We in fact did discuss this again recently, but the challenge isn’t so much the code base, but the dependencies we use for various features. The activation system, PDF viewer, the code that takes imported web pages and encapsulates them into MHT, the conversion engine that allows for import and export of DOC, DOCX and ODT, and a few other things as I recall.
It would in short be potentially even more work than it would take to make a Linux version, given the replacements that would have to be researched and integrated. Not a lot but a fair amount when you consider that the Windows version isn’t even finished.