Author Archives: rdmyers75@hotmail.com

Is AI speeding up computer programming efficiency?

Yes, by about 25% according to a serious look at the hard evidence by some heavy-weight academics (MIT, etc.)

The Effects of Generative AI on High-Skilled Work: Evidence from Three Field Experiments with Software Developers

This study evaluates the impact of generative AI on software developer productivity via randomized controlled trials at Microsoft, Accenture, and an anonymous Fortune 100 company. These field experiments, run by the companies as part of their ordinary course of business, provided a random subset of developers with access to an AI-based coding assistant suggesting intelligent code completions. Though each experiment is noisy, when data is combined across three experiments and 4,867 developers, our analysis reveals a 26.08% increase (SE: 10.3%) in completed tasks among developers using the AI tool. Notably, less experienced developers had higher adoption rates and greater productivity gains.

“Intelligent code completions” kind of matches my own experience with how I have found AI most helpful so far – as software help. Whether it is helping with obscure code syntax or complicated nests of drop-down menus and check boxes, AI makes it much faster to find the exact thing you are looking for. This should in theory give workers a bit more time for planning and creative thinking, but predictably the market wants us not to do our jobs better, but to do them barely adequately as fast as possible. And what passes for “barely adequately” erodes over time while “as fast as possible” gets faster. Which I suppose is efficiency on paper.

One question is whether this is more like the automated loom, which sharply reduced the demand for textile workers, or the cotton gin, which sharply increased the demand for (involuntary, brutalized) workers by removing a bottleneck in the process. Early signs seem to point to the former, but all this will take time to play out.

Biden and the “shovel ready” problem

This Politico article from May 2024 does a good job summarizing Biden’s legislative achievements and then gets into a central problem that seems to be facing our country in recent decades – implementation is very, very hard. And because it is very hard, politicians who promise they can deliver substantive, tangible results often have trouble demonstrating clearly that they have delivered what they promised.

Now, I think Biden was a great president with substantial accomplishments, for about three years. I think his legacy, unfortunately, is likely to be determined by that last year. He should have announced in early 2023 that he was going to retire gracefully at the end of his first term, and allowed a full Democratic Party primary to play out. If that had happened, maybe we would be exactly where we are today, after a Kamala Harris nomination and close loss. And maybe we wouldn’t – maybe she would have nominated and run a more organized, successful campaign that reached an extra 2% of voters. Or maybe a more dynamic leader would have emerged. Anyway, that is not what happened here in this particular universe which is real as far as we can tell. So let’s talk Biden.

Biden had four major legislative accomplishments. The dollar figures below include what was appropriated (approved/required to be spent) by Congress plus tax breaks:

  1. the 2021 pandemic relief package (“American Rescue Plan”) – $45 billion
  2. the 2021 bipartisan infrastructure investment law (aka the “no catchy name” act?) – about $840 billion
  3. the 2022 “CHIPS and Science Act” – maybe $60-70 billion? (Politico has done a decent job laying out the numbers in graphics and text but it is still not perfectly clear)
  4. the 2022 climate and energy-focused “Inflation Reduction Act” – about $500 billion, largely through tax breaks?

The Politico article paints a picture of Biden being frustrated after he expected to spend the last couple years of his term at ribbon cuttings taking full political credit for accomplishments produced by these bills. I think a few things have happened here.

First, implementation is slow. Realistically, investing a trillion dollars productively is going to take time, and it is probably good for it to take time. Good investments require planning, and planning takes time. Well planned, slow and steady investments in infrastructure, research and development, and manufacturing capacity seem like a great idea to me in the real economic world. In my rationally planned infrastructure fantasy world, well-thought-out, frequently updated comprehensive plans would exist at the metropolitan area scale with construction projects queued up for bid as soon as funding can be found. The real world is not like this, but an idealistic vision can provide a direction to steer our ship. Funding would come more from the private sector when unemployment is low, and more from the government when unemployment ticks up and private credit is tight. There also has to be money and a plan for operation and maintenance of whatever is built, which is also politically unsexy. All of this could be legislated, but it would have to be done in advance as an automated rule, rather than requiring Congress to react in real time to the business cycle, which it can’t do.

