would radiation pressure not counterbalance the radiation pressure from the sun?
given solar cells are only 30% efficient. if you put the radiators behind the panels then you have the structure already and always keep the radiators shaded
They seem to have solved the car break-in problem since you left. Overall the city feels much safer and cleaner than it did 10 years ago, although the character of the city has also changed.
While books > articles > audiobooks > podcasts > listening to music > TV > short-form content, it's all content consumption in the end, and while that's fine, I think it's a little overhyped.
I prefer to draw the distinction at creating vs doing vs consuming, and focus on balancing the time spent consuming (which is a natural part of enjoying life, but usually people consume way too much) unless it's in the service of one of the other two categories (e.g. consuming articles to learn/inspire the creation of something better or plan a trip to do something, etc.)
No, it is not all just ‘content’ which is roughly the same and can be consumed like a snack. That’s a word used by people who think thoughts, feelings and stories are just carriers for advertising and selling you things.
Words matter, don’t take your word choice from ad factories instagram which popularised this term.
Reading (I won’t say consuming) a good book can stay with you forever. People from 2000 years ago can speak to you in their own voice from behind the grave. And you have the audacity to call it content and compare it to cat videos and call it ‘overhyped’?
Reading a book from 2000 years ago is not the same class as reading a recently published modern romance fiction. The latter is basically consumed like a snack.
On days when I am too lazy to be productive or be “creating” something, I sometimes take out my old Harry Potter collection and read for two straight hours. I don’t really feel that’s meaningfully different from watching online videos for two straight hours. It is indeed just consumption. And Harry Potter series is in fact one of the better ones among contemporary literature.
Although I can't personally imagine re-reading YA lit, it still requires you to use huge amounts of grammar, vocabulary, and cultural knowledge in a seamless automatic way. To imagine an alternate world, people, and culture, while your eyes scan constantly across page after page invisible to your captured mind.
Those are the same language tools you need to use to create and evaluate your own writing. You're using a mental tool at a high level, even if it's not Ulysses (or Homer's Odyssey). This is fundamentally different from passively viewing cat videos, or thirst traps, which may well not involve language at all.
A 2 year old can be entertained by short form video, but can not even sound out a single word of a book.
Reading is the least stimulating than other forms of "content" mediums.
Reading about a spell duel is neurologically (and physically) going to be a very different experience than watching flashing colours on screen with loud blasts.
The visualisation of the world printed in words is more active than having it painted on screen for you.
I struggle to read, the last fiction I read was now 8 years ago but the value of it is not lost on me...
I didn't say it was all roughly the same or that you shouldn't do it, but it's obviously still consumption. If nobody ever consumed, there would be less value in creating art. Like all things in life, it should be done in moderation and balanced with creative activities.
I find the words consumption and content problematic. They encourage you to look at the world in a certain way. What would happen to your world view if you step outside the confines of those words, defined by advertisers who don't care about quality but instead measure clicks, engagement and profits and have encouraged the worst sort of 'content' to thrive (short form videos with very little substance)?
Most good art or writing requires engagement from the reader and something is created in engaging with it, something new that exists between the original work and the world of the reader. So I fundamentally disagree with the framing that reading is a passive activity and requires no creativity or engagement, or that creating is fundamentally different from engaging with the creations of others.
> Most good art or writing requires engagement from the reader and something is created in engaging with it
This is an interesting point, and as I said before, reading a book >> than short form content (although you could also argue a 3Blue1Brown video watched intentionally could also provide some level of engagement, while still being categorically different than creating a video or novel in the first place). I would agree that the best forms of art, whether in novels or music or otherwise, invite some level of creative participation. But that is rare, and it's still (in my opinion) a fundamentally different and constrained form of creativity.
> I find the words consumption and content problematic.
They're just words, they're appropriate here, and they had meaning long before digital ads.
That being said, your needless hostility suggests you attach a significant amount of ego to reading. Maybe you aren't aware of that, so I leave it as an observation here. If you like reading books, no one here is questioning your life decisions. If you use reading to signal superiority, well, that's a whole different issue.
I use a similar heuristic, but I think what we "do" for leisure is still consumptive, albeit an active sort. Reading, for it's part, exercises cognition better than something like TV.
I rank listening to music way above audiobooks and podcasts. Music is a revelation.
I knew someone who worked at some sort of homeless services clinic, and I asked how many people they were able to help get back on their feet in a sustainable way. The answer was 1; I don't know what the denominator was, but the takeaway is that most of the visibly homeless people you see are uncoachable and you really can't protect people from themselves. With a huge caveat that the visibly homeless is a subset of the homeless/impoverished population.
