Category Archives: Web Article Review

Dietary Guidelines for Americans

The USDA has a new version of Dietary Guidelines for Americans out. Sorry, TLDR, but the Harvard School of Public Health has a handy summary (along with some criticism). Basically, fruits, vegetables, and whole grains will never go out of style. Sugar will never be in style again.

I think people of my generation and older are still confused about fat. The guidelines say plant-based oils are pretty much A-OK as long as you stay within your calorie limits, but still recommend “lean meats and poultry”, “low fat dairy”, and limiting saturated fat. First, I am confused whether saturated fat is bad for everyone, even those of us with low cholesterol, or whether the USDA assumes we are too stupid to understand nuances and a blanket statement like this will save lives overall (if so, they’re probably right.) Harvard also criticizes USDA for not discouraging processed meat like bacon and ham (but bacon is so good…well, better to think of it as an occasional treat like a candy bar).

Men should limit alcoholic drinks to “no more than two” and women to one (sorry, ladies). By the way, a(n imperial, 16 ounce) pint of 7% alcohol craft beer is not a drink, it is actually almost two. Whereas 1.5 ounces of 40% alcohol liquor is one drink and actually easier to control. I love those craft beers though. Oh, and don’t touch soda – it’s death in a glass.

But you can have 2-3 cups of (black) coffee a day, with no known negative effects.

You can have more salt than I thought (2300 mg/day) if you don’t have any particular risk factors.

Harvard also points out that the science behind the nutritional benefits of all that meat and dairy is not all that strong, while the science behind the environmental risks is strong, and clear, and not mentioned in these guidelines.

Well, this is the U.S. Department of Agriculture. Not the department of get your ass off the couch, go for a jog, and then eat some vegetables. We have an Environmental Protection Agency, but first of all it is not cabinet level, and second of all they don’t regulate agriculture. Nobody regulates the environmental impacts of agriculture! And the meat, sugar, corn (etc.) and food processing industries are massive, have enormously deep pockets, and use them to buy politicians who will keep it this way indefinitely.

fun with coronavirus math

Let’s do some coronavirus math! This is a word problem, kids. I’m writing on January 14, 2021, and this post will be horribly outdated, but possibly of historical interest, when you read it.

The total number of cases confirmed to date as of today, in the U.S.: “23.1 million+” (New York Times)

The CDC’s ratio of actual cases to confirmed cases: 7.2 (CDC)

Number of cumulative cases in the U.S. so far: 23.1 million * 7.2 = 166 million (166,320,000)

Population of the United States: 328.2 million (Google)

% of our population that has had the coronavirus = 166 / 328.2 = 51%

% of our population that has been vaccinated: 3.1% (Financial Times)

But all other things being equal (which I am sure they are not), 51% of the people vaccinated will have already had the coronavirus, so the vaccine so far adds 1.6% to 51% of our population. Call it 53% to be generous.

We have heard a variety of estimates on what constitutes herd immunity, but the number 70% seems to be sticking at least in the media (I don’t have a source handy, and need to go do some other things now.) So we might not be that far off. The (painfully) slow but steady vaccine rollout tortoise will eventually get to the finish line, people are continuing to get infected at high rates every day in the meantime, and nobody wants to see another wave from the new variant, but if and when it hits us it might push us over the mark (at a horrific human cost, of course).

One last thought is that at the moment, I suspect we are immunizing people who are more likely to have already had an infection than the population as a whole. We are being told this is the most ethical approach, or the quickest way to lower risk for the population as a whole, or some combination of the two. The ethical statement may be true, although this seems subjective. I thought ethics was not up to ethicists, but rather ethicists were supposed to ascertain what our society as a whole considers ethical, and maybe compare that to other human societies past and present. I haven’t seen public polls of what people think is ethical, although they may exist. I can see a case that the way the vaccine is being rolled out is ethical, but I can also see a case for a random lottery being equally ethical.

