Re: [apps-discuss] media type for partial JSON updates?

["Paul C. Bryan" <[e-mail]>]

On Thu, 2011-12-29 at 18:08 +0100, Julian Reschke wrote:

> On 2011-12-29 17:58, mike amundsen wrote:
> > ...
> > i do not, my message here was too vague.
> >
> > 1) i assumed you were proposing a new document that discussed PATCH
> > for partial updates.
> 
> Oh, sorry.
> 
> > 2) i meant to say that the document you propose "might be the proper
> > place.... etc."
> > 3) i assumed your text "define a matching media type ... It might be
> a
> > light-weight alternative to what
> > <https://tools.ietf.org/html/draft-pbryan-json-patch-04>
> describes..."
> > meant you were proposing a NEW media type to register w/ the IANA
> and
> > i suggested the XML PATCH (RFC5621, i think) as a possible template
> > for designing a JSON variant that would be this NEW media type.
> 
> Ack; understood.
> 
> I think it would be simpler to define it as something that can be 
> trivially transformed to the format defined in 
> https://tools.ietf.org/html/draft-pbryan-json-patch-04; in that case
> we 
> wouldn't even need to describe the actual semantics...


Is this something that just reduces the verbosity of the operations, or
are you talking something probably non-JSON for higher compactness? If
the former, then I'm game to either choose shorter operation names or at
least allow for aliases.

If the latter, what have you got in mind? I'm curious how it might
compare such an approach to merely taking a JSON Patch document (with
possibly shorter op names) and encoding in a binary variant such as
BJSON or UBJSON...

Paul