voluntaryXchange

Poor Editing at the Washington Post

Not sure whether we should call this illiteracy or innumeracy

Admittedly, it’s a hard graph to interpret, and following the link helps, but it turns out it has nothing to do with the statement that the smoke is “… a thousand miles above the Pacific”. With this word order, a better choice for above would be “over”, and a best choice would be “across”.

Innumeracy: Beirut Explosion Edition

Reports are that the Beirut explosion was caused by 2,750 metric tons of Ammonium Nitrate.

That’s 2.75 Kilotons.

BUT, ammonium nitrate is not as explosive as TNT. In fact, it’s 0.42 times as explosive.

Equivalent tons of TNT is how a nuclear weapon’s power is estimated, so this is roughly 1.3 Kt. (Got to factor in that it’s metric tons, and Kt are measured in American tons too). That’s a lot smaller than Hiroshima or Nagasaki, but a lot bigger than the tactical nukes that the big countries armies have intended to use on the battlefield for decades.

So, here’s the innumeracy:

… a blast the equivalent to three million kilotons of TNT …

This innumeracy is from the Daily Mail. Not a great source, but certainly a popular one with enough money to employ journalists who know a thing or two.

Or not.

The author doesn’t appear to know that:

• A ton of something (even something explosive) does not mean the same thing as a ton of TNT.
• A ton of ammonium nitrate does not have the same power as a ton of TNT, or
• Most egregiously, that megaton and kiloton are variations on the same measure, and are almost never used to compound one another. When we’re talking about megatons of kilotons, we’re in the scale of all explosives on the planet, ever, all in one place.

Three strikes and you should be out.

A Puzzling Comparison

The headline this past weekend was that three fires in Australia have merged into one the size of Manhattan.

This is wrong on more than one level.† Here’s Business Insider:

A wildfire the size of Manhattan is currently burning through the Australian state of Victoria, CNN reports.

The blaze was formed when three separate fires joined up on Friday night, resulting in a blaze 23 square miles wide. The New York borough of Manhattan is just under 23 square miles in area.

Where to start?

How about “23 square miles wide”. That makes no sense. It’s either 23 square miles, or 23 miles wide. It could be both, but only if it’s 1 mile deep (which is possible, given how brushfires work). Which one is it?

If it’s 23 square miles, that works out to 14,720 acres, which is indeed a bit larger than the island of Manhattan. But, is that big? Around here, we had a fire 4 times that large 3 summers ago. And, overall, the fires in Australia have burned about 14,700,000 acres this summer … so this Manhattan-ey fire is just 0.1% of that, which isn’t that big.

Better context is provided by this document from the Congressional Research Service showing that this is more acres burned than in any fire season in the U.S. in the last 30 years (50% more actually). And, the U.S. numbers include Alaska, even though Australia is roughly comparable in size to just the contiguous 48 states.

† Nothing here should be construed as minimizing the horror of this fire season. I just find the grasping for context problematic … many have been wondering for decades what’s going to happen when journalism is filled with the clueless.

Better Algebra Forces an Election

In Guyana, a majority no confidence in parliament forces an election for president.

They had a vote last December, 33-32 in favor of no confidence.

The sitting president, who lost the vote, argued that this is not a majority, and got a judge to agree with him!

How so?

For the sake of exposition, let’s define an operator called “odd_to_even” that adds 1 to an odd number, and leaves an even number as is.

I’m going to use this because there were an odd number of member of Guyana’s parliament. Let that be n=65. So you’d need 32.5 or more to get a majority, right??

The president said no. He claimed that the way to determine the number of votes for a majority in a body with n members went like this:

odd_to_even(n)/2+1

A Guyanese judge went along with this.

But, Guyana (and other countries) doesn’t have a supreme court like the U.S. Instead, they share one called the Caribbean Court of Justice. And it ruled that this is silly, and the correct rule should be:

odd_to_even(n)/2

The latter works out as:

(n+1)/2 =(65+1)/2 = 66/2 = 33

The earlier formula works out as:

(n+1)/2 + 1 = (65+1)/2 + 1 = 66/2 + 1 = 34

I have seen some commentators say this is an order of operations issue, but I don’t really think so. Others also invoke the formula for the median of an oddly sized sample, but again I don’t see the relevance. To me it’s just an algebraic operator issue.

Via Political Calculations.

