The most official water costs $120k a gallon (signoregalilei.com)

83 points by surprisetalk 5 hours ago

sudb 4 hours ago

Reminds of me of the "most expensive peanut butter" at ~$2.44/g - sold by NIST, I assume for some sort of peanut butter factory process calibration: https://shop.nist.gov/ccrz__ProductDetails?sku=2387

cyphar 2 hours ago

For more context, sample reference materials from the NIST reference catalogue are not meant to be exceptionally high purity examples of the sample -- SRMs are usually commercially-produced products. The real value is that each sample comes with a datasheet with incredibly detailed and accurate measurements that are (as you said) used for calibration of measuring devices in industrial settings. To quote a Veritasium video about the topic from a few years ago[1], "the thing you're paying for is not the peanut butter, it's the knowledge of exactly what is in the peanut butter".

[1]: https://youtu.be/esQyYGezS7c?t=226

buildsjets 3 minutes ago

Truth. I once paid nearly $500 for a set of 17 certified, calibrated wooden pencils, with documentation showing that each one met their rated hardness levels of 9H thru 6B. And for $500, they don't even come with erasers, that is an added-cost option.

https://www.gardco.com/product/pencil-hardness-test-kit-astm...

cjs_ac 3 hours ago

SRM 2387 pairs well with a cup of ISO 3103.

https://en.wikipedia.org/wiki/ISO_3103

buildsjets 3 hours ago

Cap your meal off with a NIST 1196a standard issue cigarette? I am told they taste like drywall and taxes.

https://www.nist.gov/news-events/news/2020/02/nist-releases-...

Perhaps you would prefer some standard issue meat.

https://www.sigmaaldrich.com/US/en/product/sial/nist1546a

sushisource 2 hours ago

tshaddox 2 hours ago

georgeburdell an hour ago

Yes, it’s a bit frightening to know what a hidden dependency NIST is on many things. In my industry, all calibrations ultimately derive from physical NIST standards for resistance, frequency, etc.

masklinn 24 minutes ago

It’s frightening that… NIST is doing the very thing it was established to do back when it was called the Bureau of Standards?

juleiie 3 hours ago

I wouldn’t put it past people in my country to buy it just to show off wealth :/

I have recently realized we imported the western world in the most superficial ways.

artisinal 3 hours ago

This is why we should ditch Celsius for Kelvin. I never liked how we coupled 0 and 100 to where water freezes and boils under certain conditions. Kelvin doesn't rely on non-invariant reference points.

warmdarksea 2 hours ago

this is actually why i like fahrenheit the most... 0F is "really cold" and 100F is "really hot", whereas in celsius 0C is "kind of cold" and 100C is "uninhabitable". it makes sense for the 0-100 range to be reserved for temperatures frequently encountered in our habitat as humans i think...

kccqzy an hour ago

That’s a biased view that leads to a scale only convenient for people living in particular latitudes. Now talk to someone living near the equator like Singapore, and 0F is unimaginably cold (even 40F is basically unheard of), and 100F is very hot but not extraordinarily so. Then you talk to someone living close to the arctic and the converse is true.

All this talk about 0F to 100F being frequently encountered probably reveals your local climate conditions more than it helps define a useful global standard for all humans. We should accept that humans live in more diverse conditions than we intuit and such a scale probably won’t be useful for all.

cyral an hour ago

al_borland 2 hours ago

This has always been my primary argument in favor of Fahrenheit. It makes sense for humans, which is all a majority of the population really cares about. In addition, it's more granular. Sometimes turning the thermostat up or down by 1C could be too much, where 1F is just right.

In science and math, I can see where having round numbers for commonly used temperatures would be useful, but if 0C and 100C are the only places that happens, it's not that hard to memorize 2 numbers.

Timpanzee an hour ago

vagrantJin an hour ago

wongarsu 2 hours ago

Celsius maps nicely to the temperatures observed in a typical Finnish sauna. 100C is among the highest typical sauna temperatures (very relaxing in dry air), and afterwards you cool off with water around 0C

Also, having 100C as water temp for black tee and 80 for green tee is nice to remember

Paracompact an hour ago

OF will still kill you faster than 100F, and most folks aren't comfortable at 50F.

It inspires the following idea: A nonlinear scale where ~72F is 0, and +X represents the same level of heat discomfort as -X represents cold discomfort.

ajmurmann 2 hours ago

This makes sense if weather is the main focus. For things like cooking Celsius is great though.

