That's…interesting, especially seeing as the NMRA is/will be/is thinking about adopting RailCom as a standard…
I believe the current draft of the NMRA's Recommended Practice RP-9.3.1 & 9.3.2 is essentially Lenz's RailCom, however 9.3.2 has been in "draft" status since the current version was issued in 2005. They don't seem to be in a hurry to adopt it, and once adopted it will be an RP, not a standard. As an RP it doesn't preclude use of other methods (like Digitrax's proprietary Transponding) as long as they don't interfere with aspects of DCC that are standardized, but it also doesn't require a manufacturer to support it.
All of which means it can still change, and there's no guarantee of interoperability in the future even if it doesn't. Which really sucks.
That's no reason not to use RailCom (or any other 9.3.1/9.3.2 compliant decoder), but it is something to be aware of.