Second, implementation is hard, and it seems to be harder in our country than it needs to be. There is a lot of debate on the reasons, but it is some combination of labor cost/scarcity, capacity/competence of domestic firms and workers, lack of competition, slow productivity growth in certain industries (particularly construction), and corruption. There are policy options to address all of these, but either they are politically inconvenient (like more visas for guest workers or allowing foreign firms greater access to our markets) or our politicians don’t understand them.

Third, at least some slow and steady implementation definitely happened, and Biden had trouble taking political credit for it. A November 2024 NPR article talked about this. The ribbon cutting press release strategy just didn’t get much media attention at a time when the public was more focused on disappearing household disposable incomes. Arguably, maybe, the administration wasn’t savvy enough with modern communication styles and tools to get the public’s fragmented attention. Or more ominously, maybe the public’s attention is so fragmented it can’t be gotten with any kind of rational, positive message. I am also thinking back to the 2021 stimulus, when my household disposable income definitely increased due to the stimulus package. This was done so quietly and invisibly as a tax credit directly to my bank account (which I had used to pay my taxes electronically), I barely noticed it. This was economic brilliance at a time when people really needed the help, and it may have saved our country. Politically, maybe the Democrats should have had party operatives knocking on my door and handing me a check. Or maybe they should have done that with some randomly chosen fraction of people and made sure they had the media in tow.

A final thought – since implementation is hard and slow, a lot of Trump’s agenda is trying to throw up obstacles to implementation of Biden’s trillion dollars. He will probably thwart some but not all of it. So Biden’s positive legacy will continue to play out.

the Programme for International Student Assessment (PISA) test

Results of this international comparative test show worldwide drops over the past 20 years, with accelerations since the pandemic. We should note the scale of the graphic, yet the trend is clear. Poverty, distracting devices, and mental health are offered as potential explanations. East Asian countries and city-states do best in math, and the United States sits a bit below the average. Our close cultural cousins the UK and Canada do notably well, while Australia sits just a hair below the average. It’s interesting that the worst performing students in the US seem to do better than the worst performing students elsewhere. Could this be because of the things we actually do right, like get kids to (a) school regardless of income and give them some calories while they are there?

Abundance

I suppose I need to take on the new book “Abundance” at some point. Perhaps I should read the book first? Well, I doubt most people talking about it have read the book. I’ve read at least half a dozen review of it, and not one of them was able to summarize it in a simple sentence or two that I am able to remember. And this would seem to be a problem politically. I personally had an impression of it as being about technological progress, because I remembered reading a 2012 book by Peter Diamandis called Abundance: The Future is Better than you Think. It is not about that. Then I thought it must be about inequality, because the U.S. is a rich country with a big and growing inequality problem, and that is why the cast masses of people do not have abundance. But it is definitely not about that, in fact it argues that the Democratic Party should mostly not be talking about inequality.

Okay, so without reading the book (yet), it seems to go back to this 2022 article in the Atlantic by Derek Thompson called A Simple Plan to Solve All of America’s Problems. If it were truly simple, again, I should be able to summarize it in a sentence or two, but I can’t. But here goes in a few sentences:

  • Unnecessary complicated Federal bureaucracy slows down or stops implementation of many things we like, such as Covid tests (dated example), issuance of visas for skilled foreigners like nurses and teachers (hoo boy, dated example).
  • The public and private sectors together have failed to invest enough to keep up with critical technologies like semiconductor manufacturing and automated port operations (not mentioned here, but in the news a lot lately, is ship building).
  • We’re not solving our massive market failures in housing (local zoning laws are cited) and health care. In the case of the latter, the author cites the government and medical industry artificially limiting the supply of licensed doctors and nurses.
  • The clean energy rollout has been somewhat of a bust, or at least very slow.
  • He talks about colleges, but only cites the fact that “elite colleges” only enroll a small fraction of the nation’s students.
  • Infrastructure…er, he only talks about transportation, as the majority of discussions on infrastructure do. But yes, it is hard, slow, and expensive to implement.

