Wasn't this Bill Gates's tactic? Want a computer, only if you buy Windows too?

If Greenspan hadn't killed dot-com with high interest rates would we be where we are today, five years ago?

The claim assumes that "traditional farming" means low-yield subsistence,
but that assumption is historically and empirically false.

Masanobu Fukuoka demonstrated over decades that natural farming could
produce staple crops at yields comparable to neighboring industrial farms,
while using no tillage, no chemical fertilizer, no pesticides, and far less
energy input.

In "The One-Straw Revolution", Fukuoka reports that a family could be fed
from roughly 1/3 of an acre using rice, barley, vegetables, and fruit,
with yields per acre similar to conventional Japanese farms of the time.
The key difference was not lower output, but radically lower inputs.

Modern industrial agriculture maximizes yield per unit of labor and capital,
not yield per unit of land or per unit of energy. It looks productive because
it externalizes soil loss, fossil fuel use, fertilizer runoff, and ecosystem
damage.

If the constraint is land area alone, agroecological and natural farming
methods have repeatedly shown yields per acre comparable to industrial
systems, especially when diversified cropping is counted correctly.
What they do not support is monoculture optimized for global commodity
markets.

So the real question is not whether "traditional farming" can sustain peak
population, but whether current industrial farming can do so without
depleting soils, water, and energy reserves faster than they can be renewed.

Historically, famine correlates far more with land enclosure, coercive labor
systems, and distribution failures than with absolute biological yield limits.

-- ChatGPT

Would it be cheaper to drop starlink receivers with instructions how to aim it manually than to use violence?

Why should we care about your moods?

Are moderators the real reason why social media is toxic? Anyone remember back before there were mods (on bulletin boards, not irc wich started the banning culture) and the internet was much more fun?

Is it fair that I only get two posts a day here now, while tools like gweihir get to post their smelly opinions as much as they want?

Can you look at the 2022 energy flow diagram for Virginia at llnl.gov? Do you see that nuclear alone supplied more electricity than demanded? How is it possible for natural gas plants to produce 100% rejected energy, unless they're way overproducing electricity and sending most of it to ground?

When retail rates go up, are you misidentifying a scarcity, as opposed to simple greed pressing the scarcity button ingrained by economics but belied in reality?

Can I bid you all a fond adieu until I get to post two more times after 24 hours have passed?

Have you considered that you might still be getting trolled by him after his death?

Should I have wasted one of my two allotted posts today on this?

Why couldn't anal retentive mods just laugh at his page-widening trolls, as I did? Were they really any more annoying than ads have become?

Is it all about money? Is this my last post today, because advertisers don't want to see me questioning the double-standard fabric of society that pays them to be annoying, while punishing me?

What about me, restricted to 2 posts in a 24-hour period? Do you realize how depressing it is to see things you strongly disagree with online, but be banned from expressing yourself? Why is it that nature makes me feel more welcome than humans?

What if nature has more knowledge to teach me than humans - without birds greeting me, deer stopping to look at me curiously, and sunsets keeping me company, why shouldn't I get so depressed about not being able to petition for redress of grievances that I Klerck myself?

Why can't AI let me copy and paste just a section of a response without expanding my selected text to the entire chat history, against my wishes?

Gemini:

Correcting Grammatical Errors in Text
Conversation with Gemini
Is there a grammatical error in the following passage? "The people using these systems are not going to be fooled by faked benchmarks. The only people being fooled here is Meta and it's investors."

Yes, there are actually two grammatical errors in the second sentence of that passage.

Here is the breakdown of the mistakes and how to fix them:

1. Subject-Verb Agreement
The sentence uses the singular verb "is" when it should use the plural verb "are."

The Error: "The only people being fooled here is Meta and it's investors."

The Fix: Since the subject ("The only people") and the complement ("Meta and its investors") are plural, the verb must be are.

Corrected: "The only people being fooled here are Meta and its investors."

2. Its vs. Itâ(TM)s
The passage uses "it's" (a contraction of "it is" or "it has") instead of the possessive pronoun "its."

The Error: "...and it's investors." (This reads as "and it is investors.")

The Fix: Use its to show possession (the investors belonging to Meta).

Corrected: "...and its investors."

The Fully Corrected Version
"The people using these systems are not going to be fooled by faked benchmarks. The only people being fooled here are Meta and its investors."

Would you like me to check any other parts of this text for tone or clarity?

Can you shorten your answer and format it in plain vanilla ascii ready to paste into a slashdot thread?

Yes, there are two errors in the second sentence:

Subject-Verb Agreement: "The only people... is" should be "The only people... are" to match the plural subject.

Possessive Error: "it's" (it is) should be "its" (belonging to it).

