A year or eighteen months ago, if you were to ask me how much I would personally pay to have access to a then-frontier model for work, I would have said: not much. They were frustrating to instruct and use. The cost of play was too high.

Six months ago, I think I would have paid $100 per month.

Today, for a current frontier model and harness with all-I-can-eat, I think I would pay a big chunk of my salary. Easily more than $2,000 per month.

Everything improved: models, harnesses, and instructions from me.

This makes me realize that I have no advantage over LLMs in general software development.

I like Stratego a lot. Next time, I will try this and put three bombs in the other corner to keep the opponent guessing and wasting an expected 4 extra moves.

I speculate a big part of the game is moving your power pieces (1, 2, 3) to places where they can actually attack the opponent safely. Any other placement of the flag and bombs creates a bottleneck for left-right movement on your side.

     -L      Display the logical current working directory.
     -P      Display the physical current working directory (all symbolic links resolved).
     If no options are specified, the -L option is assumed.

https://fleuret.org/francois/lbdl.html

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.

https://paulgraham.com/power.html

https://www.nytimes.com/2010/11/30/nyregion/30bigcity.html

https://archive.ph/vQkwl

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?

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.

https://travel.state.gov/content/travel/en/passports/how-app...

https://fred.stlouisfed.org/series/M12MTVUSM227NFWA

Which makes our total gasoline consumption about the same as or higher than the 1990s, within 10% or so:

https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=C...

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

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.

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.

[0]: https://assets.publishing.service.gov.uk/media/5eeb975b86650...

[1]: https://www.ageuk.org.uk/latest-news/articles/2017/september...

[2]: https://www.sciencedirect.com/science/article/abs/pii/S00255...

Going to sleep now.

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.

[0]: https://youtu.be/Ol_SB5Zfv-Y?si=O6CL-kjDcgBB0bWm , 4-minute overview

TicketMaster ~ bookrunner/sponsor; scalper ~ underwriter;

The general population cannot participate in IPOs, just like we cannot buy primary-market tickets to popular shows.