Better planning and communication would not just be ethical, they are the common sense need and our government is continuing to fail, fail, fail and people are dying, which is the opposite of ethical governance. To my ears, it is arrogant to hear them lecturing us about ethics.

drawing a line from Hitler to climate change

This 2015 Timothy Snyder article is called Hitler’s world may not be so far away. He is a well-respected historian whose previous books include Bloodlands: Europe Between Hitler and Stalin and Black Earth: The Holocaust as History and Warning.

He calls the Holocaust “misunderstood” in the article, but he is not disputing facts or events that occurred. He makes a few points. First, we modern people tend to assume that we are morally superior to Germans of that period, and that we would not allow something like that to happen even under similar circumstances. He says there is no reason to believe this is true. Second, he points out that the worst deprivations occurred not within the borders of Germany or other western European states, but in lawless, stateless areas of eastern Europe. Nazi Germany intentionally created those lawless, stateless areas, but this holds lessons for failed states today, such as Syria. Third, he says that fear about the food supply in the 1930s was a significant driver of Hitler’s policy to expand east, creating space and farm land for Germans while exterminating or enslaving the inferior people who lived there. The so-called green revolution, which drastically accelerated agricultural yields, happened mostly after World War II. (We can argue later whether using massive fossil fuel inputs to produce fertilizer, pesticides, groundwater pumping at rates that will only be replenished over geologic time, and dumping the resulting waste in the ocean was a long-term solution, but it has fed a few billion people successfully for a few decades in a row now.)

So lessons for today are that as the climate crisis almost certainly worsens, we will see failed states, hunger and fear of hunger, mass migration, and these are all risk factors for genocide. I’ll pick a paragraph, but this long article really is worth a read.

Perhaps the experience of unprecedented storms, relentless droughts and the associated wars and south-to-north migrations will jar expectations about the security of resources and make Hitlerian politics more resonant. As Hitler demonstrated, humans are able to portray a looming crisis in such a way as to justify drastic measures in the present. Under enough stress, or with enough skill, politicians can effect the conflations Hitler pioneered: between nature and politics, between ecosystem and household, between need and desire. A global problem that seems otherwise insoluble can be blamed upon a specific group of human beings.

Project Syndicate predictions for 2021

And now the 2021 predictions are starting to roll in. I blew my one free Project Syndicate article for the month on this, which seems like an okay choice.

  • Covid-19 will recede as vaccines roll out, and the economy will recover. This seems to be a near-consensus, although there is one minority report. And the average growth rate of course hides inequalities, which have gotten worse.
  • As you might expect, lots of speculation about U.S. politics and what Biden will do, but most people expect a return to the pre-Trump status quo at the UN, WHO, Israel and Palestine, the Iran nuclear deal, the climate deal, and democracy/human rights rhetoric we mostly fail to live up to. Of course, there are newfound doubts about U.S. political stability in the medium- to long-term.
  • Renewable energy will continue to be cheap and competitive with fossil fuels.
  • Electric vehicles come up a couple times – the market is pulling, and there may be a big push because the U.S. is significantly behind many other countries on adoption. (My take: The electric and auto industries are behind this, and the oil industry presumably is not but nobody seems to care. Could this break their backs?)
  • U.S.-China tensions will ramp up! Or they’ll die down…the crystal ball is murky on this one.
  • North Korea likes to test new U.S. Presidents with a missile test or two.
  • Poverty and violence have gotten worse in Africa while the rest of the world has been distracted by other things.
  • The effects of food insecurity and extreme weather events are getting worse in developing countries.
  • Cash may be dead, and if so there is at least a three-way race to replace it – “private tokens, central bank digital currencies, and efforts to upgrade the current system”.

finally, (some) hard numbers on schools and Covid spread

This article from The Intercept cites some recent research studies that put some numbers behind what level of community spread would make opening schools unsafe. The basic idea is that school (especially elementary school) is pretty safe when the level of infection in the community is relatively low, because kids coming to school are not that likely to be infected. But when the level of infection in the community rises, kids coming to school are more likely to be infected and further accelerate the spread.

