I can’t imagine having a hobby that involves passing by, and in some cases climbing over, the exposed remains of others who died doing that same activity.
In other words STPA is a design review framework for finding some less obvious failure modes. FMEA is more popular but relies on making a list of all of the knowable failure modes in a system, but the failure modes you haven’t thought of don’t make it on the list. STPA helps fill in some of those gaps of failure modes you haven’t thought of.
A notebook is a REPL with an inline wiki. Of course it is not intended for running production code, it’s just an R&D environment to document, share, and test ideas
Article mentions people were raiding beaches for free building materials like sand and stone for concrete, which would be a problem at a large volume. Enforcing the same rules on individual souvenir collectors seems excessive.
Way too soon! He recently did an interview with Dave and Krist from Nirvana about the 30th anniversary of In Utero. They described recording prank phone calls during the recording sessions.
Strongly recommend watching Andrej Karpathy’s “Lets build GPT-2” videos on YouTube which dives into an actual PyTorch implementation, then download the code and study it carefully. Then study “Spreadsheets is all you need” to see what the internal data structures look like.
LLMs have a limited context size, i.e. the chat bot can only recall so much of the conversation. This project is building a knowledge graph of the entire conversation(s), then using that knowledge graph as a RAG database.
OpenSearch perhaps? The search query results returns a list of hits (matches) with a text_entry field that has the matching excerpt from the source doc
That’s pretty much correct. An LLM is often used rather like a forecast model that can forecast the next word in a sequence of words. When it’s generating output it’s just continuously forecasting (predicting) the next word of output. Your prompt is just providing the model with input data to start forecasting from. The prior output itself also becomes part of the context to forecast from. The output of “think about it step-by-step” becomes part of its own context to continue forecasting from, hence guides its output. I know that “forecasting” is technically not the right term, but I’ve found it helpful to understand what it is LLM‘s are actually doing when generating output.
For this type of unikernel project C makes sense. I’m a fan of both C and Rust. I like that Rust prevents typical code safety problems, but I like that C hardly changes over time whereas younger languages like Rust are constantly changing. It’s plenty possible to write correct, clean, memory-safe, and understandable code in C, especially if verified by extensive fuzz testing and code scanning.
I wonder how deeply the board and the rotating CEOs actually understand the technology and practicality of achieving AGI, or whether they are falling over themselves due to buying into the LLM hype.
I can believe it. There are those skilled at performative intellectualism, i.e. sounding very smart and insightful, but the performance itself is the product, it is all they sell and offer. Whenever I hear a public speaker appealing to supposed ancient wisdom of obscure tribes and traditions who had better-than-modern diets, laws, monetary policy, medicine, footwear (or lack iof), etc then that tells me that it’s just more hucksterism
Lookup edge service providers like Cloudflare or Akamia. If you’re hosted by a large cloud provider then look into their free and premium edge services offerings. Traffic can be filtered to allow legit users, and consider scaling up servers too if cost of downtime outweighs cost of edge services and scaling up.
What really turned developers away from Apple after the Apple II was not only the price but that you couldn’t sit down at a Macintosh and start programming it like you could with an Apple II. You couldn’t really do that sitting in front of a PC either unless you were happy with GW BASiC, or you bought a Borland compiler, but at least there were less obstructions in your path if you wanted to develop a PC application versus an Apple application. The Macintosh felt like a locked down black box, like you bought a car with the hood welded shut. Overall, PCs were just much more accommodating to developers than the Apple products until Steve Jobs came back and switched everything over to UNIX-based operating system.
The added complication is also because insurers refuse to pay providers unless every procedure, billing code, time of service, justification, physician notes, physician network accreditation, etc, etc is to their standards (which changes often) Health care providers have teams of full time admins who doing nothing but chase after insurers for missing payments and billing discrepancies. Insurers will delay payments, change billing codes, and refuse to pay pending audit, and then may decide not to pay at all. Doctors are pulled into doing extra admin work mostly because insurers wont pay them if the paperwork is not to their standards, or the patient doesn’t qualify.
Concurrency is like a single cook preparing many different recipes at the same time, and parallelism is multiple cooks in the same kitchen. Single-threaded processes can use async/await to context switch and it feels like parallelism but it's not unless you're executing each task to its own OS thread.
... and that didn't click for me until I understood that concurrency (async/await) and parallelism (multiple OS threads and processes) are different executions modes which the Tokio runtime in rust allows you combine those different modes.
Good preview of the near future of software dev, but I’m also wondering of the trend of companies forbids use of AI generated code do to copyright ambiguity. I suppose pressure to reduce cost as and move faster will overcome that concern.
That’s the whole point: the cost of commuting to the job plus taxes plus mandatory meal credit makes the entire job barely profitable. On top of that you’re always just one car breakdown away, or one illness away, or one gas price increase away from going broke.
I just ask ChatGPT at least 5 times using different prompts then take the most frequent response. The result is aligned correctly as long as you don’t resize the browser window.
Kids need to be taught how to use ChatGPT because it’s here to stay and they will use it, but it’s also important to demonstrate that ChatGPT is both useful and senile, and so you need to validate its responses with secondary sources.