Category Archives: Online Tools / Apps / Data Sources

a free alternative to Adobe Photoshop

Leave a reply

Adobe Photoshop is what professional graphic artists have used for quite a while now. But for the rest of us, who might want to just make minor edits to a photo or change a background color in an image, it has been a major barrier. Now, Canva has decided to make their editing software Affinity free. Canva is a subscription service so they must have decided making this tool free will ultimately increase paying customers. I’m not a big Adobe fan (Die Adobe Acrobat! Die!) so I am not upset. It could also be that with even free AI tools getting pretty good at working with graphics, expensive tools like Photoshop are headed for obsolescense.

sea level rise projections

Leave a reply

A new data set aggregates sea level rise projections, of which there have been many.

Evolution of Global Sea-Level Rise Projections and Their Uncertainty

Recognizing this challenge, Garner et al. (2018)—hereafter, G18—developed a database of 21st century GMSL projections (A. J. Garner et al., 2018). An analysis of the G18 database generated several findings, including that the range of 21st century GMSL projections has varied greatly over time, despite central estimates being relatively consistent (A. J. Garner et al., 2018), and that upper estimates of GMSL rise from individual studies often exceeded upper bounds provided in Intergovernmental Panel on Climate Change (IPCC) assessments (A. J. Garner et al., 2018).

Eight years have elapsed since the publication of the G18 database and analysis, during which two additional IPCC reports incorporating new GMSL projections have been published (Fox-Kemper et al., 2021; Oppenheimer et al., 2019). To accommodate these new reports and other scientific developments and individual studies (e.g., Bamber et al., 2019; Brown et al., 2021; Grandey et al., 2024; Hamlington et al., 2024; Horton et al., 2020; Palmer et al., 2024; Weeks et al., 2025), we have updated and expanded the original G18 database of 21st century GMSL projections. The updated database (hereafter, G26 database) includes GMSL projections published through 31 December 2025 for a wide variety of time horizons—including decadal projections with end years throughout the 21st century, as well as longer term, multi-century projections. Other updates to the database include the addition of new fields, such as the country of origin for each GMSL projection study. These updates allow the assessment of the ways in which sea level science has evolved in recent years, but also to perform new analyses that offer additional insights into the history of the science, providing vital context for researchers and decision makers as they contend with challenges presented by the numerous studies and methodological choices in a deeply uncertain field.

the rise of electoral democracy

Leave a reply

Our World in Data shows is that, while there has indeed been a rollback in the number of democracies and number of people living in democracies over the last twenty years, the trend over the last 250 years or so have been a sharp and relentless uptick in electoral democracies. The US and UK were the first in the late 1700s, but New Zealand was the first to provide true universal suffrage.

I’ve been thinking, there are plenty of examples of rational and reasonably well functioning non-democracies in the world. And there are plenty of messy, imperfect, dysfunctional electoral democracies for sure. The article points out that a lot of electoral democracies exist on paper but have not had truly competitive elections in recent history. But we know the long-term relationship between economic growth and quality of life tracks closely with the rise of democracy, and we know that electoral democracy gives us some insurance against being trapped permanently in the truly terrible governance situations that sometimes arise in both democracies and non-democracies.

the current carbon price

According to academic meta-analyses, the carbon price should be in the range of $100-165 per (metric) tonne of carbon dioxide (equivalents, including other greenhouse gases adjusted for their impacts) emitted. Our World in Data found that about 30% of the world’s emissions are priced, which is up significantly but obviously still low. Uruguay, Sweden, Norway, Finland, Switzerland, and Hungary have charges in this range. The typical price across Western Europe is $65-90. Many countries that have a carbon price charge $10 or less. Still, I imagine getting a legal and administrative system in place is a big hurdle, and once you have one the price can be adjusted over time.

a deep dive on electric vehicle sales

The International Energy Agency has a deep dive on global electric vehicle sales. Sales everywhere have accelerated sharply over the last five years. Of the countries covered, only Norway has gotten close to 100% of all new sales. Other countries that are over 50% are Nepal, Iceland, China, and Singapore. Sales are particularly lagging in the U.S. and Canada. The IEA article has a visualization, but I also like one from Rest of the World based on the IEA data. (I was not able to embed either of them.)

Government subsidies are part of the reason for these trends, but the economics is also starting to favor electric vehicles in many places, even just looking at purchase price which is what consumers are largely going to respond to. But the availability of mass-produced low-cost Chinese models is also a big part of the story, particularly in Southeast Asia and in China itself.

