No, you got this right,… spot on!
Detecting “Umlaute” and other characters and converting them into Stupid-Machine-Can-Read-Code
is known for ages and should not represent any problems for today’s engineers.
The other way round is not necessary…
As to the great danger of bugs and problems in softwares…
Such is life… A tale of misery, hunger & despaire …and buggy softwares…
If it doesn’t kill you, it’ll make you even stronger.
Big K ( to strong in the middle, already…lol…)