways to use Veo:
1. Flow (free)
You can use Veo 3.1 Fast
No subscription required
No one‑month limit
You just sign in with a Google account
This is what most creators use
Flow = free access to Veo.
2. Gemini Advanced (paid after 1 month)
This is where the “one month free” message comes from.
Gemini Advanced is Google’s premium chatbot subscription, and it includes:
Gemini 2.0 Flash Thinking
Gemini 2.0 Pro
NotebookLM Pro
Extra features
And also access to Veo inside the Gemini chat interface
But this is optional.
You do not need Gemini Advanced to use Veo.
🧩 Why Google does this
Google is splitting the ecosystem:
✔ Flow = creative workspace → free
✔ Gemini Advanced = premium AI assistant → paid
Both can access Veo, but only one is tied to a subscription.
🧘♀️ So what does this mean for you?
You can continue generating Veo 3 videos without paying, as long as you use:
Flow
Vivideo
Pika Labs (when they integrate Veo)
Any third‑party platform using Veo 3.1 Fast
The “one month free” message is not about Veo — it’s about Gemini Advanced, which is optional.
If you want, I can show you:
how to open Flow directly
how to avoid the Gemini paywall
how to generate Veo videos without ever touching the subscription page
or how to compare Flow vs Vivideo for your style (cozy, elegant, soft, Québec‑light scenes)
Just tell me what you want to do next.
I would like to use Flow - and presumably Veo 3 - without ever paying anything...
You can use Flow + Veo 3 without paying — you just need to avoid the one place where Google tries to funnel you into a subscription.
Let me lay it out clearly and calmly so you feel in control.
🌼 Yes, Louise — you can use Flow + Veo 3 for free.
There is no requirement to subscribe to Gemini Advanced, and there is no one‑month countdown attached to Flow.
The confusion comes from this:
✔ Gemini Advanced = paid after 1 month
✔ Flow = free
✔ Veo 3.1 Fast inside Flow = free
Google mixes the branding, so it looks like Veo is behind a paywall, but it isn’t.
🌊 How to use Flow without ever paying
Here’s the safe path:
1. Go to Flow directly
You must open Flow from its own URL, not from Gemini.
2. Sign in with your Google account
This is free.
3. Ignore anything that says “Gemini Advanced”
That is the only part that costs money.
4. Use Veo 3.1 Fast inside Flow
This is the free model.
It is the same model Vivideo uses.
5. Generate videos normally
You can:
write prompts
upload images
extend clips
chain scenes
add audio
remix
All without paying.
🧘♀️ Why this works
Google is running two separate products:
Flow
A creative workspace.
Free.
No monthly limit.
Uses Veo 3.1 Fast.
Gemini Advanced
A premium AI assistant.
Paid after 1 month.
Includes Veo access inside the chat, but that’s optional.
You don’t need Gemini Advanced to use Veo.
🌱 What you don’t get for free
Just so you’re fully informed:
Veo 3.1 Pro (the highest‑quality version)
Longer videos
Higher resolution
Priority compute
These are likely to become paid features later, but the Fast model is free and very capable.
🌟 Bottom line
You can use Flow + Veo 3.1 Fast indefinitely without paying anything.
You only pay if you choose to upgrade to Gemini Advanced — and you don’t need that for video generation.
If you want, I can walk you through:
how to open Flow safely without hitting the paywall
how to set up your first project
how to write prompts that give you elegant, soft, cozy, Québec‑light visuals
how to avoid the uncanny valley with humans
Just tell me what you’d like next.
* * *
Beginner how-t0:
* * *
Credits explained:
Thursday, May 21, 2026
Desperate_for_Space
Did Google Chrome secretly install a 4GB AI file on your computer? Here’s what you need to know
In my case, yes, they did!
https://gemini.google.com/share/f39b769d5a97
I used the check-for-yourelf protocol:
* * *
Not deleting it, for the moment. If I am ever desperate for space....
Wednesday, May 20, 2026
Dr_Strange
Watched Dr Strange 1 and Dr Strange 2 in the last few days. I like Dr Strange as a character
and appreciate the actors and storyline, but the special effects are overwhelming. I could not
watch either film for long at a time. And the names are other-wordly and impossible to keep
straight.
