Stump. I was a tree, but now a dump.
Lunch mates some days ago.
She has grown up in the era of me lying down all the time and has grown into lounging much the same.
Disconnect from their cycle and think for yourself. The news cycle is the primary problem. It is malicious theatrics. The further I have disconnected and distanced myself, for personal reasons, I only catch the periphery, but stuff makes much more sense overall. I write my own opinion and narrative because I have the space to form it.
We need DNS filtering to work on outgoing packages by default, and to make whitelist DNS stupid simple to implement for any parent and child processes. It should be as simple as launching with the command, including a preconfigured whitelist, and a pop-up message for "approve, deny, prepend to list." System wide and incoming packet filtering is insufficient for the modern world.
llama.cpp is at the core of almost all offline, open weights models. The server it creates is Open AI API compatible. Oobabooga Textgen WebUI is more user GUI oriented but based on llama.cpp. Oobabooga has the setup for loading models with a split workload between the CPU and GPU which makes larger gguf quantized models possible to run. Llama.cpp, has this feature, Oobabooga implements it. The model loading settings and softmax sampling settings take some trial and error to dial in well. It helps if you have a way of monitoring GPU memory usage in real time. Like I use a script that appends my terminal window title bar with GPU memory usage until inference time.
Ollama is another common project people use for offline open weights models, and it also runs on top of llama.cpp. It is a lot easier to get started in some instances and several projects use Ollama as a baseline for "Hello World!" type stuff. It has pretty good model loading and softmax settings without any fuss, but it does this at the expense of only running on GPU or CPU but never both in a split workload. This may seem great at first, but if you never experience running much larger quantized models in the 30B-140B range, you are unlikely to have success or a positive experience overall. The much smaller models in the 4B-14B range are all that are likely to run fast enough on your hardware AND completely load in your GPU memory if you only have 8GB-24GB. Most of the newer models are actually Mixture of Experts architectures. This means it is like loading ~7 models initially, but then only inferencing two of them at any one time. All you need is the system memory or the Deepspeed package (uses disk drive for excess space required) to load these larger models. Larger quantized models are much much smarter and more capable. You also need llama.cpp if you want to use function calling for agentic behaviors. Look into the agentic API and pull history in this area of llama.cpp before selecting what models to test in depth.
Huggingface is the goto website for sharing and sourcing models. That is heavily integrated with GitHub, so it is probably as toxic long term, but I do not know of a real FOSS alternative for that one. Hosting models is massive I/O for a server.
tendentious milieu, or mircomilieu
Oh I am at least more aware of the limits than most. No one knows what is wrong with me, and yet my back basically fails after an hour and I am in so much pain by then that I do not trust myself around other unstable people. I'm told stem cell treatments are the only option for mysterious spinal issues like mine, then they tell me that primitive Puritan zombie apes have prevented any such treatment in the USA and I must go to Japan or just die a homeless wreck.
Anyways, in my science fiction universe, I built it on the premise that biology is an engineering corpus, that biology is the ultimate final technology, and it eventually becomes like programming with Python, but many many orders of magnitude more complex. It is fun playing with ginkgo trees used for official structures; temperamental to work with but near infinite in age. Wall surfaces are grown of chitin. Most structures and furniture are grown from the Banyan tree. There is no industrial technology on the living level of the O'Neill cylinders. Culture has shifted to custodians of resources on stellar time scales. The complex human social hierarchy is shifted to skills, guilds, reputation, and accolades. The primary currency of exchange is a heat dissipation allowance. The lack of anonymity in exploiting natural resources, and the resultant heat, render accumulated wealth hierarchies obsolete.
I think my biggest gamble is that there are sufficient incompatibilities and complexity that prevent major modifications to most species. The biggest biological achievement is the synthesis of a deterministic biological computer based loosely on the human brain. That innovation marks the end of the stone age of silicon, several centuries before my stories take place. I play in the first expansion of life outside of the Sol system using generation ships and one way travel. Wild Earth is only inhabited by a few indigenous groups. Most humans live in CisLunar O'Neill cylinders.
All it takes is one m-type astroid. Any such differentiated planetesimal core contains more rare mineral wealth than all of what humans have accessed in the Holocene. If anyone gains access to this, to more wealth than all that Earth's differentiated scarcity has ever given us through the few impact remnants on the planetary surface dross, then we have the catalyst that produces the infrastructure and we head to space in mass within a century. The need to create fully cyclical terrarium worlds without waste drives the research and push to explore biology. Each holdout technology without a biologic solution weighs more heavily as more and more are dropped.
I've been exploring how a volcanic traps like scenario builds up. I have also been working through ways to claim that those that modify their own genetics substantially have never managed to reproduce for more than 10 generations, and the problems that result are horrendous for the more generations that persist. Genetic modifications become the tin foil hat dogma tribalism fodder.
