Pinned Tweet
Starting a support group for meditators who knew about jhana before nick and feel utterly eclipsed by his accomplishments in practice and distribution; hmu
coffee on an empty stomach feels terrible because it forces cortisol release to liberate glycogen as compensation. so what is blue light is similar? triggers some metabolic activity that typically occurs in presence of energy from uv/infrared, which, if not present...
Blue-rich LED lights in classrooms makes kids aggressive.
John Nash Ott, author of Health & Light, noted: under fluorescent lights, some kids were anxious & creating havoc.
Bring in more sunlight and that same kid was front of class, answering questions.
Isolated blue lights without infrared & UV from the sun = perpetual stress mode.
More below.
Yes, Jaron Lanier! tl;dw—why do doomy AI takes dominate? Because the (young) creators of AI grew up on bleak sci-fi: The Matrix, Terminator, Marvel. "I'll go further—if Star Trek Next Generation had lasted another ten years, there would be far fewer teen suicides."
Don’t get hung up on whether you agree with his point on AI.
Just listen to what he has to say about Star Trek vs dystopian sci fi stories.
He’s grasping at something more important than AI. It’s what I’ve been trying to help people see for years:
The story you repeatedly consume becomes the story you live into.
sped up grouped-int4 kernels by 2.5x on 13-year-old Ivy Bridge CPUs, nearly doubled GLM 5.2 tok/s: github.com/JustVugg/colibri/…
Or.
NVIDIA v100 32GB
2x + r720 = $1600
- 2x qwen3.8 55 tok/s
- qwen3-coder-next-80b 80 tok/s
- qwen3.5-122b 44 tok/s
See what else you can do with a pile of rust: sloptimized.ccb.gg
two things can be true! the second being that the alpha on the table for everyone who doesnt have piles of GPUs is building a system capable of keeping the GPUs fed. how is your metaharness coming along?
A pile of legacy metal with 3x the memory bandwidth that costs 2/3 of the DGX Spark:
- v100 32GB 900 GB/s: $650 (x8)
- poweredge r720: $300 (x4)
~> $6500
bonus: it doubles as a space heater.
So you want local AI and want to know what hardware you need?
To get close to real-world, "cut the cord" local AI you need 256GB of memory in a decent package - can be DGX Sparks (TP-2), a Mac, or dedicated GPUs. It will cost you ~$10K USD or more depending on where you are and you can run GLM 5.3 Flash, Deepseek 4.1 Flash, or Qwen 3.8 Flash Next and many folks will be pretty happy. It's not going to be crazy fast but it will be usable - point OMP (or your favorite harness) at it and party on.
Caveats:
- even the best available model recipes at this level are not frontier equivalent - they're frontier-1 (or -2) at best, but most practical tasks and coding no longer strictly requires frontier
-the models often barely fit, need to be carefully optimized, and unless on dedicated GPUs ($$$) will just barely hit acceptable real world speed and concurrency for one person doing serious stuff or a small handful of people doing light stuff
- those boxes are gonna be busy and dedicated to model serving - little or no room left for sidecars < this is something that bit me and I ended up needing a 3rd spark 😂
That's not to say the next step down (in the 96GB-128GB - single spark - range) isn't useful. There are lots of very smart models you can run in that tier (and frankly on even smaller machines) that will do a lot of useful things. It's an an awesome place to start!
I've been there personally - first with a Macbook 96GB that I've been through model hell and back with, then with my first Spark that performed admirably in a solo configuration on several workloads for months. Both of them did real work 24/7 in complete privacy and were awesome learning platforms, but neither ever came close to being able to host a model to replace my daily driver.
> be me
> write a harness wrapper that does preamble composition (AGENTS.md from multiple source dirs)
> love it, use multiple harnesses daily for months
> write a nice blog, submit to orange site
> 2 points
> claude: "we now read AGENTS.md"
> 598 points
cameronboehmer.com/p/portabl…
So @VW has decided not to support lovers of privacy and digital sovereignty by disabling their mobile app on @GrapheneOS. I wish I could say I'm disappointed, but given that the app has never had respect for so much as a password manager, I can only say that I'm not surprised.
Thank you from the bottom of my heart to Analogue for their gracious support of my inquiry into the relationship between structured water and rouleaux formation (red blood cell clumping)! Have been loving my new OMAX darkfield scope 🤩
Introducing the Spring/Summer 2026 batch of Expeditions: an experimental fund organized around the quality of interestingness. Here are the six from this season:
analogue.press/p/introducing…
How EMF ☠️ your tissues: EZ water (-) builds up around hydrophilic blood cells, allows healthy spacing by ejecting positively charged particles and repelling other cells. EMF disrupts EZ water, inducing rouleaux (RBC clumps). Rouleaux > smallest capillaries, nearby tissues starve
My blood before and after 10 minutes of meditation&prayer on a grounded pillow. Amazon basics microscope, 800x. Hanging out with greensprings.life/
Yeah and good thing. EMF disrupt the structure of water in your body, leading to rouleaux formation (red blood cell clumping), which means tissues fed by your finest capillaries (which are less than one cell wide) starve.
bbc.com/future/article/20260…
My blood before and after 10 minutes of meditation&prayer on a grounded pillow. Amazon basics microscope, 800x. Hanging out with greensprings.life/
