Last week I had two curious conversations -- curious insofar as my conversation partners used numbers in a way that I, as a physicist, would not.
The topic of the first conversation was the use of cash in public transport. It worries me that more and more public transport companies replace their ticket machines by systems that do not allow the customers to use cash. This practice excludes marginal groups from essential public infrastructure. The discussion went like this:

Me: "The difference in costs for running machines that take cash and machines that do not amounts only to a few percent of the total revenue from ticket sales."
My friend: "But there are hundreds of ticket machines. This is a lot of money."
Me: "The absolute value does not matter here. Of course, the total costs are larger in Berlin than in Freiburg, but there are more customers, too. This is about the price per ticket."
Another friend intervenes: "Tanja, I am a business woman. Believe me, at the end of the year, I only care about absolute numbers."

The second discussion took place in a doctor's practice. It concerned a different topic (the privacy of medical information), but the structure of the argument was very similar:

The receptionist: "We need your email address to let you know if the doctor thinks that another consultation is needed."
Me: "Do you encrypt emails?"
The receptionist: "No."
Me: "I prefer not to communicate about medical issues via unencrypted emails. Could you call me by phone?"
The receptionist: "We have hundreds of patients! How am I supposed to make that many phone calls?"

The topics differ, the issues are the same. Absolute numbers related to cost or working hours are used to enforce an argument, while the actual question (the inclusion of marginal groups, the privacy of medical information) is avoided.

To a physicist these arguments seem very odd. When we express a physical quantity in numbers, we always do so in relation to a reference quantity. The first question to ask about a physical problem is usually: "What are the relevant energy, length and time scales?" In other words, what am I comparing to? The only number that makes sense in an absolute way is zero.

When talking about money, this might not be obvious at first glance. However, the friend who runs a business, of course, is not interested in abolute numbers. She compares all costs to the business volume of her company, but probably she does so unconsciously.
To ensure that unconscious reference quantities do not affect one's reasoning, it is important to know the relevant scales.

Tanja, 14.10.26

Recently, my employer stopped sending salary statements by mail. Instead, the employees are expected to download their salary statements as pdf-files from a server. One argument brought forward in favour of this procedure was that it was environmentally friendly. This suprised me.

Sending a letter produces 20-30 gramm CO₂ (this estimate includes the production of the envelope and the transport in a lorry) [Source].

To estimate the carbon footprint of downloading a pdf from a server is harder, because it depends on answers to questions such as: How often do the users replace their hardware? How is the hardware recycled? How is the electricity produced that runs the server? If the user is prudent, one obtains an estimate of 1 gramm of CO₂. However, if once per month the user asks an AI assistant to find the website from which the salary statement can be downloaded, the comparison is much less convincing.

In any case, we are looking at very small contributions to the total CO₂ production. In Germany, one person produces on average 10 tons of CO₂ per year. I.e. a letter contributes 1/400000 to this (that is 0.00025%). This implies, if one would completely stop sending letters and replace the entire mail service (for letters) by a CO₂-neutral technique, one would save only about 0.1 percent of the total CO₂ production.

If each person reduced their coffee intake by one cup per month, the effect would be at least ten times as large as the effect made by abolishing salary statement letters.

In many restaurants you need to scan a QR code to read the menu. Public transport companies abolish paper tickets. And my baker's till displays a QR code instead of printing a receipt. The argument is always the same: replacing analogue by digital technology is considered environmentally friendly because it saves paper.

But, what is so bad about paper?

Paper is produced from organic, renewable raw materials (mostly wood).
Paper can be recycled easily.
For paper used in Europe, we can cover the complete supply chain in Europe (or in other countries that have similar standards concerning labour law and environment protection, e.g. Canada).
You can store information on paper without using any technology produced by American companies that do not bother about European data protection laws.
You can retrieve the information easily even after hundreds of years.

It is not at all obvious that printing information on paper is always inferior to storing it digitally. As is often the case, it depends on the details.