The heartbreak of childlessness of the 'badie' in Strange 2 was tragic, not comedic.
Why do the inhabitants of other universes have to be such monsters. Yes, it is the Marvel
signature but could we get a break from all that?
* * *
Copilot:
Dr Strange 2 warned about 'product placement'; I didn't catch it in the film...
Ah, Louise — that “product placement” warning in Doctor Strange in the Multiverse of Madness is one of those odd little Marvel quirks that confused a lot of people. You didn’t miss anything. The film barely contains any noticeable product placement at all.
Here’s what’s going on.
🎬 Why the warning appears even though the movie has almost no product placement
Marvel Studios (and Disney more broadly) often uses a standardized legal disclaimer across multiple films, even when the specific movie doesn’t actually feature obvious branded products. It’s essentially a blanket clause that covers:
Background props that might incidentally show a brand
Cars, phones, or storefronts that appear briefly
Licensing agreements that aren’t obvious to viewers
International versions where different edits may include or exclude certain shots
In other words:
The warning is there to protect the studio, not because the film is full of ads.
🧭 So what was the product placement in Doctor Strange 2?
Very little — and none of it is foregrounded.
The only things that might qualify are:
1. Christine’s wedding scene (cars, venue branding)
Some real-world car models and hotel décor appear in the background. Nothing is highlighted.
2. Wanda’s suburban home (generic household items)
These are mostly unbranded or too generic to count as deliberate placement.
3. Kamar-Taj battle gear
Some of the props are made by real-world manufacturers, but the branding isn’t visible.
4. America Chavez’s jacket patches
These are fictional symbols, not brands.
5. Phones and computers
Marvel often uses non-branded or custom UI to avoid licensing issues — so even here, there’s no Apple/Samsung/Lenovo placement.
🧩 Why the warning feels misleading
Because Doctor Strange 2 is visually chaotic and stylized, the viewer’s attention is on:
spells
portals
multiverse jumps
Wanda’s arc
…not on any real-world products.
So the warning feels out of place — almost like a leftover tag from a template.
And honestly?
It is basically a leftover tag from a template.
🌟 Bottom line
You didn’t miss anything.
The film contains almost no visible product placement, and the warning is just a standard legal precaution Marvel attaches to many releases.
If you’re curious, I can also break down:
Why some Marvel films do have heavy product placement
Which MCU movies are the worst offenders
How Disney handles brand licensing in multiverse-heavy films
Just tell me what direction you want to explore.
* * *
Found this very interesting piece on Neanderthal Dentistry on the New York Times.One brave soul went through a root canal without anesthesia. Kudos to doctor and patient
from the 21st Century.
Yes, we should all take note. A no sugar diet made all the difference to pre-agriculture
oral health!!
Tuesday, May 19, 2026
Inflation_Watching
Been reading on inflation this morning; in Quebec - if we ignore fuels - it is at 2%
year over year, the Cinderella point. And we are told the government's temporary
suspension of the excise tax was a good decision. Hoorah for the Federal Government.
We may be all speaking Chinese before the end of this cycle, but our finances will be
in order.
Copilot:
Canada; Translate 'taxe d'accise'. This tax on gas was recently abolished. By how much?
Short answer:
“Taxe d’accise” = excise tax.
Canada’s federal excise tax on gasoline (10¢/L) and diesel (4¢/L) was temporarily abolished, reducing pump prices by exactly those amounts.
🇨🇦 What was abolished, and by how much?
The federal government suspended the fuel excise tax from 20 April to 7 September 2026.
This removed:
10 cents per litre on gasoline
4 cents per litre on diesel
4 cents per litre on aviation fuel
This is a full suspension of the excise tax — not the carbon tax, not provincial fuel taxes.
📌 Practical impact for you in Québec (St‑Jean‑sur‑Richelieu)
Every litre of gasoline became 10¢ cheaper at the pump.
For a typical 50‑L fill‑up: $5 saved.
Weekly fill‑ups over the 20‑week suspension: ~$100 saved.
Follow‑up question for you
Do you want me to calculate your exact savings based on how often you fill up and the size of your tank?
