Define data structures manually, ask AI to implement specific state changes. So JSON, C .h or other source files of func sigs and put prompts in there. Never tried the Agents.md monolithic definition file approach
Also I demand it stick to a limited set of processing patterns. Usually dynamic, recursive programming techniques and functions. They just make the most sense to my head and using one style I can spot check faster.
I also demand it avoid making up abstractions and stick to mathematical semantics. Unique namespaces are not relevant to software in the AI era. It's all
about using unique vectors as keys to values.
Stick to one behavior or type/object definition per file.
Only allow dependencies that are designed as libraries to begin with. There is a ton of documentation to implement a Vulkan pipeline so just do that. Don't import an entire engine like libgodot.
And for my own agent framework I added observation of my local system telemetry via common Linux files and commands. This data feeds back in to be used to generate right-sized sched_ext schedules and leverage bpf for event driven responses.
Am currently experimenting with generation of small models of my own data. A single path of images for example not the entire Pictures directory. Each small model is spun akin to a Docker container.
LLMs are monolithic (massive) zip files of the entire web. No one really asking for that. And anyone who needs it already has access to the web itself
Physics is ageist. Our financially engineered economy GenX is currently benefiting from is ageism against youth.
GenX is the only generation that has not soured on Trump since he started his second term. They're absorbing the economic shocks just fine.
I am an older Millennial who has worked with GenX for two decades.
They have been the worst colleagues. More often sexist, racist, classist, nihilistic than Boomer colleagues who kept it polite.
I can't recall a Millennial or GenZ coworker ever acting like any of those things.
GenX make up a huge chunk of the anti-woke crowd. Hop on TikTok and witness all the 50+ Kid Rock Stan's complain about blue haired lesbian GenZ women and trans people.
Millennials and GenZ aren't pushing those memes with the same numbers.
I have been paying attention.
Dave Chappelle, Foo Fighters, Will Smith, Brad Pitt; the backbone of the entertainment industry the last 25-30 years have little influence on our culture?
Feels like you're projecting your own out of touch with reality as mine
Only after we acquire language from sensory experience first.
It need not be language as we know it that fosters those outcomes either.
What you describe is reinforcement education which can be achieved without our language, without the word "blue" we can still see the portion of the visible light spectrum that we associate to the specific word.
Because we don't experience reality through language but direct sensory perception. Language is arbitrary bird song and visual representations dragged forward from history, accepted definitions never uniformly distributed.
Testing based on contextual correctness makes no sense when there is no center to the universe. No "one true context to rule them all".
We learn from hands on sensory experiences. Our bodies store knowledge independent of the brain; often referred to as muscle memory.
Gabe Newell mentioned this years ago; our brain is only great at some things like language and vision processing but the rest of our body is involved in sensory information processing too: https://en.wikiquote.org/wiki/Gabe_Newell
The most potent evidence the brain is not the center of the universe we commonly think it to be is that patient with 90% of their skull filled with fluid while they carried out a typical first worlder life: https://www.sciencealert.com/a-man-who-lives-without-90-of-h...
Shortly after Trump started to screw up the economy, Wyden stated Trump had ruined an economy that was the envy of the world.[1] The problem is that envy of the world economy was built entirely on decades of deflation of the average persons buying power.
Other world leaders were envious alright. They wanted the secret to how US leaders convinced the public to accept runaway inequality, where it now takes $800k/yr to have the buying power of $200k/yr in 1980.
He’s doing well with the whole investigating Trump angle
but like the rest of the Dems he's signed off on looting workers.
"Do what we say or lose your income" does require blind faith in the correctness of the hierarchy. The business model and whether or not useful work is being done or if the biological is just shuffling capital around as political rules allow.
It requires ignoring management is just another random person, wielding fiat authority. Physics has not imbued them with special properties. It's allegiance to made up semantics.
Everyone ran towards this Brave New World based on media fueled populism.
To me religion isn't Christianity or Islam. It's following orders of arbitrary leaders who give themselves titles via narrative. Priest, Minister, CEO, General... just words.
Provenance such as "this is what I want to do with my life" are poor justification for enabling it.
The old economy is never sustainable because the people it props up die.
Growth is slow but collapse is fast because it takes decades for those people to build, earn their status.
With our eggs in one basket, a small group of elders, they all die off within just a decade or so of each other. A much faster process than the 30-40 years it took to for them to grow their worth to trickle down on us.
Entropy tears apart all structure. Its mechanism for tearing apart society is generational churn.
Time is non-linear. No thing has the same epoch and erodes at the same tick. Endless linear economic growth will never be because once dead belief the elders were rich has to be rethought.
Remember though this forum is full of people who consider code objects when it's just state in a machine.
We have been throwing away entire pieces of software forever. Where's Novell? Who runs 90s Linux kernels in prod?
Code isn't a bridge or car. Preservation isn't meaningful. If we aren't shutting the DCs off we're still burning the resources regardless if we save old code or not.
Most coders are so many layers of abstraction above the hardware at this point anyway they may as well consider themselves syntax artists as much as programmers, and think of Github as DeviantArt for syntax fetishists.
Am working on a model of /home to experiment with booting Linux to models. I can see a future where Python in my screen "runs" without an interpreter because the model is capable of correctly generating the appropriate output without one.
Code is ethno objects, only exists socially. It's not essential to computer operations. At the hardware level it's arithmetical operations against memory states.
Am working on my own "geometric primitives" models that know how to draw GUIs and 3D world primitives, text; think like "boot to blender". Rather store data in strings, will just scaffold out vectors to a running "desktop metaphor".
Define data structures manually, ask AI to implement specific state changes. So JSON, C .h or other source files of func sigs and put prompts in there. Never tried the Agents.md monolithic definition file approach
Also I demand it stick to a limited set of processing patterns. Usually dynamic, recursive programming techniques and functions. They just make the most sense to my head and using one style I can spot check faster.
I also demand it avoid making up abstractions and stick to mathematical semantics. Unique namespaces are not relevant to software in the AI era. It's all about using unique vectors as keys to values.
Stick to one behavior or type/object definition per file.
Only allow dependencies that are designed as libraries to begin with. There is a ton of documentation to implement a Vulkan pipeline so just do that. Don't import an entire engine like libgodot.
And for my own agent framework I added observation of my local system telemetry via common Linux files and commands. This data feeds back in to be used to generate right-sized sched_ext schedules and leverage bpf for event driven responses.
Am currently experimenting with generation of small models of my own data. A single path of images for example not the entire Pictures directory. Each small model is spun akin to a Docker container.
LLMs are monolithic (massive) zip files of the entire web. No one really asking for that. And anyone who needs it already has access to the web itself