It is frankly shocking that this was ignored for so long! “The history of leaded gasoline is nuts. Scientists warned it was poison, the factory where it was made was making workers loopy, but GM/Standard Oil enlisted the surgeon general to convince everyone it was safe and rejected alternatives. Massive public harm resulted.” “A Yale physiologist named Yandell Henderson had tested tetraethyl lead as a potential nerve agent during WWI, and when asked his thoughts on putting it into gasoline, he reacted with alarm. ‘Widespread lead poisoning was almost certain to result.’ Later he deemed it the ‘single greatest question in the field of public health that has ever faced the American public.'”
interactively explore how Jeff Dean’s “Numbers Every Programmer Should Know” have changed over time (via Kishore Gopalakrishna)
A good exploration into modern CPU/memory performance behaviour, and profiling same on Linux using “perf stat -d” and “google-perftools”:
Modern CPUs are really good at sequential memory access when it’s possible to predict memory fetch patterns (see Cache prefetching). Random memory access on the other hand is very costly. Advanced data structures are very interesting, but beware. Modern computers require cache-optimized algorithms. When working with large datasets, not fitting L3, prefer optimizing for reduced number loads, over optimizing the amount of memory used. I guess it’s fair to say that Bloom filters are great, as long as they fit into the L3 cache. The moment this assumption is broken, they are terrible. This is not news, Bloom filters optimize for memory usage, not for memory access. For example, see the Cuckoo Filters paper.
Wow, this is cool — babies are born with some “pre-wired” visual connectivity networks, specifically for faces and scenes:
Where does knowledge come from? We addressed this classic question using the test cases of the cortical face and scene networks: two well-studied examples of specialized “knowledge” systems in the adult brain. We found that neonates already show domain-specific patterns of functional connectivity between regions that will later develop full-blown face and scene selectivity. Furthermore, the proto face network showed stronger functional connectivity with foveal than with peripheral primary visual cortex, while the proto scene network showed the opposite pattern, revealing that these networks already receive differential visual inputs. Our findings support the hypothesis that innate connectivity precedes the emergence of domain-specific function in cortex, shedding new light on the age-old question of the origins of human knowledge.
‘on the basis of what we can learn from the very unfortunate experiment that was the Diamond Princess, the coronavirus is probably at most 13x more lethal than the flu and likely a lot less lethal – probably closer to 5x more lethal (.3% CFR).’