Summary: o media content might contain version information, but reliance on that can result in wasted resources o negotiation via content-features may be a viable mechanism where negotiation is possible, but it is not always possible (and some means of indicating version must be available to the content negotiation mechanism) o proliferation of type/subtypes tags for the sole purpose of versioning is undesirable that leaves parameters associated with the media type as an efficient mechanism for indicating version information.
I would also add that how version information should be handled very much depends on how the versioning is done. Versioning schemes range from single integer affairs (V1,V2,V3) to major/minor/patch m.n.o schemes to feature-by-feature enumerations. Some formats (e.g., PostScript) even allow conditionalization of features that may not be supported, so documents work everywhere but work "better" (in some sense) when certain extensions are present. IMO the more complex a versioning scheme is the less amemable it is to being captured in a set of parameters. Once things pass a certain point content negotiation needs to be used, or if it cannot be used there's no choice but to rely on information embedded in the data. Ned