homerowilson

Many good points, but with respect, electricity costs in the US are near the lowest of any large economy (see, for instance, https://www.iea.org/data-and-statistics/charts/estimated-fin...).

I (American) worked in Spain (Cáceres, Extremadura) ~2015-2017 in tech. It was a wonderful experience. Extremely talented, hard-working, and friendly co-workers. Great health-care and education systems. I think since then rising housing prices partially due to migration have become an issue, but it's a really, really nice place.

Adding

% !TEX program = lualatex

to the top of your document allows you to switch LaTeX engine. This is required for recent accessibility standards compliance (support for tagging and \DocumentMetadata). Compilation takes a bit longer though, but works fine, unlike with Overleaf where using the lualatex engine does not work in the free version.

Swap "solid rear axle" with "De Dion axle" and your description fits the Slate truck pretty well.

"one mushroom species in five is poisonous"? 20% ??? That seems like a crazy high estimate to me, at least if you mean deadly poisonous to humans. In the USA there are only a few species of amanita, galerina, a few of the hundreds of species of cortinarius, maybe some gyromitra and a handful of others I can think of that will kill you. Among the many thousands of mushroom species in the USA, there are only a few dozen known deadly poisonous ones. It's a really tiny percentage. Of course that doesn't mean that the others are edible, just not gonna kill you...

As an educator, one great use for me is classroom use. Students can run R/Python/Fortran in the browser on any OS without installing any software:

https://docs.r-wasm.org/webr/latest/ https://github.com/jupyterlite/jupyterlite https://dev.lfortran.org/

There are rough edges to be sure, but the potential is great in education I think.

You may enjoy this:

https://bwlewis.github.io/cassini/

Very cool. I regularly plug logs with various spawn (lately hericium) and have had good luck covering the plug with natural clay instead of wax, which unfortunately I have tons of in my yard. In my case, the cover is mostly to prevent rodents from eating the mushroom spawn, which they seem to love.

I'm coming from an applied math perspective. A few of my favorites (and ones I find myself regularly referring to) are:

Matrix Analysis by Horn and Johnson (perhaps the best end-of-chapter problem sets of any math book I've encountered!)

Matrix Computations by Golub and Van Loan

Elements of Statistical Learning by Hastie, Friedman, Tibshirani

Functional Analysis by Reed and Simon

as others pointed out i misread the article, which says 2kwh per hour. indeed that is easily do-able.

My partner and I have been living "off grid" using PV + battery for electricity for the last 5 years in the Appalachian mountains. We track energy use meticulously. We try to use mostly wood heat in winter, but also have a buried propane tank for supplemental heat and hot water in winter. So that is arguably not fully off grid? We have a 2,400-sq. ft. A-frame house with R-52 insulation in most walls (closed-cell + batt) and low-e glass windows (but, alas, lots of windows). We mostly cook with electric or the woodstove in winter. Our woodstove is a massive, pretty efficient soapstone masonry stove central in the house by design. South-facing windows are backed by local stone floors that capture and radiate heat in winter. The house is angled to allow sun to enter windows in winter and block sun in summer for passive heating/cooling.

The two of us use just under 5kWh / day average electricity. Our big energy use is heat in winter though! Here is the breakdown:

5kWh/day electric * 365 days = 1825 kWh/year electrons

2.5 cords wood (mostly oak) = 17584 kWh/year wood heat

300 gal. liquid propane = 8400 kWh/year propane heat

That's a total of ~27800 kWh/year for the two of us including heat. That's about 76 kWh/day or 38 kWh/day per person on average. *This includes charging my e-bike for my work commute--but we also have a gas car so not nearly all transportation energy costs.[edited to add this]

I can't see any way to go much lower than that without freezing in the winter. 2 kWh/day seems crazy low to me.

Kersten led amazing research and MonetDB was really ground-breaking in many ways (columnar, shared memory interaction with client languages, etc.). A sad day.