No aliens, no FTL travel. There are kilometers scale self replicating drones that handle the required resource acquisition, any necessary industrial processes, and enable maintenance of habitats on the scale of stellar lifetimes. Interstellar travel is a major gamble where around half do not succeed. Indeed, while the generation ship is built with many of the larger structural components of the first O'Neill cylinder to be constructed at the destination, the small population will collapse and become sterile unless they get the first cylinders spinning within a few generations. The hub of the main habitat is the generation ship, which is spun to a twelfth of standard gravity.
It is all possible because of the largest construction project in history, when most if Mercury is used to create the orbital structures of the forge around Sol. There, generation ships and drones are built around the antimatter fuel required for a one way journey to any stable, type g star within 7 parsecs, and with fuel enough to arrive and slow down on the other side. There is no coming back. It is part of a goal to colonize the galaxy within 220 million years at most; one galactic orbit of Sol. The main connections to the other systems is a large packaged news stream that is, and must be constantly sent to Sol, and the same is sent from all other systems. It is how rogue breakaway colonies are prevented. The big challenge and looming scare is what will happen when the first colony systems grow large enough to produce antimatter and expand. If any system goes silent in the story era, a poison pill of sorts is launched from Sol to disable their autonomous drones and effectively doom the colony. Antimatter fueled weapons are the nuclear option.
I am certainly no expert on any of the underlying subjects. It stems from being a fan of Asimov and realizing the myth of the machine gods is a fallacy that has existed in the cultural zeitgeist since at least the dawn of the industrial revolution. The machine gods myth is our Chaos, Ares, and Typhon that will seem equally nonsensical centuries from now. Fusion will eventually happen, but we are already past the phase resource use where industrial technology will deplete all economically accessible sources. ITER is already using a significant percentage of beryllium IIRC. All present industrial tech is impossible within thousands of years or less. The equivalent biological solutions will last stellar lifetimes.
Anyways, thanks if you made it through my ramblings this far. It is a fun to be a counter culture positive futurist, and conjure a better future history, even if only a dream.
True there is a long path ahead, but O'Neill cylinders solve most issues using 1970's materials science. We are unlikely to survive the next providential scale volcanic traps type eruption like the Columbia Basin, Siberian, or Deccan Traps. We must reach the final technology of fully understood biology and an engineering corpus first. Then we unlock interstellar travel. We are already on a rock-ish ship circling the galaxy. The real challenge is when to get off the ship.
I am stuck in bed around 95% of my day. I am dealing with most of the issues you described.
PLA will be better for hardware store and hobby junk. You cannot use automotive class finishes and expect them to last. Generally stick to one brand. Most paints are formulated for steel. ABS is the closest to steel in thermal properties. The expansion is the most important attribute. PLA has a different thermal profile so catalysed 2-part paints will not work very well long term. Rattle can enamel is junk by comparison, but it never fully cures like automotive paints. That property helps it stay in place longer in general. There are special adhesion promoters like bulldog for automotive stuff, but the thermal properties will still be an issue.
Pro automotive paint is 99.9% sanding and prep work. It is far more intense and rigorous than people realize. Perfection happens in the prep work. The actual paint is just a way of showing off that perfection. Mastering automotive paint is actually all about defeating yourself. Perfection is not subject to your emotions or expectations. It is right when it is perfect.
You want the highest pressure spray cans as possible. Also, if you do not used all of the can at once, flip it upside down and clear the nozzle by letting the siphon into the empty void and spraying. If you have a compressor that does not shoot out a bunch of oil or water, a cheap Harbor Freight pink gun with the nozzle of the can beside the spray gun will work wonders by atomizing the spray far more effectively.
Decompresses my spine and makes me normal again?
Gravity prisons suck. Let's plot our escape when the guards are not looking!
From 20 minutes ago... The sonic boom took 8 minutes to arrive.
I like the hacked breadboards so far. I made a sloppy super-breadboard. I should have glued it before soldering but was worried the glue might make it into the slots with the metal contacts. I also broke up the internal rails and labeled so that 8b data and 16b address all exist on the same set of 4 power rails.
I think I am having an issue with either some crossed wires or how Bus Request works with the Z80 versus how the Bus Enable of the 65x chips work. I think one or all may have some type of routine that does not high-z the buses immediately... a problem for tomorrow.
I watched Anton Petrov's last upload on the impossible merger of intermediate mass black holes,(https://www.youtube.com/watch?v=8p6PgXqL6OQ).
Do two orbiting black holes have a gravitational resonant effect that is different than a single object of an equivalent mass?
view more: next ›
It would have been more fun to paint cars if someone named the mixing colors like this. Yellow is actually quite rare in automotive mixing systems. You never see yellow cars. There are many you might mistake as actual yellow, but nearly all are a trick. There are usually 2 yellows that are used for tinting, mostly with whites. The primary black in automotive paint is actually yellow based. A tiny amount of black with a fine metallic is how most champagne colored cars are formulated.