0cf8612b2e1e 2 hours ago

clickety_clack 2 hours ago

clickety_clack 2 hours ago

I actually came around to Fahrenheit for the exact same reason. 0°F and 100°F are kind of the top and bottom of the temperature range I’m going to be experiencing and anything outside that is waaay too hot or cold. In Celsius a large portion of people have to flip to negative numbers regularly during the normal year, and everything between 45°C and 100°C is completely unused.

mmooss an hour ago

Timpanzee an hour ago

Except 0F being “really cold” and 100F being “really hot” is a personal bias of yours based on where you’ve lived. It’s not true for everyone or everywhere.

And even if it were true, it still wouldn’t work. 100F in a dry environment is nice while, in a humid environment, 100F is awful. Same for 0F, a dry and static 0F is fine while a wet and windy 0F is killer.

TylerE 2 hours ago

Also for things like hvac temps Celsius is rather too coarse where as 1F is right around the perceivable difference.

SketchySeaBeast 2 hours ago

Guvante 2 hours ago

You want numbers to be understandable and K produces illogical numbers.

(There is a reason it is common to drop the century when talking about years, humans are less skilled at numbers over 100)

And I know temperatures in Fahrenheit go higher but it is common to refer to Phoenix as "over 100" even when it's high can peak over 120.

zinckiwi 2 hours ago

That’s 02025 thinking.

thaumasiotes 2 hours ago

> (There is a reason it is common to drop the century when talking about years,

Yep.

> humans are less skilled at numbers over 100)

No, the reason is that the century is constant and therefore carries no information. Where the century isn't constant, it never gets dropped.

Guvante 33 minutes ago

analog31 3 hours ago

I think that both the Celsius and Fahrenheit scales are now defined in terms of Kelvin and a formula. Most if not all of the units of measure are no longer related to their history except for an agreement to make the new units match the old ones at some point.

0cf8612b2e1e 2 hours ago

The US is technically metric in that the units are defined in terms of the metric equivalent. A US inch is exactly 2.54cm.

throwatdem12311 2 hours ago

Now do metric vs imperial.

artisinal 2 hours ago

Nothing screams freedom more than using ancient terms like horsepower and measuring tires in inches and percentages.

The other day I heard an American talk about a TV being 65 inches. Took me a while to understand they had a 165 centimeter diagonal television.

tancop 37 minutes ago

clickety_clack an hour ago

vel0city 24 minutes ago

alehlopeh 2 hours ago

Jblx2 2 hours ago

xgulfie 3 hours ago

Kelvin does still rely on each degree being 1% of the way between freezing and boiling so it kinda does rely on a variant reference point

chermi 2 hours ago

The reference point is the triple point, and invariant.

xgulfie an hour ago

spullara 3 hours ago

A US gallon of deuterium water costs roughly $2,600 to $3,800, while a gallon of pure tritium water (super-heavy water) would cost about $44 million.

BizarroLand 2 hours ago

Makes me want to go to the ocean and start filtering out water somehow.

inigyou an hour ago

You can centrifuge it. Heavier molecules move closer to the bottom. Repeat until you reach the desired purity.

You need shitloads of ordinary water to extract a liter of heavy water. And I'd be surprised if there's any superheavy water in that, because tritium radioactively decays.

Fun fact: Heavy water is slightly toxic to humans. If you only drank heavy water, you'd die as it shifts the chemical reactions in your body out of balance. A moderate amount is fine though.

danbruc an hour ago

delecti an hour ago

There are a variety of ways to do it, but you'd want to be careful that you don't attract attention from your local nuclear regulatory agency.

https://en.wikipedia.org/wiki/Heavy_water#Production_methods

rwmj 2 hours ago

Interesting article but left me wondering why they don't use ¹H₂¹⁶O as the "standard water"? Surely you can make that (relatively) easily in a centrifuge by spinning out the lightest fraction of water.

thayne an hour ago

At the very least that has the advantage that it's possible to make more of it, so you aren't dependent on a limited sample of standard water.

cactusplant7374 2 hours ago

How does the water taste?

pack_stimulus 4 hours ago

Loved reading this. Thank you!

dorianmariecom 2 hours ago

gallon is not a si unit ;)

sandworm101 3 hours ago

As specialized world-class lab supplies and reagents go, that is not very expensive. Stuff like pure Xe (for space thrusters) can be more by mass.