I have a few thoughts. First, it’s not a story of subsidies skewing the benefit-cost picture towards electric vehicles when it would not otherwise be. The positive benefit-cost ratio is there, and government policy is helping many countries achieve commercialization and scale faster than they otherwise would, although the shift is inevitable even in the U.S. Second, U.S. subsidies TOWARDS liquid fossil fuels, along with lack of infrastructure investment, ARE skewing the positive benefit-cost ratio to negative for our already cash-strapped consumers, while propaganda hoodwinks us into thinking the rest of the world doesn’t have nice things being denied to us. Third, with the economics clearly favoring electric vehicles over the life cycle, corporate and state/local government fleets will be able to do the math and make the right choice, and this will eventually start to move the market. The federal government will follow suit eventually when political winds shift. And fourth, even if the benefit-cost ratio were not positive, the insane Iran invasion would be incentivizing China and Southeast Asia to shift toward electric and away from liquid fossil fuels from the Middle East, to help manage geopolitical risk. So in this sense U.S. policy may have moved the needle significantly towards electric for 2 billion or so consumers abroad, even while our 350 odd million at home are being successfully hoodwinked. Some of the energy to fuel these vehicles will shift from oil to coal in the near to medium term, which is not ideal, but it still represents an increase in efficiency (I think) and the economics will continue to shift toward renewables (and nuclear?) over time.

By the way, I had to refresh for myself: The IEA is an OECD institution, while the IAEA is a UN institution. Neither should be confused with the EIA, which is a U.S. government agency. My fingers really want to type IAE for some reason, and apparently I am not alone, but that is not a real agency.

what people die from

Our World in Data has updated stats on causes of death worldwide. The focus of the written article and analysis is children, but they have the stats for all ages. This is certainly a grim topic, but it is also very informative in terms of what policies a government might want to support if it actually cared about its people.

The visualizations in the article are great and you should look at them. I’m going to do some very visually uninteresting “top 3” ranking lists below.

Let’s look at a developing country. Our World in Data picks on Nigeria so I’ll do the same.

  • Under 5: malaria, respiratory infections, suffocation and trauma
  • 5-14: infectious diseases, malaria, respiratory infections
  • 15-49: HIV, AIDS, cancer
  • 50-69: heart disease, cancer, malaria
  • 70+: heart disease, cancer, diabetes and kidney diseases

So you can see why international development and charitable organizations focus on bringing health care up to a basic standard and trying to make sure people have access to it. And you can see how a developed country reducing aid spending literally kills people.

Now, the US:

  • Under 5: preterm birth, “other noncommunicable diseases”, suffocation and trauma
  • 5-14: transport injuries, cancer, “other noncommunicable diseases”
  • 15-49: other noncommunicable diseases
  • 50-69: cancer, heart disease, other noncommunicable diseases
  • 70+: heart disease, cancer, neurological disease

The transport injuries (aka “car accidents”) stand out to me as particularly shameful because they are mostly preventable. They’re caused by our poor land use and design choices, and selling out to the greedy oil-auto-highway lobby. Cancer, brain diseases, and the various “noncommunicable diseases” strike me as causes we have less control over. Better access to affordable health care can only help, but at some point progress on these issues is a question of basic science and technological progress (um, maybe governments should not be cutting these investments either? and there is more to “technology” than just AI, and and there is more to even medical technology than a handful of drugs wealthy people are willing to pay a lot of money for). What we really have some direct control over as individuals, then, is living somewhere we don’t have to be in a car all the time, and taking care of our hearts. You can also break this down by gender – men start having heart attacks younger (i.e., heart disease is the leading cause of death for men 50-69, while cancer is the leading cause for women), but if you are lucky enough to make it to 70+ heart disease becomes the leading killer for everyone. This is the low-tech side of health care, at least for the time being – good old weight, blood pressure, and cholesterol management. Or if we just want to continue being morbid, your goal is to die of cancer or brain disease in old age, because that means you successfully avoided heart disease and fatal accidents throughout your life.

I would have guessed that accidents like drowning and falling, and mostly gun-related violence including homicide/suicide/accidents, would be higher on the list. These are all important issues, but they just pale relative to the transportation-related violent death in this country. So in public policy terms, I wouldn’t suggest giving up on gun crime and accidents, but I would divvy up attention and money somewhat in proportion to the scale and obvious solvability of the problems.

