Media Types

, Feb 11, 2008

Interesting reactions to the media type proposal. First, Steve Vinoski likes it:

I like this approach for several reasons beyond just the decentralization win. First, this allows us to gain experience with new media types before they’re standardized. Second, this augments rather than replaces the current IANA registration approach, and doesn’t break the existing IANA registration process in any way. Third, it doesn’t break or otherwise negatively impact any existing applications.

So, enough emailing and blogging about it — who’s going to write it and submit it to the IANA? I hereby offer to pitch in and help in whatever way I can.

Mark Baker points out that this has already been attempted (there’s even an IETF draft authored by Mark Nottingham on it!). He doesn’t like the idea anymore, though — because of three problems: we need less, not more data formats; parameters are a bad choice because they’re not treated as semantically relevant; being forced to follow the IANA registration process is actually a good thing. IMO, problem #2 is easily fixable, while problems #1 and #3 are actually related.

Maybe the proposal is a problem that’s self-made? At least this seems what both Mark and Bill de hÓra seem to believe. Bill writes:

Sorry, but ‘applicaiton/data-format’ seems to be the same approach to representations as overloaded POST is for protocol actions. It conflates the least meaningful signal with the most general one. It’s no surprise that a few former WS-* fans are mooting this. If you want to indicate meaning, use intrinsically meaningful interlingua, not gensyms or links to gensyms. For The Web (FTW!), RDF/OWL is the best game in town as far as interlingua go.

It’s definitely logical to view things this way — if one buys into the uniform interface argument (I do), it’s hard to argue against a uniform data format. As Mark writes,

[A]n abundance of media types is a bad thing for pretty much the same reasons that an abundance of application interfaces is a bad thing; the more that is different, the more difficult interoperability becomes.

I’m not sure yet I believe this. Interoperability isn’t furthered by having meaningless MIME types such as application/xml, or un-interoperable MIME types invented ad hoc. I find it hard to accept that the only way to achieve interoperability should be by either using something like RDF or defining a full formal standard … Admittedly, I once felt similarly about this whole uniform interface idea in the first place.