Dear Tom,
I feel like I'm not fully understanding your question or aim, but at the very least let me summarize my findings based on terms and resources mentioned today.
The RAND or FRAND mainly refers to the licensing policy associated with the standards. This especially makes sense in industries where basically everything is patented, which is just about every industry other than the software industry. Being up to date in making standards is part of the standardization process, [1] and therefore dealing with active patents is hard to avoid.
[1] https://de.wikipedia.org/wiki/Deutsches_Institut_f%C3% BCr_Normung#Grundprinzipien (German only)
That seems to be why standardization bodies have adopted policies for patents related to standards. [2] ISO offers two options, possibly free-of-charge, either way "on a non-discriminatory basis on reasonable terms and conditions.". [3] If a patent holder will not agree to these terms, even though the negotiations are handled outside the standardization organization, the standard will avoid the patent.
[2] http://www.iso.org/iso/standards_development/patents [3] http://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/Common_Guidel...
So regarding the discrimination of free software, the ability to make use of a liberal licensing policy of essential. I'm not sure if free-of-cost would suffice, as this might not cover modifications (freedom 2). Whether or not liberal licensing policies are preferred in the standardization-process I don't know.
That leaves the subject of Open Standards. The definition of Open Standards seems to be inherently incompatible with active patents. I'm curious whether or not the process adopted by the standardization bodies is sufficiently open for the Open Standards definition, as I haven't looked into the standardization process that much, and it might vary between organizations.
Does that answer your questions? If not I'd be glad to hear about them and delve into it.
Kind regards, Nico Rikken