I have never worked in an it operations where the skills of the developers was high enough to use eventsourcing - good on paper and for neckbeards - no junior can code up against it
Not my experience. We use event sourcing in production for enterprise level software and inexperienced developers have no problems using it after about 15 min of introduction. Just start with your bank account list of events as an example and build from there.
I’m confused this seems exactly like a skinned glm 5.3 - flash with a few context injection tricks and guardrails - whats going on with these “startups”?
For what it's worth, they claim they applied "large-scale reinforcement learning" to a base of glm 5.3. If that's true, then it's not as trivial as a lot of the "startups" have been lately.
It wasn't chance, it was fundamental to the starting conditions of the two continents and the nature of the exchange.
First there were just more old world diseases which could infect humans than new world diseases. Prior to contact, the old world had about 80 to 90% of the world's total population, so a priori if a disease was to emerge in humans it was overwhelmingly more likely to emerge there. On top of that the old world saw much more domestication of livestock (and particularly livestock animals that had evolved alongside great apes) which in turn created more opportunities per capita for diseases to jump to humans.
For the exchange itself, Europeans were building the ships and sailing them to the New World. It was primarily Europeans on those ships, and especially those passengers who would be most likely to be incubating diseases (the old, young, and infirm) were overwhelmingly more likely to be Europeans headed west. Europeans afflicted with diseases after they reached the Americas would generally not have been allowed on ships back to Europe, or if they were they would likely not survive the voyage. Syphilis likely came from the new world, and it's easy to see how, but if there had been some American equivalent to small-pox it would simply have not been able to make the journey before going extinct as the native populations harboring it were wiped out by Old World diseases.
For it to have been "the other way around" the Americas would have needed a vastly different ecology and the people living there would have needed to beat the Old World to the level of technological development necessary to make contact.
This book came up so many times in school, always as some way to say Eurasians never had any inherent advantages. It's not that I think they do, it's just that I'm wary of people trying to reach a conclusion only because it's politically correct.
unless you plan to read a dozen academic books about a variety of topics, i dont really think there is a better option for a popular audience that will answer the question
"why were Eurasians able to so easily take over the Americas" than this book.
Within five years as many as 15 million people – an estimated 80% of the population – were wiped out in an epidemic the locals named “cocoliztli”. The word means pestilence in the Aztec Nahuatl language.
i think this is a wash at best, but might be ok if you have a very specific model that is completely unoptimized - but these days you could just point your astra level llm at it and say "make this faster"
and the results are what i kinda expected to begin with, this adds next to nothing? - also the repo was pivoted from a playwright sub assembly to this not long ago - so my conclusion - THIS MIGHT be worth some watching if you have a model where no-one!, has optimized it at all - and where it does not use anything native to your platform.
results (averages)
VIA rapid-mlx :8902 (MLX)
Decode: 175 tok/s
TTFT: 64 ms
prefill (~760 tok cold): 594 ms
Magnitude 0.2.1 (GGUF/llama.cpp+Rust)
Decode: 161 tok/s
TTFT: 111 ms
prefill (~760 tok cold): 669 ms
Hi, M5+ Macs have a known optimization gap since we do not fully utilize the Metal 4 matmul operations in our kernels yet - so should be able to do much better on this specific comparison soon!
Nah llm optimizes for the middle of the bell curve - most designers and designs are meh so they’ll output meh across the board
. Takes rare skill to do it right
As a guy who used to do art direction and was a creative director i can with some confidence say that: “most “designs sucks balls” and 95% of the worlds population DO NOT CARE.
I wish it was different but estetics, color sensitivity and for a lack of better word “taste” is a skill that needs training - the kerning alone on those make me sad - but the dirty secret is that meticulous design and craft is a dying breed - why spend days on designing whitespace when you can get something the client likes in 10 minutes from a prompt?
reply