Utah Beer Law Innumeracy

Utah is relaxing its well-known laws limiting alcohol content in beer. The change takes place on November 1, 2019.

State level alcohol laws are a hodge-podge (the link also has some details about sub-state level laws, which Utah does not have). So while Utah’s laws are different, they’re not that unusual if you travel around the country. For one thing, Utah never had package stores for beer, so in that sense we were more liberal than many states for quite a while.

Anyway, here’s the current system:

Beer is sold cold in grocery stores. It is limited to 3.2% ABW. Bars and restaurants that sell just beer sell 3.2 beer.

Beer is also sold warm in state (liquor) stores that close in the early evening and on Sundays. It’s ABW percent is unlimited. Bars and restaurants that sell liquor can sell beer with higher ABW.

The thing is, most localities measure alcohol by volume (ABV). But they’re not the same. Check out the conversions:

3.2% ABW = 4.0% ABV

4.0% ABW = 5.0% ABV

Most beer in most localities is marked as 5.0% ABV (the bold above). But beer in Utah in grocery stores has been limited to 3.2% ABW (the italics above).

The innumeracy is that people compare the bold with the italics and conclude that right now 10 beers in another state is roughly the same as 15 beers in Utah (10 x 5.0 ≈ 15 x 3.2).

But that’s not right. You have to make the units the same before you do the comparison. So 12 beers in other states are equivalent to 15 beers in Utah (12 x 5.0 ≈ 15 x 4.0 in ABV) or (12 x 4.0 ≈ 15 x 3.2 in ABW).

What Utah is changing to is a system in which alcohol content is measured by ABV (to make inter-state comparisons easier), and also one in which beer with higher alcohol is sold in grocery stores.

The bottom line is that it is a change, but not really a very big one, and definitely not as big as most people think.

I’m not much of a beer drinker, but there’s a bit of smoke and mirrors going on here. Most people are worried about the change in the number, but the bigger sociological change is the increased availability of beer with more alcohol that’s sold cold for immediate drinking. So drinking in Utah now requires less prior planning. I’m not sure how big a deal that is to most people.

N.B. I find it ironic that so many cases of innumeracy revolve around not getting the units right, rather than not dealing with the numbers properly.

Relying On Innumeracy to Perpetuate the Fib

Women work more than men. Everyone knows that. Except the data don’t support that, unless you’re really selective with the categories you call work.†

Of all places, Men’s Health came up with this whopper (the article appeared in my feed from Pocket):

… In 1965, women spent on average 42 hours a week on chores and childcare and nine hours on paid work. In 2016, they spent 32 hours a week on those domestic duties and almost tripled their paid work hours. Men today, by comparison, spend 43 hours a week on paid work and only 18 on housework and childcare.

Let’s unpack that (and in their defense note that the article was mostly about time spent doing household chores, where women still do more). The key point is that womens’ hours of work outside the home nearly tripled from nine. That means 27 at most. So women do at most 59 hours of work inside and outside the home. Add up the men’s total and you get 61.

Face palm.

† If you’re interested in how men and women in the U.S. actually spend their time, consult the American Time Use Survey. Men and women spend their time differently, but it’s pretty hard to argue that women more. I am not sure what source was used by Men’s Health.

Ridiculous Optimism or Innumeracy

We got a new building. It has water bottle fillers:

Note the counter. It took about a month for us to hit 1,000. That pace has been sustained.

But that display has 8 digits.

If it took a month to get to the 4th digit, it will take about a year until the 5th digit turns from zero to one.

And thus, about a decade to turn the 6th digit, a century to turn the 7th one, and a millennium to turn the 8th one.

By analogy, this is like a car’s odometer going up to 99 million miles, rather than the 6 digits (on newer cars), or the 5 digits on older ones. Next time I drive to the sun I might need that.

It’s Time for Not Just Legal, but BRANDED Opiods

Tim Worstall makes the point that drug legalization is now about quality control: the stuff is so strong that it better be dosed in a lab because it’s close to impossible to do on the street.

Fentanyl related deaths are soaring. Check.

Here’s the old school problem. It costs $5K to make a Kg in China, and it’s not hard. That sells for $50K in America. That can be divided into 330K doses. Check.

The new school problem is not how much money that is worth. The problem is 330K doses means each one is 3 micrograms.

