Just a note here: Open Office makes an attempt to read and write the proprietary Office file formats. If often ends up in real sorrow if trying to open a complex Word document and then save it again. Reverse-engineering has a lot of limitations. A quick view at Open Office may show it as a good alternative to Microsoft Office. But besides a lot of missing features, it is not a good idea to try and share more complex documents in a mixed environment.

Next thing is that the owner of a proprietary file format may decide to - or not to - request royalties or not. Or if they should try to take legal actions against competitors that makes use of (or tries to) the file format.

And they may decide if they are selling - or giving away under an NDA - the documentation. And they may decide if the use of the file format should require royalties or not.

Next thing is that, depending on country, it matters a lot if if is a file format used to store primary information owned by the creator of the document, or a file format storing secondary information. When it comes to a Word document, the document is an original. The author of the document owns the rights to the information represented by the document, which can have a big difference in what a country allows - or doesn't allow - when the user wants to get access to the information expressed by the document. That greatly affects Microsofts choices when it comes to requesting royalties for the use of the Word file format. But the classic *.doc file format is still proprietary. And it doesn't change Microsofts abilities to decide if they want to sell the documentation or not. Or if a reverse-engineering of existing documents will show all possible formatting combinations, or if there are other features already specified for the file format that will break any file reader created through decoding of existing documents.

The debugger output from a compiler is not primary information. The developer still have access to his source code in the clear-text C files. If the developer can't access all debug information from the Keil output files, the developer is always free to select a different compiler to feed his source code through. That can make a big difference in legal standings when it comes to reverse-engineering of the file format.

And different countries have different views on file formats regarding how to protect. Do they represent an algorithm that may be patented? Or do they represent something that should be covered by copyrights? Is it a device or a pattern?

The big thing here is that you have to separate "proprietary" from "royalties required" or "patent infringement" etc. The important thing is if it is an open file format, where the full specification is openly available, or if the full ownership of the file format is by a user or company and where you may have to buy the specification, or where you may only find third-party documentation from reverse-engineering. Protection of a proprietary file format is a secondary parameter.

Being proprietary is just a requirement for potentially trying to protect the documentation or use of the file format. Proprietary means that you must figure out if you will break any laws by reverse-engineering or use - and how your situation changes depending on where in the world you, or your customers, live.