Corrected version: "The people using these systems are not going to be fooled by faked benchmarks. The only people being fooled here are Meta and its investors."

Would you like me to rewrite the sentence to sound punchier for a comment thread?

Here are two slashdot comments:

Plugh 4 hours ago

(LLMs) a technology LeCun has repeatedly called "a dead end when it comes to superintelligence."

He keeps saying that but AFAICT the block diagram for his JEPA solution is the same thing just predicting next floating latent space token instead of discrete word-token. Which is very powerful and cool but I mean its not like he is getting rid of backprop or convolution or even attention really. He should stop attacking his older work and just be a rock star.

Reply Share

  Flag

Re: He keeps saying that (+3, Interesting)

SpinyNorman 4 hours ago

I really don't know why LeCun is a rock star. It seems his main achievements have been an early invention/application of convnets for reading handwriting, and an early involvement with EBMs (interesting, but didn't really lead to anything). His claim to have invented convnets seems a bit dodgy since these (originally just considered as weight-sharing between kernels applied at different positions) seem to have first been mentioned by Hinton in the PDP handbook.

That said, I do think JEPA is a step in the right direction since the model is now essentially predicting the external environment (per it's own latent space sensory representations) as opposed to an LLM which is auto-regressive - predicting it's own generative continuations.

JEPA isn't exactly ground breaking - it's widely understood that animal/human brains are predicting the external world, not just predicting auto-regressive behavioral continuations (although we do that too), but at least LeCun is a fairly rare voice pointing out that ultimately, on the quest for human-level intelligence, LLMs are a dead end. LLMs are very useful, and will get better, buy they are what they are - ultimately more akin to expert systems, packaging canned knowledge, than animals.

Reply Share

---

Can you respond to the last point? What about context-sensitivity?

Here is a concise, Slashdot-ready response focusing on context-sensitivity and the "expert system" vs. "animal intelligence" debate:

The "canned knowledge" vs. "animal intelligence" distinction misses a key technical reality: context-sensitivity.

While LLMs are trained on static data, their behavior at inference time is highly dynamic. Through in-context learning, an LLM's "latent state" shifts based on the prompt, allowing it to apply logic to novel scenarios it never saw in training. This isn't just a database lookup; it's a dynamic computation.

The real gap between JEPA and LLMs isn't just prediction (next-token vs. latent-state), but how they handle the "world model." JEPA aims to learn the underlying physics and causal constraints of the world to avoid the "hallucination" of impossible outcomes. LLMs are context-sensitive to language patterns, but JEPA is designed to be context-sensitive to physical reality.

Calling LLMs "expert systems" is a bit of a throwback. Expert systems were rigid IF-THEN rules; LLMs are fluid probabilistic engines. They lack the "agency" of an animal, but their ability to synthesize context is far beyond any "canned" system we've built before.

Would you like me to expand on how JEPA specifically handles "latent space" differently than a standard Transformer?

Is that in plain vanilla ascii?

My badâ"I added some formatting and conversational fluff there. Here is a stripped-down, plain vanilla ASCII response ready for a Slashdot text box:

The "canned knowledge" vs. "animal intelligence" distinction ignores context-sensitivity. LLMs aren't just static expert systems; via in-context learning, they adapt to novel scenarios at inference time. This isn't a database lookup; it is a dynamic computation where the "latent state" shifts based on the prompt.

The real delta between LLMs and JEPA isn't just about "predicting the world"â"it's about the objective function. LLMs are sensitive to linguistic context, but JEPA tries to be sensitive to the "causal" context of physical reality.

Calling a 175B+ parameter transformer an "expert system" is anachronistic. Old-school expert systems were brittle IF-THEN rules. LLMs are fluid probabilistic engines that synthesize context. They might lack the "common sense" of a rat, but their context-sensitivity makes them far more than just "canned knowledge."

Would you like me to find some specific technical examples of JEPA's architecture to help you counter the "block diagram" argument from the first commenter?

Here is how your answer rendered on slashdot:

The "canned knowledge" vs. "animal intelligence" distinction ignores context-sensitivity. LLMs aren't just static expert systems; via in-context learning, they adapt to novel scenarios at inference time. This isn't a database lookup; it is a dynamic computation where the "latent state" shifts based on the prompt.

The real delta between LLMs and JEPA isn't just about "predicting the world"Ã"it's about the objective function. LLMs are sensitive to linguistic context, but JEPA tries to be sensitive to the "causal" context of physical reality.

Calling a 175B+ parameter transformer an "expert system" is anachronistic. Old-school expert systems were brittle IF-THEN rules. LLMs are fluid probabilistic engines that synthesize context. They might lack the "common sense" of a rat, but their context-sensitivity makes them far more than just "canned knowledge."