Maybe? There exist some extreme-performance ARM CPUs like A64FX from Fujitsu.

Wow, this was almost my exact experience (except that I had somehow gone through life never using Windows). Using WSL is working out pretty well for me for the most part, except networking seems to break when the corporate VPN is on that I'm required to use (which is very irritating). I find myself prototyping things on my personal Linux laptop and only using the corporate machine for required company stuff. It's been an awful experience.

The current work

https://proceedings.mlsys.org/paper/2021/hash/3636638817772e...

seems mostly about tuning the original idea instead of expanding its scope. But it's still a neat idea. I guess it could be possible to adpt many of the approximations used in the SLIDE idea to GPUs too though...

when i was a kid in the 60's and 70's milkmen delivered milk in glass bottles to your door (a locked insulated metal box at the door in our case), talking away empties to be washed and re-used. it seemed like a pretty good system.

I used OpenBLAS on my cheap last-generation AMD Ryzen 7 4700U laptop like so:

git clone https://github.com/xianyi/OpenBLAS && cd OpenBLAS && make PREFIX=/opt/openblas install && curl https://jott.live/code/blas_test.cc | sed -n "/<code>/,/code>/p" | tail -n +2 | head -n -1 > blas_test.cpp

inspect blas_test.cpp file, and then...

g++ -I/opt/openblas/include/ blas_test.cc -lopenblas -std=c++11 -O3 -L/opt/openblas/lib/ -o blas_test && ./blas_test 512 512 512 100 100

and got a peak of about 192 gflops, averaging closer to 180. So yeah, the M1 is > 6x faster in this simple single-precision matrix test.

I enjoyed this comparison, thanks! Here is a related generally R-centric comparison that you might enjoy of DuckDB, dplyr, data.table, etc. applied to five data-sciency problems I wrote up a few months ago: https://github.com/bwlewis/duckdb_and_r

A nice, short, recent paper along these lines (not mentioned in the article) is "Every Model Learned by Gradient Descent Is Approximately a Kernel Machine": https://arxiv.org/abs/2012.00152

Yes, to the R foundation.

Also my first computer, bought as a kit and I soldered it together. Mine lasted a long time. It was so limited though I learned Z80 assembler, which turned out to be really cool. The ZX-81 was a wonderful machine.

Also don't forget the really interesting Transmeta processors, also ultimately a failure. VLIW is very difficult in practice.

yes, we've had this in Ohio for over a year ($200/year for full EV, $100/year for hybrids).

Another good example of a pretty neat, autonomous personal rail transport system is in Morgantown WV: https://en.wikipedia.org/wiki/Morgantown_Personal_Rapid_Tran...

We foragers and amateur mycologists use smell, touch (slimy, dry, etc,) sometimes taste (bitter, acrid,...), habitat (on wood, ground, type of wood, is there a bulb below ground or a root-like structure, time of year, spore prints, sometimes color change due to drops of chemicals (especially on boletes), sometimes even microscopes to view spores, and more.

all of these variables could of course be coded for a good classification algorithm.

just saying, it's often more than simply visual.

Indeed. We don't need regulation. Nobody is forcing anybody to pay Zoom for the privilege of sending their data to Facebook.

I've been using Jitsi a lot recently and it's great! It's high-quality free (libre) open source software.

That's funny since singularity is now also written in Go!

I must be calculating this wrong, I get:

400ft * 35tons * 2000lbs/ton * 0.00037661609675872 watt hours/ft. lb = 10545 watt hours.

that's only about 10kW hrs, what you get in a just one very compact Sonnen battery (https://sonnenusa.com/en/sonnenbatterie-eco/).

This doesn't seem like much. Did I make a mistake somewhere?

I think this is a good question. Perhaps the author means older than the current position of the continents? The Appalachian mountains, though not as old as the Makhonjwa, would also fall into that category.

compare with the cool vintage Morgantown PRT: https://en.m.wikipedia.org/wiki/Morgantown_Personal_Rapid_Tr...