Surfing close to Robert Horn's Information Mapping. Which is a useful construct, but it's a dangerous idea to think that content information types just sort of

Surfing close to Robert Horn's Information Mapping. Which is a useful construct, but it's a dangerous idea to think that content information types just sort of live . . out there, somewhere. Content typing will always be context - dependent, which, well, can boil down to "know your audience". But the content types aren't General Truth, they do need audience to be defined, which is where I disagree with the DITA folks.