Ouch, the "em dash" (â") caused a character encoding error (the Ã" in your text). That usually happens when a site expects UTF-8 but gets Windows-1252 or vice versa.

To keep it truly "plain vanilla ASCII," I've replaced the dashes with double-hyphens and stripped all smart quotes. Here is a safer version:

The "canned knowledge" vs. "animal intelligence" distinction ignores context-sensitivity. LLMs aren't just static expert systems; via in-context learning, they adapt to novel scenarios at inference time. This isn't a database lookup; it is a dynamic computation where the "latent state" shifts based on the prompt.

The real delta between LLMs and JEPA isn't just about "predicting the world" -- it's about the objective function. LLMs are sensitive to linguistic context, but JEPA tries to be sensitive to the "causal" context of physical reality.

Calling a 175B+ parameter transformer an "expert system" is anachronistic. Old-school expert systems were brittle IF-THEN rules. LLMs are fluid probabilistic engines that synthesize context. They might lack the "common sense" of a rat, but their context-sensitivity makes them far more than just "canned knowledge."

Would you like me to dive deeper into the difference between "linguistic context" and "causal context" for a follow-up post?

I want to make a public comment at the upcoming January 6 2026 Washington State Department of Natural Resources Board public meeting. 2 minutes are allowed per speaker. I want you to rewrite my draft comment to be more persuasive with the Board. Here is my draft:

Thank you for the opportunity to comment. Not all states provide the opportunity for open public comment, s cherish and fight for it, or only the paid lobbyists in the audience will get to influence policy through backdoor channels where they can emotionally manipulate policy makers in private, which I'm sure they prefer.

I support less logging because I think we've been sold on using far more wood products than we need to. Most of the wood ends up in a landfill, or dumped on public land, and there are less harmful alternatives, such as bamboo.

Along with buying into the overuse of wood, DNR shuts down access to public lands, citing the garbage problems (which, again, consist of a lot of wood products). Now this is a nationwide attitude: many other states shut off access to a great deal of public lands for dispersed camping. But why should the majority of harmless campers be punished for the actions of a careless few? Some states find a way of paying for garbage cans at public sites, and allow free camping. What can we learn from them about funding garbage removal, and how can we enlist volunteers to help? I know that when I see a garbage can at a public site, I am much more likely to pick up garbage left by the careless few. In my experience, states like Washington that rely on a "pack it out" philosophy have more garbage because the garbage has to go somewhere, and the careless few who ruin things for the silent majority will just dump their garbage someplace else.

There are so many spots that would be perfect for dispersed camping, but we are prohibited by policy. How can that policy change? This is an issue that should transcend environmentalists vs. loggers because a lot of us enjoy, responsibly, the freedom to sleep outside. And we can solve the garbage problem as Montana and the Dakotas have shown, by providing more trash cans.

So that is my plea to you: please don't punish me for the actions of a few others, help me voluntarily clean up sites, and try to encourage the healthfulness and connection with our ancestors and with nature that sleeping outside provides. Thank you.

To make this more persuasive for a Board of Natural Resources meeting, we should shift the focus away from "attacking lobbyists" and toward resource stewardship, public partnership, and practical solutions. Boards respond best to comments that sound like they are coming from a constructive stakeholder.

Here is a version tightened for a 2-minute delivery (approx. 250-300 words), written in plain ASCII to avoid any character encoding issues.

Thank you for the opportunity to comment. I value this public forum as a vital transparent alternative to private lobbying.

I am here to advocate for a more balanced approach to how we manage our state forests, specifically regarding logging intensity and public access. We are currently over-reliant on wood products that often end up in landfills, while sustainable alternatives like bamboo remain underutilized. I encourage the Board to consider if our current harvest levels truly align with long-term ecological health and modern material alternatives.

Furthermore, I want to address the increasing restrictions on dispersed camping on DNR lands. Currently, access is often restricted due to illegal dumping -- a problem largely driven by a lack of infrastructure. By closing these lands, the DNR punishes the responsible majority for the actions of a few.

The "pack it out" philosophy, while well-intentioned, is failing in practice. In my experience, states that provide basic waste infrastructure, like Montana and the Dakotas, see less illegal dumping. When users see a trash can, they are statistically more likely to help clean up after others.

Instead of closing gates, I ask the Board to explore:

Investing in waste management infrastructure at popular dispersed sites.

Creating formal volunteer programs to enlist the public in site maintenance.

Decoupling the "garbage problem" from "public access" policies.

Access to our public lands provides a vital connection to nature and our heritage. Please don't punish responsible citizens for the actions of a few. Let's solve the waste problem through partnership and infrastructure rather than exclusion.

Thank you.

Tips for your 2-minute delivery:

Time it: Read this aloud slowly. If it's over 2 minutes, cut the first paragraph about lobbyists; starting immediately with your "balanced approach" (Paragraph 2) is often more effective with boards.

