A couple of problems that I see with the ISO-codes idea is that the
codes include h, q, w, and that many letter combinations will not be
valid in Lojban phonotactics. Converting h -> x, q -> k, w -> v or u
would probably create many duplicate codes, and I'm not sure how to
deal with the phonotactics issue.
Unfortunately, the phonotactics of Lojban, being so restricted, this is always going to be a problem, I think. A greater issue caused by 'c' (whose idea was that letter, anyway?!), as it could be, phonetically, /s, k, or tc/. My gut reaction is to go for the sound over the actual letter.
In terms of the ISO codes, the original idea (in my head, at least) would be to have the fu'ivla forms based on the ISO to as large as an extent as possible, but fully acknowledging that it might not always be.
Developing these isn't really all that hard. Here's a sample of what I came up with earlier, which took maybe 20-30 minutes. In all of the below, everything before the ' is derived as much as possible from the language code, the second part from the language family ISO code.
Language name in English (native name) fu'ivla (actual language ISO code/language family ISO code)
Afar (Afaraf) fra'afa (aar/afa)
Albanian (Shqip) ski'ine (sqi/ine)
Amharic (-amharinynya-) xma'afa (amh/afa)
Argonese (Argonés) gra'ine (arg/ine)
Assamese (অসমীয়া) sma'ine (asm/ine)
Belarusian (Беларуская) ble'ine (bel/ine)
Breton (brezhoneg) bre'ine (bre/ine)
Chechen (нохчийн мотт) tce'acu (che/cau)
etc.
Chris