Convert legacy Word .doc files to Markdown in your browser. Old OLE binary documents, no upload and no Word licence needed.
A .doc file is not a zip. It is an OLE compound file — a small filesystem inside a single file, with named streams for the text, the formatting tables and the embedded objects — and that container predates the whole modern XML lineage by more than a decade. Reading one means walking that directory structure and then reassembling paragraphs from a piece table that stitches together fragments of a text stream, so the work has almost nothing in common with parsing a .docx.
The obvious place for this to go wrong is text encoding. Word 97-2003 wrote runs in whatever code page the authoring machine used, so a document typed on a Russian or Japanese system contains no UTF-8 anywhere — the bytes only mean anything once you know which code page was declared inside the file. That declaration is honoured here: a .doc holding CP1251 Cyrillic comes out as Привет, мир! rather than as mojibake, and the same goes for Shift-JIS Japanese. You do not need to know what encoding your file used.
What you do lose, compared with converting the same document as .docx, is everything Word only implied. Bookmarks vanish, so a cross-reference that jumped to a named place in the document becomes ordinary text with nothing to click. Images disappear entirely rather than leaving alt text behind. Text boxes lose their contents, not just their position. If the .docx version of a document is still around, convert that instead; if this is all you have, the prose, the lists and the tables still come through.
Encrypted or password-protected .doc files are rejected by the engine. Two things that survive a .docx conversion do not survive this one: bookmark anchors and embedded image alt text. Both are simply absent from the output rather than flagged, so a document that leans on internal cross-references will read as though the links were never there.