Javac uses the platform encoding [0] by default to interpret Java source files. This means that Java source code files are inherently non-portable. When Java was first developed (and for a long time after), this was the default situation for any kind of plain text files. The escape sequence syntax allows to transform [1] Java source code into a portable (that is, ASCII-only) representation that is completely equivalent to the original, and also to convert it back to any platform encoding.
Source control clients could apply this automatically upon checkin/checkout, so that clients with different platform encodings can work together. Alternatively, IDEs could do this when saving/loading Java source files. That never quite caught on, and the general advice was to stick to ASCII, at least outside comments.
[0] Since JDK 18, the default encoding defaults to UTF-8. This probably also extends to javac, though I haven’t verified it.
Source control clients could apply this automatically upon checkin/checkout, so that clients with different platform encodings can work together. Alternatively, IDEs could do this when saving/loading Java source files. That never quite caught on, and the general advice was to stick to ASCII, at least outside comments.
[0] Since JDK 18, the default encoding defaults to UTF-8. This probably also extends to javac, though I haven’t verified it.
[1] https://docs.oracle.com/javase/8/docs/technotes/tools/window...