* * *
A regular snoop, I also follow the rent situation in New York City. On the internet, this
often comes down to smaller and weidly arranged expensive appartments with shining
Monster appliances. It is a known fact that New York is the hotel capital of the world, and
sees more influx of business conferences than anywhere. The demand to live there sees no
bounds!!
Monday, May 18, 2026
Easy_Cakey
Queen Victoria reportedly ate a piece of Victoria Sponge every afternoon
with tea. And I'm not averse to a slice of comfort myself. Had a talk with Copilot
about what might be easy to make...
Is there some kind of cake I can make that uses a lot of milk; I have eggs, flour, sugar, vanilla...
Absolutely — if what you want is a cake that drinks milk like a sponge, you have several excellent options with the ingredients you already have. And since you’re in Québec, these are all very “comforting, cozy, dimanche après‑midi” desserts.
🥛 1. Gâteau au lait chaud (Hot Milk Cake)
Uses a LOT of milk in the batter and gives you a soft, velvety crumb.
Quick recipe (your ingredients only)
Ingredients
2 cups milk
4 eggs
2 cups sugar
2 cups flour
2 tsp vanilla
2 tsp baking powder
½ cup butter (if you have it; if not, I can adapt)
Method
Heat the milk + butter until hot (not boiling).
Beat eggs + sugar until pale and thick.
Add vanilla.
Add flour + baking powder.
Pour the hot milk into the batter and mix.
Bake at 350°F / 180°C for ~30–35 min.
Texture: soft, moist, nostalgic, perfect with berries or maple syrup.
🥛 2. Gâteau renversé au lait (Milk‑Soaked Cake)
This is like a simplified tres leches but using only regular milk.
How it works
Bake a simple vanilla cake (flour, eggs, sugar, vanilla).
Warm 2–3 cups of milk with a bit of sugar + vanilla.
Poke holes in the cake and pour the warm milk over it.
Chill 1–2 hours.
Texture: ultra‑moist, pudding‑like, but still sliceable.
🥛 3. Far Breton‑style “Milk Cake” (no prunes version)
A rustic French custard‑cake that uses a huge amount of milk.
Proportions
4 cups milk
1 cup flour
½–¾ cup sugar
4 eggs
Vanilla
Mix everything, pour into a buttered dish, bake 45–55 min at 375°F.
It sets into a silky, custardy slab — almost like a firmer clafoutis.
🥛 4. Clafoutis (but as a “milk‑heavy” version)
If you have fruit (fresh, frozen, canned), this is perfect.
Base batter
2 cups milk
3–4 eggs
½ cup sugar
¾ cup flour
Vanilla
Fruit of your choice
Pour over fruit, bake 35–40 min.
Light, custardy, elegant.
⭐ Recommendation for today
Given your pantry and your love of soft, drapey, comforting textures, the Hot Milk Cake or the Milk‑Soaked Cake will feel like a warm hug — and they use the most milk.
If you tell me how much milk you want to use up (1 L? 2 L?), I can tailor the recipe so you use exactly the amount you want.
Today
Just made the 1)-gateau au lait with 2% milk. What is the calorie count on that cake?
Here’s a clear, realistic calorie estimate for the gâteau au lait chaud you just made, based on the classic proportions you used.
I’ll calculate it ingredient by ingredient, then give you:
total calories for the whole cake
calories per slice (for different slice sizes)
⭐ Calorie Breakdown (Hot Milk Cake, 1 full cake)
Ingredients you used
2 cups 2% milk → ~244 kcal
4 large eggs → ~280 kcal
2 cups sugar → ~1548 kcal
2 cups all‑purpose flour → ~910 kcal
½ cup butter (if you used it) → ~814 kcal
Vanilla → negligible
🧮 Total Calories
If you used butter (classic recipe)
Total = ~3,796 calories for the whole cake
If you did not use butter
Total = ~2,982 calories for the whole cake
🍰 Calories per slice
With butter
1/12 of the cake → ~316 calories
1/10 of the cake → ~380 calories
1/8 of the cake → ~475 calories
Without butter
1/12 → ~249 calories
1/10 → ~298 calories
1/8 → ~373 calories
⭐ Quick takeaway
Your cake is roughly 300–475 calories per slice, depending on size and whether butter was included.
If you tell me whether you used the butter and how many slices you cut, I’ll give you the exact number for your cake.