The implication (given that the US is an outlier here) is that prevention is much better than cure here; that means reform of healthcare and social housing so people are less likely to wind up in that trap.
I can't imagine who is using something like this for agentic coding, but I see exciting opportunities on the horizon when we can have hundreds of reasonably rational and conversational agents working on local machines to simulate emergent behavior (simulating crowds, markets, ecosystems, game NPCs, etc.)
> We recommend using it for agentic workloads, tool use, data extraction, RAG, and long-context workflows. It is not recommended for agentic coding and knowledge-heavy tasks.
Really curious about people's workflows with these agentic-but-not-for-coding workflows. Are there some interesting people to follow there or just good testbeds/environments to get an idea ?
Yeah. Clearly the target for these models is on-device assistant/analyst tasks where really everything is agentic.
Thinking out loud: any text-based generative AI application is on a continuum between:
- document-to-document: solve this language problem using language skills by outputting only new symbols, and
- document-to-tasks: solve this language problem by only operating these tools.
In a sense, given access to vast compute to train a very large model, agentic coding is one of the easier applications that is somewhere between the two. It is manipulating symbols in a mixture of languages that are biased towards context-free (code), it exploits the embedded knowledge in a vast number of weights, and it calls fairly simple tools. The user-focussed solution is happening inside the LLM.
A small model can't do this job well, even if it has a good understanding of programming languages, because it lacks the world knowledge to understand the problem.
At the other end of the continuum is: these words mean do this thing. That Cactus Needle 2 model mentioned earlier is here. It doesn't even produce a language description of what it did, because it one-shots tasks.
In the middle but near this end is the fantastically hard job: train a model that understands language and reasons well enough to respond to queries about the tools it has access to, operate them and reply in natural language, without being large or slow. That is, understand language without being overburdened by details of scenarios that caused the need for the words in the first place.
This is maybe the great grand challenge of LLMs: make it know how to speak and reason and think and work for us while discarding everything that is just knowledge of unnecessary facts.
Small reasoning models are where most of the really big challenges are.
Removing spam from email, extracting attachments (invoices) from emails, and otherwise managing E-mail. Those are my use cases, although I suspect this model will be too small for that. I currently use 27-35B models (MoE mostly).
I feed these things my tax transcripts and it found all sorts of refunds and ways to waive penalties that the IRS and my state agreed I was eligible for, because my current CPA doesn’t do this kind of tax matter
The high parameter 2026 models know the latest tax law, revealed deadlines for court rulings to me that are in fact real and new branches of service on the IRS website, or can learn while also agentically search
While many debate the utility in legal matters and 2023’s hallucination issue, the rest of us just do
“High parameter […] models” was mentioned above. It’s still not a good take however, as laws of civilizations getting more convoluted over time is the general behaviour, not specific to LLM-producing countries.
I am getting to know Hermes agent and in the last couple of days I used it for searching the web for medical specialists or finding the best deal for a product online, manipulate with excel and word documents, gather data from APIs, finding stuff in my computer or keeping track of itself running in another process and keep the computer awake and when detecting the process is finished, put computer to sleep.
I have noticed that these cheaper and faster models are very great for Ops-work. Luna max is beast when you use some stronger model to write detailed instructions/run book what to do and when to stop.
You can use it for things like classifying documents, spam detection and detecting certain events, etc. You're right that it's not strong enough for coding but it can still be quite good for question answering and such. Or even simple things like converting documents from one format to another.
And yeah, for NPCs, this model and their other A1B model can fit inside of a game and run at a decent speed on CPU-only. It would probably be workable for game characters.
> The risk is the same as what Intel had. There is a group who are pushing them to go bigger and with a resource no limit approach, who have a lot of dollars to push you that way. Follow them and they lead you to a pile of money, but then you risk something like Apple Silicon happening.
The problem was that Intel stopped making things faster and better, and shifted to wearables and mobile chips instead of investing in their hard-tech strategy that had worked for decades.
Anthropic's fear-mongering and marketing is the reason we have these restrictions in the first place.
Despite their virtue signaling, Anthropic is the only major lab that has never released an open weights model, has been caught intentionally nerfing a model after release (Opus 4.6), intentionally and silently degrades performance for suspected competitors and AI researchers, complains incessantly about distillation when everyone is doing it (and after they settled for pirating books), and wants to pull the ladder out from everyone trying to catch up.
They're anti-consumer and only concerned with holding the power themselves.
