Tale is the Afrikaans word for languages, and in English is an imaginatively recounted story. Due to my naiveté in both spoken/written languages and prior work in this area programmatically; I'm just going to start with the simplest methods I can think of.

I only know English, so the translation is via the internet.
(Checked some of them with google translate.)

Currently, the demo menu builds in 12 languages:

German
English
Finnish
French
Indonesian
Italian
Japanese
Korean
Polish
Portuguese
Russian
Chinese

Obviously, other languages or refinements are welcome.

GPT-4 suggested that I add some additional languages (purely based on population metrics):

Bengali
Hindi

https://github.com/bitRAKE/fasmg-umbrella/tree/main/examples/in-tale

Next, I will add some right-to-left languages.

FYI: GPT-4 was able to grok the file and only replace the text in quotes with translations. There was no need to change the produced text. This included changing the language constant symbol at the start of the file, and not changing the quoted text referencing key-combinations. Fine tuning consisted of advising that the ampersands probably no longer coincided with valid letters (which I could be incorrect about, but GPT-4 did remove them).

The above technique works well because it's the only resource in the executable. Windows isn't given another option - it doesn't matter what the user/system settings are.

Maybe you work in a more complex environment and want to respect Windows settings. This can be done by having multiple resources with the same ID, and windows will select those closely matching the user/system's preferences.

I would say this is the next easiest method to support multiple language. The downside is memory bloat - all the resources in one executable. Without changing the system/user settings, alternate language preferences can be tested with: ... at the start of program.