I used butter; and justement, I'm wondering what a human-scale serving of that cake might be. Because, you know, when it comes to cake...
Ah, oui — the eternal question: what is a “human‑scale” serving of cake, especially when the cake is soft, warm, and smells like childhood comfort. Humans are notoriously unreliable around cake. You’re in good company.
Let’s anchor this in something realistic, physiological, and a bit compassionate.
🍰 What counts as a “human‑scale” serving of hot milk cake?
This cake is light, tender, and not overly rich, so the natural serving size tends to drift upward. But here’s how it breaks down in real life:
1. The sensible, nutrition‑label serving
1/12 of the cake
≈ 316 calories
This is the “I’m being reasonable” slice. It’s modest, satisfying, and doesn’t feel stingy.
2. The normal human serving
1/10 of the cake
≈ 380 calories
This is the slice most people actually cut without thinking. It feels like a real dessert, not a sample.
3. The honest, cozy, rainy‑day Québec serving
1/8 of the cake
≈ 475 calories
This is the “I baked this for comfort and I’m going to enjoy it” slice. Still absolutely fine in the context of a normal day.
4. The “I just made it and it’s warm” serving
1/6 of the cake
≈ 630 calories
This is the slice that happens when the cake is still warm, smells like vanilla, and you’re standing at the counter with a fork. Entirely human.
⭐ My recommendation for you, Louise
Given your love of soft, drapey, comforting textures and the fact that this cake is basically a warm hug in edible form, a 1/10 or 1/8 slice is the most “human‑scale” for pleasure and satiety.
It’s enough to feel cozy and satisfied without tipping into “I regret nothing but I need a nap.”
If you want, I can also give you:
calories for your exact slice if you tell me how big it was
a lighter or richer variation for next time
a way to turn leftovers into a milk‑soaked dessert (dangerously good)
Just tell me what direction you want to go.
* * *
On 35 minutes, the cake has risen, is starting to get brown but is not cooked properly
in the precise middle. I returned it to the oven, put foil on top and gave it 15 minutes.
The result was perfect, nothing runny, but the bottom part had a texture resembling clafoutis.
Great with strawberry Jell-O and Cool Whip. Queen V might well have approuved!
* * *
found this cake on the Web. It is a Depression-era goodie. Same recipe and instructions,
only it uses 1 cup of milk instead of two:
Mine is still good, but this one looks perfect!
Holiday
It's a holiday for many. I am pleased to share a little breakthrough
I have just had with the problem of opening Pop bottles.
Because, yes, when the weather turns warms, I bring home large bottles of Pop
to drink over ice on hot days. And pretty darn often, there is a batlle to initially open
the said bottle... Well, not anymore.
So there I was battling a 2liter Diet Tonic, having just borrowed the large elastic from the
frozen peas bag in the freezer compartment. No matter how i forced things, nothing was
moving. Just before taking the deadliest knife in the kitchen out of its sheet to knock out the cap,
I tried one last time the nice way. And just squeezed the cap as well as turning it. Psitttt! like
a charm. The bottle was open without much force at all.
A lifetime of bottle caps, and now I just know how!!
* * *
Sunday, May 17, 2026
Mexico/Paris
Chatted with Copilot about why smexico and Paris hae such difficulties:
Mexico City’s car‑pollution problem is strongly linked to fuel‑quality standards that lag behind those in the U.S. and Canada, especially regarding sulfur content in gasoline and outdated national emissions regulations. This combination leads to vehicles emitting far more pollutants than similar vehicles elsewhere in North America.
Core issue: Mexico’s fuel and emissions standards are older and less strict
The key difference is that Mexico’s national vehicle‑emission standards were last updated in 2005, while the U.S. and Canada have repeatedly tightened theirs over the past two decades. As a result:
Gasoline in Mexico often contains higher sulfur levels, which directly increases emissions of hydrocarbons (HC), nitrogen oxides (NO), and carbon monoxide (CO).
High sulfur fuel also degrades catalytic converters, making even newer cars pollute more.
Hydrocarbon emissions from new 2022 Mexican vehicles are comparable to U.S. vehicles from 20+ years ago, showing how fuel quality and standards affect real‑world performance.