YOU do not know how big that is. Here you go: imagine taking an eyelash and dividing it into 25 pieces. Each one would be 3 micrograms. Here’s the equipment you’d need to just weight that.

Alternatively, you could cut your kilo. Here’s how that would work:

• Start with 1 Kg uncut, and add 1 Kg corn starch.
• Now take your 2 Kg, and add 2 kg more corn starch.
• Take your 4 Kg, and add 4 Kg more corn starch.
• Take your 8 Kg, and add 8 Kg more corn starch.
• Take your 16 Kg, and add 16 Kg more corn starch.
• That gets you to something that’s still 3 times as strong as pure heroin. But, whatever:
• Now divide that 64 Kg in two, and
• Divide each of those 2 piles of 32 Kg in two, and
• Divide each of those 4 piles of 16 Kg in two, and
• Divide each of those 8 piles of 8 Kg in two, and
• Divide each of those 16 piles of 4 Kg in two, and
• Divide each of those 32 piles of 2 Kg in two, and
• Divide each of those 64 piles of 1 Kg in two, and
• Divide each of those 128 piles of 500 g in two, and
• Divide each of those 256 piles of 250 g in two, and
• Divide each of those 512 piles of 125 g in two, and
• Divide each of those 1,024 piles of 62.5 g in two, and
• Divide each of those 2,048 piles of 31,250 mg in two, and
• Divide each of those 4,096 piles of 15,625 mg in two, and
• Divide each of those 8,192 piles of 7,812 mg in two, and
• Divide each of those 16,384 piles of 3,906 mg in two, and
• Divide each of those 32,768 piles of 1,953 mg in two, and
• Divide each of those 65,536 piles of 976 mg in two, and
• Divide each of those 131,072 piles of 488 mg in two, and
• Divide each of those 262,144 piles of 244 mg in two, and
• Divide each of those 524,288 piles that are the typical street size of 0.1 g that goes for $10.
• Oh … and don’t make any mistakes … and don’t breath in the dust or touch it with your bare skin.
• And yes, the arithmetic doesn’t work out exactly, so you’ll have to fudge it a bit.
• Oh … and … I have a Ph.D. and I did this math incorrectly by hand and had to set up a spreadsheet to make sure it checked.

What our society does is presume that a bunch of minimally educated drug dealers can do this, with quality control provided by a bunch of addled users. Because it’s illegal to do it otherwise.

Get a clue: we haven’t lost the drug war so much as we’ve lost at the math problem. This is now an innumeracy issue.

And it can’t get better since the chemists are not done finding fentanyl variants that are stronger, and make the math tougher.

I’ve been in favor of legalization (on the demand side) since the early 1980’s. Now I’m stepping up my game: we need legalization (on the supply side) of accurately dosed product for end users.

Technorati Tags: fentanyl,drug,legalization,dose

Innumeracy and Those Sugar-Free Gummy Bears

The vXwife brought them to my attention last night. It’s a big enough deal to make it into The Atlantic. Apparently they … hmmm … give some people lower intestinal disstress (read the comments on Amazon).

I don’t suffer from this sort of problem, but I have good friends that do, so I’m sympathetic.

But, after a bit of reading it struck me that innumeracy is part of the problem. Yes, there are many people that eat a small number of these gummie bears and have problems. But many of those commenters appear to be eating rather large quantities. How large?

• A five pound bag is big enough to be economically priced, but not so large that one person wouldn’t buy it for themselves. That’s about 2,250 grams.
• If you’ve ever cooked with sugar, to get something to the gummi consistency, it has to be almost pure sugar. The squishiness is from a fairly small amount of retained water.
• By the way, it’s technically true that they’re sugar free. But this is because they’re made with Lycasin, which is mostly sugar-alcohols (not sugars, but still carbohydrates). Those are what are hard for some to digest. It’s not as calorie dense as sugar, or as sweet, but if you make a candy out of Lycasin, it’s technically sugar free.
• There’s clinical evidence that Lycasin is problematic: “Consumption of 40 g lycasin HBC or isomalt by adults significantly increased the mean frequency of bowel movements and the number of subjects passing watery faeces.”
• There’s a threshold effect too: “In adults, 40 g isomalt and lycasin HBC provoked significantly more gastrointestinal responses compared to 25 g of either product.”
• Haribo is pretty clear about the serving size:

Note the serving size: 39g, most of which is carbohydrates, but which will clearly come under the 40 g threshold.

