[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [xri] Example local resolver API and corresponding proxy resolver output.
Comments below marked with ##### marks. I-Name: =les.chasen -----Original Message----- From: Drummond Reed [mailto:drummond.reed@cordance.net] Sent: Tuesday, February 28, 2006 1:56 AM To: xri@lists.oasis-open.org Cc: 'Andy Dale' Subject: RE: [xri] Example local resolver API and corresponding proxy resolver output. Les, 1) I agree that it might make sense to return the entire XRD (with the addition of the new URI elements as describe below) but with only the *matched* Service elements. This is because Service selection has already happened, so you are saving the consuming application the job of needing to select them again. However there may be more than one, and this lets the consuming application have the option of using the others. Steve, what do you think of this, i.e., sepResolve returning "all of XRD less non-matching SEPs"? ########### [Chasen, Les] I was not referring to the contents of the XRD. I only meant to highlight that there are three out parameters one of which is an XML service blob. We could just have one out parameter that is the XML XRD blob. Just a suggestion, I do not have strong feelings one way or the other. ############ 2) There's a strong case for calling the URI element by the same name at both the XRD level and the Service level. Besides the fact that it has the same contents (a URI), there are other uses for a URI element at the XRD level. For example, a long time back (early in CD01 days) we discussed returning a URI at the XRD level as one of the items of metadata an XRD author could assert about a Resource. It would be the equivalent of saying, "the resource represented by the query XRI is at this URI". ############### [Chasen, Les] That's all fine and well but I thought it would be more self describing to have another name since the resolver mucked with the field. The URI in the service block is pure while the one in the XRD level is modified by the resolver. I really do not have a strong preference one way or the other. Just highlighting a small area that IMHO can be made more understandable. ################
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]