Amazing! The 2009 Lincoln & Joyce paper you cited catalyzes one bond per hour on average. (Doubling time = 1 hour, but only one bond between oligonucleotides needed to double.)
OP's Gianni et al 2026 paper connects 45 nucleotides, taking 72 days (1700 hours) to yield 0.2%.
The latter effort is like drawing the whole owl.
That is incredible patience. Without access to the full article, I read only the abstract. I wonder if they used simulations to narrow the candidates?
I learned Perl after trying C; and after struggling with `scanf` (not even getting to tokenization), the ease and speed of `while (<>) { @A = split;` for text-handling made it easy to fall in love. This (in the mid 90s, before Java, JavaScript, and C++ TR1) was also my first contact with associative arrays.
I was also drawn to the style of the Camel Book.
More than most other languages, Perl encouraged one-liners. When I later read PG's "Succinctness is power" essay, I thought of Perl.
which for 2019 describes "0.5 accidents per million departures" and "40 fatalities per trillion revenue passenger kilometers". Considering that many or most passengers fly close to 800-1000 km/h, we're still quite a bit above above 1 fatality per 100 million passenger hours.
Would a factor of 10 be enough? Suppose we go from one major accident per fortnight to one per five months (10 fortnights). Is that higher than what we have seen in the past thirty years?
Not a lawyer, but liability waivers may not apply if there is determined to be gross negligence or recklessness.
Making a reasonably-designed API available, only if connected to an inaccessible network, doesn't sound dangerous, but the goodwill gained might be hard to weigh against a miniscule chance of malware, which would revise everyone's opinion of the degree of negligence or recklessness.
Can confirm. US passport photos want a neutral expression and explicitly say (not in the below page, but elsewhere during the renewal process) not to smile.
Yes! Automobile fuel efficiency is now probably the classical example of Jevon's paradox: some cars are more fuel-efficient; and (therefore?) we in the United States are driving 35% more miles each year in 2024 (3.3e12) than in 1994 (2.4e12).
Nice! One nit: if playing an archive game (from a previous date), when opening the calendar, would you show not the current month, but the month of the game? This would make it easier to go back to other dates.
When I subscribed to Kagi, my block/lower/raise/pin lists were very highly correlated with these aggregate ones.
It makes me think that for Kagi customers, search engine rankings optimize for something other than useful sites such as docs.python.org and cppreference.com
Would you say more? Which part is the violation - the recording, transcription, storage, scanning of other data, interpretation, or something else? Does it matter if you are in a one-party or two-party (for consent to recording) state?
If a police officer is using a body camera and you happen to be speaking nearby, is that a violation of your privacy? What about the same situation, but the body camera user is not a police officer?
363 BB tons / (1 ton / cu m) / (1 BB cu m / cu km) is the mass of a 9-km-diameter ice ball, smaller for rockier materials.
363 BB tons / 10 years = 100 MM tons / day.
By comparison, Mount St Helens (1980) and Eyjafjallajökull (2010) each produced 500 MM tons of ash. So the amount of material is about equivalent to having a similarly-sized volcano eruption every few days for ten years straight.
The Earth's surface area is 500 MM sq km. If we were to distribute 363 BB tons uniformly, that would be an average of 1000 tons per sq km. If the layer had a density of 1 ton per cu m (same as water), the resulting layer would be 1/1000 of a meter thick. This seems to be thinner than the spherule layers described.
Interesting, I initially thought this factor was probably negligible, because the cohort was aged "62.8 ± 7.8 years" (either before or after the 7-year observation period). I still think it's small, but perhaps not negligible.
I estimate that some 2-4% of the group could be heavily involved in caring for grandchildren, sorting themselves into the irregular category. It's well-known that school-aged children pass flu to grandparents [2], and then grandparents die, just in time for the 7-year post-birth observation window. The absolute death rate due to flu is 10^-4 to 10^-3 per year, which would be visible on the paper's mortality time course chart.
Estimation details:
* some 20% are grandparents (and very few are parents) of young children [0]
* 20-50% of grandparents care for grandchildren regularly [1]
So maybe 4-10% of the cohort as an upper bound. If the birth rate is 12 per 1000 per year, and babies cause sleep disruption for about two years to two persons, then that's about 4% also, but perhaps mostly not the same age group.
I speculate your comment has been downvoted because you haven’t explained why you’d “like to see generative AI try to reconstruct the full image”.
What would be the purpose of that reconstruction? To help visualize the most likely original scene using a contemporary-biased lens? Why would we want that?
Imagine you found a box of macaron sandwich cookie fragments, and your model only knows about Oreos. Would such a reconstruction have value? Could it also apply inappropriate bias?
On the other hand, the right models can be critical. Look at how researchers created an image of black holes. [0] This is a reconstruction, and it is an average, and it relies on physics models; but those assumptions are in some ways more transparent than generative AI.
Agreed, the presentation of summarized statistics in the web page seems intended to shock, and it’s hard to reconcile some numbers:
According to the Census 2020 Current Population Survey[1]:
* There are 44.9 MM foreign-born in the US, of which 95.9% are age 15 or more. Of the 39.5 MM (88% of foreign-born) age 25 and over: 13.6% do not have/86.4% have at least 8th-grade education; 75% have at least high-school diploma.
* Total population of US is 331.45 MM, of which 81.4% are age 15-and-older, so foreign-born is 13.5% of total, or 16% of 15-and-older.
If we believe that 21% of 15-and-older lack literacy proficiency, and 34% of those are foreign-born, then we multiply to get: 7% of 15-and-older are both foreign-born and lack literacy proficiency.
But then comparing with the CPS, 7% of 15-and-older is nearly half of the 16% foreign-born fraction of 15-and-older, and seems inconsistent with the CPS statistic that more than 69% of 15-and-older foreign-born (75% of 25-and-older, which represent 88% of all foreign-born) have at least high school education.
There are some basis risk problems (literacy statistics vs educational attainment; 2024 data vs 2020 data); but I would think those are probably rounding errors. I suspect the top-line 21% number is inflated.
https://fleuret.org/francois/lbdl.html