I am not a real historian, but until real historians turn their attention to obscure corners of the computer gaming world, someone has to discover and preserve knowledge about early roguelikes. This means not just talking about past roguelike events, but listing my sources and explaining my reasoning so that others can be convinced of my results or correct them.
Recently, akeley, creator of the Roguelike Archive, contacted me about a version of UltraRogue that uses the number "2.0". It's called "Ultra Rogue 1.03 (DOS)" on that page, but the file name is "ur20.zip", and it calls itself 2 in its version message.
UltraRogue was developed mainly by a programmer named Herb Chong. It was a fork of Advanced Rogue, first released in 1985, with further releases continuing until 1995. This can be deduced from the license file. A few years later, the Roguelike Restoration Project received the UltraRogue source code and developed version 1.07. The RRP SVN repository contained two earlier source trees, numbered 1.02 and 1.03. How does 2.0 fit into this numbering?
If you look at the changelog, you will see the number 2.01 at the beginning. UltraRogue originally used numbers inherited from Advanced Rogue, and started its own versioning scheme later. On January 15, 1985, the version changes to "0.00 Alpha", with the entry "version and name of game changed to UltraRogue. hopefully, it will live up to it's name".
Advanced Rogue 5.8 states that it was released in January 1985. UltraRogue must have been forked from an earlier version, as its history file begins the previous month. The Roguelike Restoration Project found a copy of Advanced Rogue 1.0. Unfortunately, neither that game in any form nor any further information was ever posted. It appears that between Advanced Rogue 1.0 and 5.8 came a 2.0 which was the ancestor of UltraRogue. But we have no artifacts or evidence of release dates to confirm its existence.
Does Advanced Rogue 2.0 have anything to do with the UltraRogue 2.0 for DOS found on the archive site? Probably not. Remember, Herb Chong never called anything "UltraRogue 2.0": he changed the version numbering at the same time he invented the name. That was before he had made a first release. Would the game have been ported to DOS before it was even finished?
I'm inclined to trust the text file bundled with the DOS version, which says it is a port of UltraRogue 1.03. This does not explain why someone applied the number "2.0" to it. But sometimes people change version numbers in unintuitive ways. Large software companies have been known to pick numbers for marketing reasons. Maybe the creator of the DOS port made a few changes and thought they deserved a new major version.
Without the source code to this DOS version, it's hard to confirm where it belongs in the UltraRogue development line. Someone could attempt to compare it with the various surviving source trees, deducing its behavior by disassembling it or repeatedly playing it. That's a project for a real roguelike historian. Or a roguelike archeologist.