And…I’m out of time, but I’d like to come back to each of these at some point. Each one has a complicated, messy set of origins and potential solutions. I am having trouble seeing a sound bite version of these solutions. But the idea of “Abundance” seems to be that if we solve these problems, we get abundance, so they are worth solving.

“how genetics is changing our understanding of ‘race'”

This is a New York Times article by David Reich in 2018. This can be a taboo subject of course, but I think it is useful to know some facts on what the key studies have been and how serious scientists think about it. As Reich points out, because it is a taboo subject, serious scientists self-censor to an extent and this creates a vacuum where un-serious and ill-intentioned people step in. So here are some facts:

  • A key genetic study was done in 1972 by Richard Lewontin. He concluded that about 85% of human genetic variation is explained by differences in individuals and 15% by ancestral, aka racial, categories (which he created, resulting in a slightly circular logic). The categories he chose were “West Eurasians, Africans, East Asians, South Asians, Native Americans, Oceanians and Australians”. Without digging into the paper, I imagine he tinkered with these categories to make the proportion of variation explained by the categories as large as possible, and this is what he came up with.
  • That study become the basis of a broad consensus that the term “race” has no real biological meaning, and is therefore a “social construct”.
  • Reich goes on to argue that even though race is a social construct, it is useful because the race that a person self-identifies as is correlated to certain genes, which in turn are predictive of the risk of certain diseases. So, it makes complete sense for doctors to use a person’s self-identified race as part of health screening. [At least until we just all get our genome sequenced and stored in a medical records?]
  • Reich then goes into the taboos against, and some studies that have dared nonetheless, the delve into correlations between genes, behavior and “cognition”. He doesn’t use the term “intelligence” by itself, but rather “performance on intelligence tests”. [To me though, the examples he gives all seem very marginal, such as a study of people in Iceland showing that certain genes are correlated with years of educational attainment. How well can we truly control for all the factors other than genetics that affect this?]
  • Reich points to an interesting study of the ancestry of modern western Europeans (aka “white people”). They (we) are a mix of ancient middle eastern farmers, western European hunter-gatherers (sometimes called “barbarians”?), and people of Asian ancestry from the Siberian steppes. One interesting thing is that those people from the Eurasian steppes have some genetic similarities to Native Americans. So if a white North American has their DNA sequenced and finds some Native American ancestry, that could have happened in North America in the last 500 years or so, or in Europe a lot longer ago.

I’m not sure I have great words of wisdom to end this one with. Continuing to study the genetic basis of disease seems like a good idea. Trying to link “race” to “intelligence” seems like a waste since neither of these concepts is clearly defined, and even if they ever are, most peoples’ failure to live up to their innate potential is going to be due to factors other than genetics. “Highly intelligent” people who can beat me easily at checkers are not much use to society if they fall for obvious lies and logical fallacies coming from politicians and advertisers. In fact, they are a danger to society. So we need to focus on removing barriers that prevent people from living up to their potential.

electrification in China, US, EU

Here’s an interesting stat from OilPrice.com:

According to a study, China’s electrification rate has hit 30%, significantly ahead of the U.S. and the EU and US where the electrification rate has plateaued at ~22% in recent years.

The study defines the electrification rate as the share of electricity in final energy consumption versus energy coming from fossil fuels. According to the study, the U.S. still leads the world in the electrification of buildings; however, China recently caught up to the U.S. and Europe in industrial electrification, and has overtaken both in the electrification of transport. In 2024, electric vehicles (EVs) made up approximately 47.9% of the total passenger car sales in China, a huge increase from 2020, when plug-in EVs accounted for just 6.3% of total sales. In comparison, electric vehicles accounted for less than 23% of new car sales in Europe over the timeframe.

I’m all in on electrification. For one thing, it reduces air pollution and carbon emissions even with our current energy supply mix, as I understand it. But it also allows us to substitute cleaner fuels for electric generation over time, starting with natural gas for coal and oil, and moving towards nuclear, renewables, and as an aspirational goal, maybe even fusion.