This is where the innumeracy comes in. I really am sympathetic to those suffering, but these bags have around 60 servings in them. And a lot of those commenters assert that they ate a substantial part of the bag. I think a lot of people would have some intestinal distress if they ate ten servings of anything.

P.S. I am reminded of the time the vXboy ate the entire package of dried apricots in one sittting. I was sympathetic then too, but clearly he had met the enemy and it was himself.

Technorati Tags: haribo,sugar,free,gummy,gummi,bear,lycasin,malitol

Oil Industry Factoids

You might benefit from these:

The natural gas distribution pipelines in the US alone could stretch from Earth to the Moon 7-8 times. There are millions upon millions of miles of pipe on the planet to distribute crude oil, refined products, and natural gas. (Mostly gas.) Consider this: if your home has natural gas heat, it is connected via a continuous network of pipes to tens of thousands of wells drilled into subterranean rock strata that were laid down tens of millions of years ago. That's pretty cool, really. Your house is directly connected to the Pliocene era -- by the world's oil & gas infrastructure.

I would add that, if you’re not in favor of more pipelines (like the Keystone XL that was blocked by Obama), then you are in favor of tanker trains and tanker trucks going through your neighborhood.

About 40% of all seaborne cargo is oil, and there is literally more seaborne cargo at any given time (by weight) than there are fish in the sea. Oil is in transit for a much shorter amount of time than the lifespan of most fish, so the total amount of oil that moves via water each year is much, much higher than the total amount of fish biomass. Think about what that means for a minute. The ocean isn't full of fish, it's full of oil cargoes.

I tend to think people are ignorant of a lot more than this next bit (and that it could be appied to just about any large industry.

People who think renewables can replace oil with a few decades of Manhattan Project style effort are simply ignorant of how big oil really is.

Even if we assume the energy-storage problem is solved soon …

Oh yes, that energy storage problem. You know … the one that plants solved millions of years ago with … sugars and starches that break down into hydrocarbons. Plants didn’t evolve batteries for a reason; just because we invented them doesn’t mean they won’t always be second best.

Here’s where the public’s innumeracy really starts to kick in:

… Wind and solar are growing exponentially, yes, but from such a small base that it doesn't even make a dent -- the use of renewables as a percentage of total world energy consumption only increased by 0.07% from 1973 to 2009.
Let me break down some numbers.

• World oil production was 82 million barrels per day in 2010. At roughly 6 gigajoules per barrel, that's about 5.7 terawatts of power production.
• World wind power production in 2010 was 0.3 petawatt-hours. Averaged over a year, that's about 34 gigawatts.
• World solar power production in 2010 was 0.03 petawatt-hours. Averaged over a year, that's about 3.4 gigawatts.

Let’s run some math: that makes oil 167 times as big as wind power, and 1,670 times as big as solar. Let’s call those two figures, roughly, 128 to 1, and 1024 to 1. Those estimates are both advantageous to supporters of wind and solar. With those ratios, power from wind will have to double 7 times to match that from oil, while power from solar will have to double 10 times.

Where are ya’ gonna’ put all those windmills and solar panels? Right now, the U.S. has about 49K utility scale wind turbines (you know … the 3 bladed windmills that are a few hundred feet high). That’s one for every 6-7K people. If we’re going to double that 7 times, it means one turbine for ever 50 people or so. That’s far less people than live on most suburban residential blocks. The math is actually better known for solar: about ten 4x4 solar panels per person, all located in the best spots in the southwest, and (uh oh) not operating half the time, or at peak efficiency when most needed.

Even so, for wind power to double 7 times, at the not necessarily believable growth rates experienced in China, would take 70 years. For solar it would take 100. And that’s assuming zero growth in the energy needs of the world, so it will take a lot longer if we actually care about people in Africa and Asia having better lives. It’s also assuming that the oil industry doesn’t grow, so we’re just playing catch up to a stationary target. That seems unlikely.

I like wind power. I like solar. But I can do the math. It’s not pretty.

The quotes are from a thread on Quora entitled “What are the top five facts everyone should know about oil exploration?" Due note that the original has citations to all the factoids. Most of the criticisms in other parts of this thread do not.

Technorati Tags: oil,gas,coal,wind,solar,growth,scale