[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [search-ws] Media type vs. response type
I’m happy with that. Ralph From: Ray Denenberg,
Library of Congress [mailto:rden@loc.gov] From
the meeting minutes: "Media type vs. response type. We have not sufficiently
distinguished between these. Since application/sru+xml is both a media
type and a response type, and because it is expected to be the predominant
response type, we have overlooked the fact that a media type is not always a
response type. If you want a response in ATOM then you will indicate that as
the media type (via the httpAccept parameter or via http accept header). But
ATOM is not itself an SRU response type. There is an however an ATOM
extension supplied in an annex. But there is no way to indicate in a
request that that particular extension (vs. another ATOM extension which is
also an SRU response type) is desired. Discussion on this will continue
via email. There are two suggestions: (1) add additional values for
recordPacking parameter; (2) add an additional parameter, responseType.
Objections to the first suggestion are that this is at the response level, not
record; it stretches the semantics of the recordPacking parameter; and it is
not extensible. Objections to the second suggestion is the complexity of adding
another parameter. " So
let's please move this discussion along:
I propose that we add a parameter responseType. It's
value would be a URI, identifying a defintion of some extention (not necessarily
registered) to a media type that tells how that type is to be used as an SRU
response. --Ray |
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]