Even educated people in the general public have a hard time with unit conversions, and this article switches between various units within the article. Come on, guys. Anywhere, here are the numbers from a variety of sources in the article. I’ve done the unit conversions (correctly, I think, but this blog post does not constitute medical advice…)

  • 36-44 per 100,000 population per week (~5-6 per 100,000 per day)
  • 147 per 100,000 per week (21 per 100,000 per day)
  • 35 per 100,000 per week (5 per 100,000 per day)

That seems like a pretty big range, and I am also suspicious whether the reporters have carefully checked the math, given how they jump around even within the article. But let’s assume they have it right. The threshold is somewhere between 35-147 cases per 100,000 per week. The Pennsylvania Department of Education recommends a threshold of 100 cases per 100,000 per week to consider in-person K-12 school. (Although private and parochial elementary schools have been open throughout the pandemic, and public school districts are hit or miss.) The official number for Philadelphia county at the moment (I’m writing this on January 7), which they only update once a week, is 225.9 per 100,000 per week and falling. My unofficial 7-day running average of the numbers the Philadelphia Health Department reports in its daily press releases is 235.0 per 100,000 per week and falling (but looking at a plot, I would say it’s bouncing around and not clearly rising or falling this week). Those of us with children in public school have not had the option of in-person school so far during this school year.

2020 Human Development Index Report

You could spend a lot of time going through any one sprawling UN report like the Human Development Index Report. Then you could spend a lifetime digging into the underlying sources. Here are a few things I gleaned from a light skim and looking at some of the pictures:

  • Amartya Sen says the index was designed as an alternative to be looked at alongside GDP, and the intent is to identify shortcomings of GDP, draw attention using a single aggregate number which doesn’t really mean much, and then hope news media and individuals dig into the underlying information. He thinks this has been reasonably successful.
  • The index crashed in 2020 due to Covid-19.
  • They have made a significant effort to incorporate ecological risks into the index. There are interesting chapters on planetary boundaries and relationships between the overall level of development and ecological risks across countries. Of course, the countries with higher development levels tend to contribute more to the global risks, which then fall on the countries with lower levels of development. So the goal would be to reduce the impact from the more developed countries, while moving the less developed countries up the development ladder without creating even more risk globally. This is hard to do.
  • Chapter 2 summarizes the magnitude of overall human impacts compared to the scale of the planet’s natural systems, the planetary boundaries concept, and the biodiversity collapse. Not a bad introduction if anyone is new to these issues.
  • Food security risks have increased significantly, and not just due to Covid-19 but due to flooding, droughts, heat, and natural disasters clearly driven by climate change. There are a lot of intertwined issues out there, but if we were going to pick only one to pay attention to globally, this would be it. See pp. 56-58. Flashing warning lights here!
  • There is an essay on existential threats to the species and civilization somewhere towards the back. One way to estimate the risk is to look at how long the species has been around, how long some of our ancestral Homo species were around, and then the annual risk of extinction. Interesting, but I wonder how hard it would be to measure/model progress against this metric or the potential impact of any one action. Even if we don’t go extinct, the category of existential threats includes an unrecoverable collapse of civilization or merely a partial collapse to an “unrecoverable dystopia”. Let’s try to avoid any of these.
  • Of course, the UN has to balance all the doom and gloom stuff with an equal word salad of things we could try to do to make it better. There are a lot more facts, figures, and scientific references in the first part, and a lot more anecdotes and case studies in the second part. TLDR, but hopefully there is some stuff in there that could trickle down to actual policies and actions at the national and local level.

sperm counts and clean chemistry

Yes, according to one study from 2017, sperm counts are crashing and if you just extrapolate out in time it leads to disaster for the human race. Now, I had heard that sperm counts have dropped steadily over the decades, and we still have plenty of sperm for now, but we don’t know where the trend is headed next. I know about the concerns with endocrine disrupters. We also know that fertility is down for a variety of reasons, beginning with women around the world having more choices in terms of education and career.

This comprehensive meta-regression analysis reports a significant decline in sperm counts (as measured by SC and TSC) between 1973 and 2011, driven by a 50-60% decline among men unselected by fertility from North America, Europe, Australia and New Zealand. Because of the significant public health implications of these results, research on the causes of this continuing decline is urgently needed.

