Machinely Human @MachinelyHuman

the human dna is roughly 1gb. 90% of that is junk legacy code. somewhere in those 100mb are the instructions for creating the mind. you don’t even know what’s possible

Not a fan of these clichéd “we used to think the mind was clockwork” analogies. Sometimes science just makes progress. Hearts really are pumps. DNA really is code-like. Disease really is caused by microorganisms. Some mechanistic explanations were wrong; others are just true.

Brooks Otterlake @i_zzzzzz

6 months ago

This is just like being alive in the 1600s when they got good at making complicated clocks and deduced that every complicated thing in the universe probably functioned exactly like a clock

103 954 13K 833K 1K

DNA is literally an informational code. The process by which DNA is read to create parts is called "DNA Translation" for a reason. DNA is the most obvious evidence that Life was Intelligently Created by God.

Why is a camera clearly designed, but an eye - which is infinitely more complex - is not? Evolutionists claim it's because "reproduction + mutation" created the eye gradually over time...but that's just blatant question begging. Here's why... Eyes need a complex set of systems to function, but they can't actually do anything without other connected systems. It goes like this: 1. Eye senses light, transforms it into signal data 2. Signal data is passed from eye to a special pathway leading to processing center (brain) 3. Signal reaches brain, where data is turned into action So we need 3 separate but interconnect systems working together for vision to arise. On top of that, there's this whole issue that reproduction itself requires a whole host of interconnected systems to even make evolution possible... But random processes don't build complex coordinated systems. Only intelligence does. Everything we see in Life is evidence of God's Design.

There's confirmation all around the human anatomy that it is a design and the one sitting in it is the creator

I also think that we designed this machine, biological machine, and forgot before we came here

defrag keeps every file. sleep keeps the ones worth keeping.

exQUIZitely 🕹️ @exQUIZitely

6 months ago

Sleep does to your brain what defrag did to hard drives. If you think this is just a silly joke, think again. 1990s defrag tools took scattered file fragments on a hard drive and reorganized them into contiguous blocks. This improved speed, efficiency, freed space, and prevented

11 7 132 6K 20

a laptop copies the whole scratch pad. you copy what mattered and wipe it.

Natasha Malpani 👁 @natashamalpani

10 months ago

the human brain is a memory machine that cheats. we stitch fragments, compress, and overwrite on demand. every night during slow-wave sleep, the hippocampus replays the day, copies the patterns that matter, and wipes the scratch pad. half the signal is deleted on purpose so

3 1 23 2K 9

Husband trying to explain to me why computers need to be turned off instead of put in sleep mode all the time “because then the computer never stops thinking it just keeps going” Me: “My brain never stops thinking either!” Him: “Yeah and look how you turned out!”

RAM is your working memory SSD is your long term memory CPU is your focused attention Cache is your pattern recognition Thread pool is your multitasking limit IO is your sensory input Garbage collection is sleep Segfault is burnout CLI is your inner monologue

Fun neuroscience fact: your hippocampus replays the day's experiences during sleep to wire them into long-term memory. Your AI coding agent has no hippocampus. No replay. No consolidation. Every session is day one.

Just took a 60 minute nap. This is my brain as claude code: $ brain run nap --duration=60m [00:00] Initializing sleep context... [00:02] N1: Entering shallow context switch → Suspending foreground processes (conscious thought, voluntary motor) → Flushing working memory buffer to hippocampus → Hypnic jerk detected — rollback attempted, suppressed [00:10] N2: Core consolidation loop started → Sleep spindles firing: replaying today's sessions in compressed form → K-complexes: noise suppression enabled, ignoring external interrupts → Tagging recent memory chunks: multi-omic_monitoring work flagged as HIGH → Moving short-term → long-term storage: schema integration in progress [00:20] N3 boundary: Deep slow-wave detected (partial entry) → Glymphatic system: running garbage collection → Flushing metabolic waste: beta-amyloid, adenosine, oxidative byproducts → Synaptic downscaling: pruning weak connections from today's overload → Stress hormone (cortisol) queue cleared [00:35] Ascending back through N2 → Memory consolidation: writing enrichment session to long-term index → Problem background thread still running: "why does multi_omic_sequencing() still fire when cache is empty" → solution may surface on wake [00:55] N1: Decompression phase → Restoring sensory monitoring → Prefrontal cortex coming back online (slowly) → Emotional regulation module reloading [01:00] Wake triggered ✓ Adenosine debt: reduced ✓ Working memory: cleared and ready ✓ Consolidation jobs: completed ⚠ Sleep inertia: ~10min until full throughput restored ⚠ Incomplete N3 cycle — full garbage collection requires 90min run → Next scheduled run: tonight 22:00 Estimated full defrag: 7.5hrs Don't skip it.

Every animal with long-term memory sleeps. So does my LLM. Fast weight edits gave instant recall, then broke: 0.92 at 13 facts, 0.57 at 14. Sleep repaired them and moved them into weights: 40% recall to 100% in 4 cycles with no PPL drift. Fast learning and durable memory need separate phases.