This article is not about absolute differences in mortality rates between countries, but those are probably elsewhere on the site. You could somewhat obviously look at countries where car accident and heart attack rates are much lower, and then ask what policy or behavioral factors are contributing to those differences that you might want to try to emulate.

Normally you can embed these charts, but I can’t seem to figure out how to do that today.

my summer 2026 experiment with financial and economic literacy education for a young audience

This is a Google Slides experiment. I’ve come up with a summer financial and economic literacy thing for kids. It’s easy to find basic financial literacy content, but my idea here is to link those ideas to concepts like time value of money, endogenous growth, and ecological economics in a way that is accessible to an elementary to middle school audience. And honestly, an intelligent but unaware adult audience, because I don’t think most adults have a conceptual understanding of these concepts. It is a test of my own conceptual understanding if I can explain them clearly to children. This has been a challenge, and I am certainly happy to hear feedback on my attempt.

McKinsey on high-growth industries

McKinsey has a post with a data visualization on industries it predicted in 2022 would be growing quickly by now (May 2026 as I write), versus how they actually turned out. I find it interesting both for the industries/technologies themselves and for which are overperforming and underperforming. Overperforming ones include, of course, “AI software and services” and semiconductors. Robotics, however, has not kept up with expectations at least in terms of widespread commercialization (I think it is still coming, just behind schedule). Electric vehicles are also both high-growth and overperforming, while “shared autonomous vehicles” are high growth and were not considered in the original study due to “negligible baseline revenue” – more evidence that in the U.S. we are being duped as this combination of technologies explodes globally. Interestingly, batteries have not kept up with expectations as a high-growth, profitable industry/investment even though we know the technology itself has seen massive improvements in cost-efficiency. Biotechnology is a mixed bag – “obesity drugs” have exploded while “non-medical biotechnology” has seen no growth in the profitable investment sense. The holy grail of turbocharging construction productivity by making it more like manufacturing (“modular construction”) is about 50 years behind schedule. Maybe the robots can help with this eventually. And finally, even with all our fossil fuel woes the nuclear energy industry never seems able to capitalize, probably because of its long lead times and public risk-aversion on this particular technology.

My big picture analysis – technological progress is slow and steady, but when it comes to which will “hit” in a widespread profitable commercialization/investment sense, it is hard to identify the needles in the haystack at least in any sense of precise timing. In a personal investing sense, you can either gamble and go for broke, or you can diversify and be patient. In a broader economic sense, governments can use policy to try to give a particular industry a nudge, but there is a gambling aspect to this too, and my view is they would be better off focusing on reducing economic friction (great infrastructure, ease of starting a business, predictability, level playing field in terms of taxes and regulation) while protecting the environment and workers. Maybe provide childcare, health care, and education so people can start a business without worrying about those things, and have healthy skilled workers available when they do.

https://www.mckinsey.com/mgi/our-research/The-race-takes-off-in-the-next-big-arenas-of-competition

exploratory data analysis

There probably is no “one stop shop” for exploratory data analysis. I’d like to see some plots myself – box plots, cumulative distributions, time series – and why not put all those side by side with the relevant summary statistics? But still, this is a nice entry.

The describe() function from R’s psych package (Revelle, 2023) provides a comprehensive statistical summary of your dataset. Unlike R’s base summary() function, it includes additional metrics that are particularly useful for data exploration and assumption checking.

All this works best for data points you assume are independent of each other in time and space, which is not how the actual universe tends to work. And heah, I know there is a thing called machine learning, but I like to start simple to begin building a picture of a data set in my simple human brain.

bibliometrix

Bibliometrix is an R package for literature review and synthesis of past research on a topic. It now has a Shiny graphical interface.

bibliometrix: An R-tool for comprehensive science mapping analysis

The use of bibliometrics is gradually extending to all disciplines. It is particularly suitable for science mapping at a time when the emphasis on empirical contributions is producing voluminous, fragmented, and controversial research streams. Science mapping is complex and unwieldly because it is multi-step and frequently requires numerous and diverse software tools, which are not all necessarily freeware. Although automated workflows that integrate these software tools into an organized data flow are emerging, in this paper we propose a unique open-source tool, designed by the authors, called bibliometrix, for performing comprehensive science mapping analysis. bibliometrix supports a recommended workflow to perform bibliometric analyses. As it is programmed in R, the proposed tool is flexible and can be rapidly upgraded and integrated with other statistical R-packages. It is therefore useful in a constantly changing science such as bibliometrics.