I'm not a fan of Altman, but Anthropic is the worst actor in the space, and I hope they lose.
Everyone in the space was talking about the automation of work from about day 2. People couldn’t stop themselves from talking about the way it was going to end work, and tech firms were firing people left right and center over AI.
Notably, Anthropic is the firm that stuck to its guns with the US Government, meaning they likely believe in their own spiel.
I mean it's fear-mongering until it isn't. I think people have become a bit too comfortable with dismissing the dangers of misaligned AI as simply "marketing hype".
What about openai's fear mongering, or googles, or JP Morgans, or Frank Herbert's, or Arthur C. Clarke's or Samuel Butler's?
If you can't envision plausible scenarios where very bad things happen because of a malevolent actor, ChatGPT 6, and a little bad luck - you need to think harder.
Yeah honestly I don't get any of the fear-mongering from any side. If access to intelligence and knowledge is scary to you, that's a you problem.
Is it going to change the world? Yes. In more positive ways than negative ways. Websites will continue to get hacked, as they are already getting hacked. People who are afraid of AI are really just afraid of change.
Even the original idea (if The Social Network is a trustworthy source) was copied -- Zuckerberg just has a complete lack of vision, but is clearly an intelligent operator with good business sense. Jagged intelligence, like an LLM.
We just had a thing that would export MSN messenger contact list, you'd add your email there and everyone would add you on MSN. Way less creepy than the stalking fb was doing.
has zuckerberg done anything productive of his own accord? facebook was ripped off. most meta companies were acquired. it could be argued that propagandizing the entire world was productive relative to vested interests but that was done on the behest of those interests, not zuck himself. the only thing im aware of zuck actually spearheading was the metaverse, a conclusively unproductive pursuit costing tens of billions to achieve literally nothing. it isnt objective to unilaterally behave with vitriol ... still this person seems more comparable to cancer itself than any actual human. i guess you could collapse productivity to 'making money' in which case clearly he is productive, im more referring to accomplishing anything useful for humanity. i also dont consider mass surveillance to be useful for humanity as bad actors will always get away with it whether they are on or off camera.
I don't think you can say that the idea was "copied". It was a very obvious idea. I had the same idea before I heard of the Facebook. Do you know why it's called the face book?
You may as well say Bezos copied the idea of Amazon from book shops.
Really, he was in exactly the right place at exactly the right time.
Except there is plenty of public evidence Facebook was. I think they were even on the cap table and thiel / palintir are explicitly partners in wiring up tech data pipelines into the knowledge graph. Thats why thiel was an early investor, this is known public info.
Thanks for that. I am confused about your prior comment. Maybe we are both cia and have forgotten because we were part of mk ultra and melted each other's memories
We've been hearing about the risks of engineered viruses and homemade superweapons since GPT 3.5, so where are they? We've had abliterated open weights models much stronger than GPT 4 for over a year now.
It's been interesting seeing how OpenAI pops up to counter the threat of AGI being controlled by Google, and then OpenAI and every spinoff company from its employees has become a far larger threat to the public, for different reasons.
As much as it seems like Anthropic's self righteous leadership truly believes in what they're preaching, they've shown themselves to be tied for the worst stewards of this technology. Google actually seems like the best option to me, by far. Anthropic is also the only major lab with no open weights releases.
They'll have burned a lot of goodwill with the community by the time another lab takes the tech lead, which I guarantee will happen.
> We've been hearing about the risks of engineered viruses and homemade superweapons since GPT 3.5, so where are they? We've had abliterated open weights models much stronger than GPT 4 for over a year now.
What I can't understand, is that they act like the _knowledge_ is dangerous.
I don't know if I'm biased from my BSci (chem/maths), but: knowledge isn't dangerous, the reagents needed are incredibly easy to control. Thats what we already do!
What is this comment? If they occurred we would face a huge disaster; isn't it better to err on the side of caution to make that risks as low as possible???
No no we should push the limits until a bioattack happens, then when those people are all dead we comment angrily on the hackernews thread and say that someone should have seen this coming
What are you willing to give up to be less afraid of all these extremely hypothetical risks? (which in the case of bio/nuclear/chemical is mostly access to controlled materials and capital anyway)
Access to information and knowledge is probably the very last thing I'd be willing to give up, personally.
I think most people would be a lot happier if they were less fearful in general.
>We've been hearing about the risks of engineered viruses and homemade superweapons since GPT 3.5, so where are they? We've had abliterated open weights models much stronger than GPT 4 for over a year now.