I’m not surprised the US is lagging on electrifying transportation, because the oil, auto, and highway lobbies are politically powerful and have money at stake. The regulated electric utility and nuclear industries don’t have the same political pull. (There is no particular reason oil couldn’t have been a regulated public utility.) It surprises me a little that the US and Europe are at the same level.

offloading thinking to AI

It’s disturbing if professionals and students are trying to use AI to avoid hard thinking, as this duo of articles suggests. Ideally, at least in the near to medium term, we need to be doing the opposite. Using AI to perform mundane, repetitive, or just plain frustrating tasks that take up a lot of our time but don’t require deep thinking. Figure out coding syntax is an example, or which of the 99 drop down windows and dialog boxes in Microsoft Word will fix the frustrating formatting problem. (Actually, these last two things are kind of the same as you think about it, just two different ways of accessing a complicated menu of options and trying to communicate with a computer in its version of logic.) If AI can free us from these time wasters, we can have more time for deep thinking and creative thinking. I’m not saying this is the general trend, but this is my personal goal for how I am using AI. For now, I want it to help me do something I could have done myself faster or better. Asking it to think for me would be like asking another person to eat, exercise, or poop for me – I won’t gain any benefits from that.

I’ve been trying to use CoPilot to help me debug a simple stock and flow model. It can’t. It gives me sophisticated-sounding answers that do not even come close to working in the software I am playing with (Vensim PLE in this case).

May 2025 in Review

Most frightening and/or depressing story: The India-Pakistan conflict seems to have died down a bit (or did the media outlets I pay attention to just lose interest?). But both the potential nuclear conflict and the long-term loss of glacial ice billions of people depend on are terrifying.

Most hopeful story: I came up with four keys to my personal happiness in the moment: sleep, coffee, exercise, and down time. What, no family, community, career accomplishment, or making a lasting difference in the world you ask? No, those are about reflecting on life satisfaction, not being in the moment. No “fun”? Well, my idea of fun may be different than your idea of fun. I wish you joy and happiness as you pursue your idea of fun, only try to have some empathy and don’t force your own idea of fun on others. So there.

Most interesting story, that was not particularly frightening or hopeful, or perhaps was a mixture of both: The U.S. approach to R&D is a partnership between government (through both grants and procurement power), universities, and the private sector (historically, including regulated monopolies like Bell Labs). Other countries including China have copied this model somewhat successfully, and our own government taking a monkey wrench to our own system that has worked so well seems like a really stupid idea. First we need to stop the damage and then let’s hope it can be repaired.

the ggplot2 “ecosystem”

In the beginning there was R. Or, S? I’ve heard that R actually rests on a foundation of C++ or Java. Anyway, then there was the tidyverse, sort of another whole programming language that rests in R (or a metastasizing cancer that has grown to dominate R, if you ask certain people, but I personally am a big fan). Now within the tidyverse was always ggplot2, which I have grown to rely on almost exclusively for plotting. Now ggplot2 itself has grown into an “ecosystem” of related programs and extensions. Here is a useful guide. I’ve always been interested in finding the really good ones for things like interactive charts (plotly) and animations (gganimate). And awesome as ggplot2 is, there are some things that are just clunky, like scales and legends (seriously, legends are a big pain point for me – I hope there is an extension out there that really streamlines legends). But I am also wary of using extensions that might be buggy or not updated/supported long term, which could make my code obsolete sooner. So I usually try to do things with ggplot2 proper first, and if that doesn’t work with a reasonable effort I will try one of the extensions. So this guide seems timely and useful.

updating the science on nuclear winter

Jeff Masters has a nice summary of the science on the global devastation of even a limited regional nuclear war. He starts with the accounts popularized by Carl Sagan and others in the 1980s, which really did move the needle on global public and political consciousness on the issue. There was also a 2008 paper about the global consequences of a relatively small India-Pakistan exchange. Since then the science has been updated several times, including using the latest climate models. The results are always bad, with even the limited regional war disrupting global agriculture for up to a decade and killing 2 billion people. It would just be cold, in the summer, where food is normally grown, for ten years. Human beings would starve on an unimaginable scale. By contrast, a huge volcanic eruption like the one in Indonesia in 1815 could cause a similar effect, but it would last only a few years. (Nonetheless, we should have a plan for that one, no? Something that happens every few hundred years is common and you have to have a plan for it!) This should probably be the #1 political issue no matter what else is going on. Where are the courageous leaders today?