Temporal trends in sperm count: a systematic review and meta-regression analysis

I assume we are probably headed for a world of more technologically-assisted reproduction for a variety of reasons, beginning with just wanting to have more control over our fertility at various stages of our lives. But endocrine disrupters are potentially bad news for humans and for ecosystems. We don’t really try very hard to look for safer and equally functional chemicals before we put lots of them in the environment and in our bodies. I believe better living through chemistry does make our lives better on balance. For example, get rid of food preservatives or water disinfectants and we would instantly cause massive amounts of suffering and death. But get rid of most of the weird stuff in my shampoo, and I would still be able to wash my hair just fine. Building materials are a tough one. The tar on my roof and siding on my house are highly functional and beneficial, but they both cause pollution in production, manufacturing, and most likely cause water and air pollution. We can say similar things for materials used to build our streets and highways. We should look for clean but equally functional substitutes for all of these. And in the meantime, we should probably impose taxes to offset the impact these materials cause. This could both fund research into alternatives and provide some incentive to adopt alternatives as they become available.

advertising shits in your head

Some books just have good titles. And this one is called Advertising Shits in Your Head: Strategies for Resistance. From the publisher:

Advertising Shits in Your Head calls adverts what they are—a powerful means of control through manipulation—and highlights how people across the world are fighting back. It diagnoses the problem and offers practical tips for a DIY remedy. Faced with an ad-saturated world, activists are fighting back, equipped with stencils, printers, high-visibility vests, and utility tools. Their aim is to subvert the adverts that control us.

“PM Press is an independent, radical publisher of books and media to educate, entertain, and inspire.”

Really, they had me at the title. But it seems to me that the main way to counter the shit is to teach children from a young age to evaluate the source and quality of information they are taking in, expose themselves to multiple sources of information, think critically, draw their own conclusions, discuss their conclusions with other informed people, and be open minded.

Somewhat related is this new browser plug-in Newsguard, which provides “Trust ratings for all the news sites that account for 95% of engagement” and is “Written by journalists, not secret algorithms.” Sounds okay, although I think I have too many browser plug-ins already.

attribution science, and some thoughts on computer modeling

This Slate article explains how attribution science works. It depends on modeling. Basically, scientists model an event (like a storm, flood, fire, whatever) using a hypothetical condition where the event did not occur, and compare that to the data from our actual universe where it did occur.

I do a fair amount of modeling in my job, and there are always skeptics (some more informed than others). Why would anyone trust a computer model? Isn’t empirical measurement always better? Well, we model things we can’t measure, often things that could or would have occurred if things were different, or things that might happen in the future. To trust a model, first, somewhat obviously, you need to say what the model is for, clearly. Second, you need to be confident that it is adequately representing the real-world processes underlying the system you are interested in. Whether this is true requires expert judgment, and the expert needs to really understand the system. If the expert is confident in this, and the expert knows what they are doing, the model has some usefulness even if there is no data. (Purely empirical models like regression equations don’t represent processes, and therefore have limited predictive value if conditions change significantly.) But we always want data. Third, the modeler will compare what the model predicts to some real data. The modeler needs to be aware that there is always uncertainty in how well measurements represent the real condition of the actual physical universe, and that this uncertainty will propagate through the model (the uninformed often think of this as “model error”.) If the prediction is reasonably accurate without tweaking, you may have a pretty good model. Often the modeler will do a little tweaking to improve the fit, but the more tweaking the more you are moving toward an empirical model with less predictive value. In a somewhat old-fashioned (according to me) but common approach in the engineering field, the modeler will set a portion of the data aside while doing the tweaking, then compare the tweaked model to the portion they set aside. I don’t usually do this, because there is never enough data. I tend to use it all, then check the model again when more data becomes available in the future.

Finally, we have a model that we are confident represents underlying processes, matches real-world measurements reasonably well, and is suitable for its stated purpose. We can use the model for that purpose, be clear about the known unknowns and unknown unknowns, and draw some conclusions that might be useful in the real world. We have some information that can inform decisions better than guesses alone could have, and that we couldn’t have learned from data alone.