They have existed far longer than LLms, and kill tens of thousands of Americans every year.
Fentanyl is the best example. Any chemistry graduate can figure out how to manufacture it. Used by China as a Chemical weapon against the American public.
I had a long process with this that mostly manifested as exercise intolerance and general inflammation/discomfort, and sleep struggles. I made no progress for 2 years, lost most of my muscle (I had been very active before) and started thinking "is this how it's going to be forever?". After not finding anything promising from traditional medicine or supplements, I finally made some dramatic life changes. I'm fully past it now (with persistent lifestyle changes), but I really had to rethink my relationship with food.
Ended up doing a paleo diet, avoiding stressors (some of which are not obvious like just being on your phone scrolling, bad posture/circulation/sitting for too long), improving sleep hygiene, and ramping up consistent cardio exercise, with an emphasis on getting up to 4x/week zone 5 cardio without triggering intolerance.
Since then I've discovered a lot of other things that are great for overall health, like HRV-reset breathing and long-duration water fasts (around 3 days is optimal for me). I imagine those would have been very helpful if I had tried them earlier. A water fast is a complete metabolic and inflammatory reset of the body, and it's not as hard as you might think.
Hopefully most affected folks have recovered and are living normal lives by now, but if not, there are things you can do! It seems like the more challenging those things are, the more efficacious.
Having gone through this too, I also had to accept that a lot (not all!) of it was in my head and made worse by it. When I convinced myself that “this will pass” and “this slow steady plan will get me out of this eventually” was when I finally saw regular progress (progress, not immediate relief).
I'm glad that this was the case for you, but presenting it as an universal truth is extremely cruel and harmful to people for whom it is not, and you should be really careful with phrasing. Statements like these is why people's symptoms are not being taken seriously.
How do you prepare/deal through a water fast? What kind of supplemens would be needed for the water?
Can I read up on this anywhere? I'd welcome a suggestion over surfing the many many pages I found through a simple search. A book or paper reccomendation to read up on would be nice as well
Honestly AI overviews are a pretty good guide, ask about a 3-5 day water fast. I was nervous going into my first one, but now I don't worry about it. The main thing is to drink a lot of water and have 1-3 LMNT electrolyte packs (unflavored, no sugar) per day, depending on how much you're sweating/exercising (which you absolutely can do, especially the first 2 days).
You can expect to feel colder as your body doesn't conserve as much heat, and after ~2 days more lethargic physically, but your mental energy may actually be higher. I don't sleep as well when fasting, so 3 nights is about my limit. That being said, you feel rested on less sleep, because your body is probably producing a lot less waste.
A non-inflammatory rocket shock diet can certainly aid in symptoms of long covid in many users, often people megadose on antioxidants to dilate their recovery window and not regress. Glad to hear you are feeling better and I totally agree that movement and diet are key in recovering from inflammatory disease.
Zone 5 is usually 1-2 minutes out of a longer 30 minute cardio session for me, I do it as a final sprint. I am not talking about repeated hill sprints where you would get 10+ minutes of zone 5 cardio in a session, which I agree would not be something a normal person should do 4x/week.
From my own experience, it seems like hitting that Zone 5 briefly is a good nervous system reset (overrides any dysfunctional breathing and heart rate effects from long covid); it's less about training the heart, although that's an excellent side effect.
> hitting that Zone 5 briefly is a good nervous system reset
I ended up doing this inadvertently earlier this week due to some medication side effects (yay akathisia!) making me want to go outside and just seriously run fast, including up the nasty hill nearby. Which I could, so I did. I'm amazed at how much it seems to have helped, for how simple it was. (Another one from the past, though for different situations: alternating hot and cold showers.)
Thanks for this, reading "water fast" and "3 days" gave me a shot of adrenaline. The "water" prefix is just confusing, the word for abstaining from food is just "fast" for those interested.
dry fasts aren't always what they appear. if you have significant glycogen stores in your body as you begin your fast you wont be dehydrated for the first day or two as water is freed. what usually happens is someone who starts glycogen endowed discovers that they aren't thirsty when they start fasting and tout it as dry fasting.
What you thought of (not even drinking water) is called a dry fast. It is a thing, but for obvious reasons is much more intense and shorter in duration.
Thanks. Unfortunately, not really, especially where diet is concerned. For example, I saw significant improvement after removing wheat, white rice, red meat, and dairy entirely, which is not something your typical US doctor would suggest. The first doctor I saw wanted to put me on antidepressants.
I also think exercise recommendations are generally too low, especially with respect to high intensity cardio.
reply