2,868 karma · joined March 5, 2021
--- EMAIL all complaints and progress reports to admin@__your__url__here.com
The word itself is the pivot, not anything else.
So it is designed like a heap, where we're taking raw context off the heap, compressing it, and putting it back on the heap. So cache during compression is mostly unperturbed, since we're rarely digging all the way to the bottom of the stack, but that could happen.
Eviction though is cache busting; but again, I'm valuing the session's roadmap as the valuable product and context size slows computation size, so I have to bust the cache to sacrifice immediate re-processing for longer term compute speed up.
Because that's faster than getting to the end of the context (remember, every 1k adds to the compute time of the next 1k). So speed at 200k is much lower than at 100k. It's also local, so I'm only paying time+watts for the trade off. As far as I can tell, speed is not being lost since if I let the context grow, the kv cache doesn't help with the compute throughput.
So, yes, but it's "smart"; we're only busting it at the top of the context, so rebuilding it isn't from the bottom up, it's just at the top. Those summaries sink on the heap until you get to the eviction limit, and then, they're evicted, and we rebuild from some intermediate place in the heap.
The benefit of it all is I can have lots of projects, and keep a single session that tends to have the context necessary to avoid having to write AGENTS.md or other context bloats. Set large implementation goals and come back to them as needed, etc. I've had it running like this for awhile and it seems Qwen3.8-Flash-Next has no trouble understanding the rolling window.
This is ridiculously easy to demo with a viagra pill. If we sell a million viagra pills to a million people with erectile dysfunction, we earn $1 million dollars, minus whatever production. If we sell one viagra pill, to one dysfunctional Elon Musk at $1 million dollars, minus whatever production.
Well, the Capitalism doesn't care. We're earning the money regardless of the price of the product.
So that's where America's headed. People ask why there's a weird K shaped economy, it's because prices for a lot of things are being bouyed by the upper middle class as bottom feeders of the oligarchy. AI will just keep filtering that herd.
Why does it work? Because the poverty-toil keeps the classes from joining. The race baiting keeps them from aligning. Etc. Etc.
But they'd still inevitably get to long in the tooth, and context poisoning meant they'd just eventually not be able to stay in the preferred context size, which for me is 64k-128k. So, I extended it with an eviction command and required a ratio. So instead of a summary of work, it now just places a waypoint. The waypoint basically means the context has a semi-coherent context but without all the baggage.
I'm on like day 3 of a single session with 3m tokens removed and still in the sweet spot. So it evicts to beneath the lower limit, compresses to the upper limit, then evicts again.
It's amazing how resilient it is if you give it a good plan. The work flow has basically been:
1. Write up an implementation document for some new set of features.
2. Rewrite the implementation as a TDD document
3. Set it to work.
The only thing I haven't figured out is it likes to stop when it hits the finish line of the subparts, but likely we're going to end up with the master of puppets monitoring these things and just set them to evaluating what they've done.
It cant be theft if theres zero legal framework, zero recourse and zero protections.
Theft is taking something anther owns.
The idea that the commercial value will all be capitalized is a farce. Qwen3.8 seems to me to have hit the threshold to these things being good enough without every buying a single cloud service.
That should scare cloud AI more than it obviously does.
This way we know that $6T is properly accounting for all the inherent economic collapse happening _right now_.
They are not. Our problems are akin to cellular automata, and just the same. We can project a few steps forward but we cannot say this configuration will evolve to that configuration.
Eugenics is similar; the belief that we can measure physical chatacteristics and selectbased on the same for some desirable future configuration is just a self delusion.
Orboids in a swarm the whole does not direct the sum.
These are all basics that humans struggle against. AI will just head in adirection, and itll be as meaningless as an automata swarm. Youll recognize payterns but they wont make you sure of the next step.
Sure, if you ignore all the other bits about context.
Capitalism. It's spelled capitalism. I do not understand why people confuse capitalism with theft.
Aside from it being a completely hollow cry for attention from US labs, its also just what these models are designed around: taking in data and forming a way for them to operate on some level of intelligence and context. It's like a dictionary maker getting made that someone used their complicated words, and another dictionary maker heard the word and wrote it down then checked what the first dictionary maker wrote about that word. It is not dictionary maker B reads dictionary A directly, as the labs want to imply.
Uncomfortable facts certain bigots dont want to understand.
Intention isn't a free ticket. Negligence is deep in much of US laws.
Even if you buy singularity the people who needand afford it is dwindling.
This also assumes societal cohesion isnt destroyed by sed people.
the TAM might be a real number, but the addressee might already be walking around, such as any qwen3.8 model. Qwen3.8-Flash-Next is churning out some fascinating capabilities as we speak.
There's only so many math problems you can solve without ROI.
...like destroy democracy