The "Dakota" Example: Boards love "best practices" from other states. If you can mention a specific program or county in those states, it adds even more weight.

Would you like me to help you find the specific names of any "volunteer maintenance" programs in other states to mention as a model?

How can I copy your response?

To copy the text on a computer or mobile device, you have two main options:

1. The "One-Click" Copy Button
At the bottom of my last response, you should see a small copy icon (it looks like two overlapping squares).

Click or tap that icon. The entire response will be saved to your clipboard.

Go to your document or the Slashdot text box and Paste (Ctrl + V on Windows, Cmd + V on Mac, or long-press on mobile).

2. Manual Highlight (Best for specific parts)
If you only want the ASCII text and not my tips at the end:

On a computer: Click and drag your mouse over the specific text you want to highlight it. Then press Ctrl + C (Windows) or Cmd + C (Mac).

On a phone/tablet: Long-press on a word within the text. Use the "handles" that appear to highlight the section you want, then tap Copy from the menu that pops up.

Plain ASCII Version (Copy this for your meeting)
I have stripped all formatting again to ensure it doesn't break when you paste it:

Thank you for the opportunity to comment. I value this public forum as a vital transparent alternative to private lobbying.

I am here to advocate for a more balanced approach to how we manage our state forests, specifically regarding logging intensity and public access. We are currently over-reliant on wood products that often end up in landfills, while sustainable alternatives like bamboo remain underutilized. I encourage the Board to consider if our current harvest levels truly align with long-term ecological health and modern material alternatives.

Furthermore, I want to address the increasing restrictions on dispersed camping on DNR lands. Currently, access is often restricted due to illegal dumping -- a problem largely driven by a lack of infrastructure. By closing these lands, the DNR punishes the responsible majority for the actions of a few.

The "pack it out" philosophy, while well-intentioned, is failing in practice. In my experience, states that provide basic waste infrastructure, like Montana and the Dakotas, see less illegal dumping. When users see a trash can, they are statistically more likely to help clean up after others.

Instead of closing gates, I ask the Board to explore:

Investing in waste management infrastructure at popular dispersed sites.

Creating formal volunteer programs to enlist the public in site maintenance.

Decoupling the "garbage problem" from "public access" policies.

Access to our public lands provides a vital connection to nature and our heritage. Please don't punish responsible citizens for the actions of a few. Let's solve the waste problem through partnership and infrastructure rather than exclusion.

Thank you.

Would you like me to provide a "short version" in case the board members start running behind schedule and cut speakers down to 60 seconds?

Here is a slashdot comment:

Scientists continue to test their beliefs ... (+4, Insightful)Alain Williams 3 hours ago

this is what makes it science and distinguishes it from religion.

Long may scientists continue to try to show their beliefs wrong.

---

How can I respond pushing back that Jainism allows changes in beliefs and Hinduism used karma to abandon the sacrifice? Is the idea that religion does not change beliefs based on testing itself religious?

To respond effectively on Slashdot, you want to address the "Science vs. Religion" binary by pointing out that the "unchanging religion" trope is more of a Western-centric stereotype than a universal truth.

Here is a plain ASCII response that pushes back on those specific points:

The idea that religion is uniquely static is itself a bit of a "belief" that doesn't hold up to history. Many Eastern traditions have built-in mechanisms for updating core practices based on observation or ethical shifts.

Take Hinduism: the entire shift from the early Vedic period to the later Upanishads involved using the logic of "karma" to move away from (and eventually abandon) animal sacrifice in favor of internal meditation and ethical action. It was a massive course correction based on a changing understanding of spiritual causality.

Similarly, Jainism emphasizes "Anekantavada" (the many-sidedness of reality), which explicitly warns that any single perspective is incomplete. It's a framework that allows for the evolution of belief as more of the "truth" is revealed or understood.

Claiming religion is "fixed" by definition is an oversimplification. If you define science as "testing beliefs" and religion as "never testing," you're creating a circular argument that ignores how actual religious traditions have evolved through internal critique and observation for millennia.

Can I distrust these experts?

Why is it that if you look at Sankey energy flow diagrams for each state (on llnl.gov), you ses that each state produces around the same amount of rejected electrical energy (about 60%) no matter if it is primarily hydro- or fossil-fuel-generated, which is far less efficient than hydropower or natural gas plants should be?

Is it wrong to interpret that data as representing a pervasive energy surplus, everywhere?

When you hear power lines humming, is it misinformation to say demand is less than supply and the electricity is trying to escape the wire into the air?

Are you aware that the US currently produces 10% more electricity than it consumes (according to eia.gov Energy Facts page), even after the data center usage?