🚕 Why Mexico City is hit especially hard
Mexico City’s pollution problem is amplified by:
A large fleet of older vehicles, which are more sensitive to poor fuel quality.
Taxis emitting 2.7–3.9× more pollutants than non‑taxi light‑duty vehicles.
Higher altitude, which affects combustion efficiency and can worsen emissions (noted in broader Mexican emissions studies).
Even when comparing similar model years, vehicles in Mexico City emit significantly more pollutants than those in Colorado, Virginia, or Alberta.
...
Yes — Paris has a vehicle‑pollution problem, but the cause is very different from Mexico City’s. Paris does not suffer from poor fuel‑refining standards. Instead, its pollution comes from diesel history, urban density, and traffic volume, not from substandard gasoline or sulfur levels.
Below is the clear comparison.
🧭 Short answer
Paris does not have Mexico City’s fuel‑quality problem.
Its pollution comes from decades of diesel dominance, high traffic density, and urban geography, not from inferior refining standards.
Mexico City’s problem is tied to outdated emissions standards and higher‑sulfur gasoline.
🌍 Why Paris struggles with air pollution
Paris has long battled smog episodes, but the drivers are:
1. A legacy of diesel vehicles
For decades, France incentivized diesel cars with tax breaks. By the 2010s, over 60% of cars in France were diesel. Diesel engines emit:
NOₓ (nitrogen oxides) → major contributor to smog and respiratory illness
Fine particulates (PM2.5) → harmful to lungs and cardiovascular health
Even modern diesel engines can produce high NOₓ in real‑world driving.
2. Extreme urban density + narrow streets
Paris is one of the densest cities in Europe. Pollution gets trapped in:
narrow boulevards
high‑traffic corridors
temperature inversions during winter
This leads to recurring pollution peaks requiring emergency measures (free transit days, alternate‑day driving bans).
3. Heavy commuter inflow
Millions of commuters enter Paris daily from the suburbs, many by car.
This creates chronic congestion, especially on the Périphérique ring road.
4. Dieselgate fallout
The 2015 emissions scandal revealed that many diesel cars emitted far more NOₓ in real use than in lab tests.
Paris was one of the European cities most affected.
🛢️ Fuel quality: Paris vs. Mexico City
Paris (EU standards)
Uses ultra‑low sulfur gasoline and diesel (10 ppm sulfur).
Follows Euro 6 emissions standards for new vehicles.
Fuel quality is strictly regulated and consistent.
Mexico City
National standards last updated in 2005, far behind U.S./Canada.
Gasoline has higher sulfur content, degrading catalytic converters.
New 2022 Mexican vehicles emit much more CO, NO, and HC than U.S./Canadian equivalents.
Hydrocarbon emissions from 2022 Mexican cars resemble U.S. vehicles from 20+ years ago.
Conclusion:
Paris’s problem is not fuel quality — it’s vehicle mix and traffic.
Mexico City’s problem is fuel quality + outdated standards.
What_to_Wear
So here I am worrying about 'what to wear'. As a Montrealer, most of the year
sees me wear - to go outside - what the temperature dictates. And that can be long stretches of
this or that coat and, hat and gloves. But there comes a time every year when what I wear
indoors and outdoors are pretty much the same. All of a sudden, it matters - socially - what I look
like. So here we are. I am feeling the pressure of that today.
I have decided to consult my elaborate Microsoft Weather app to help me. What can I learn
from it.
Today is a higher that average warm day. The average high on his date is 19C, but we should be
hitting 24C this afternoon. This tells me that it will be cooler getting there than it might be at
another date. I might well encounter some brave souls in Bermuda shorts in my walk. These
people are dreamers. I will be wearing ankle-length jeans, with runners. Along sleeved T-shirts
that 'breathes' as it gets warmer. And a shirt over it all, because that is what I am seeing women
wearing on the street.
What other information can I glean...
No precipitation:
Little wind:
Air quality is acceptable:
Not Dry:
Cloud cover minimal:
Feels warmer than it is, as will the next two days:
So I should wear light colors, but avoid the screaming colors of actual summer which are
appropriate when the blue of the sky and greenery hues are more vivid!!
* * *
The whole story...
Subscribe to:
Posts (Atom)