- From: Seaborne, Andy <Andy_Seaborne@hplb.hpl.hp.com>
- Date: Tue, 3 Jun 2003 17:56:53 +0100
- To: "'Sean B. Palmer'" <sean@mysterylights.com>, "Seaborne, Andy" <Andy_Seaborne@hplb.hpl.hp.com>
- Cc: "'www-rdf-interest@w3.org'" <www-rdf-interest@w3.org>
> Cool. I see from [1] that queries are of the type "GET > http://josekihost:2020/model?query=...&lang=RDQL HTTP/1.1"; Confession - you can have a URI instead of a short name :-) Andy -----Original Message----- From: Sean B. Palmer [mailto:sean@mysterylights.com] Sent: 3 June 2003 12:43 To: Seaborne, Andy Cc: www-rdf-interest@w3.org Subject: Re: RDF Queries in HTTP "Range" Headers > An alternative to using HTTP "Range" is to use the > query string of a GET. Quite, although one has to balance the advantage of addressability with the disadvantage of not having 304 Not Modified ('stead of 206 Partial Content) returned on failures. This means that the query would have to be repeated by the client, though one might want to do so even using Range queries... With both methods there are some minor privacy issues, of course, but one isn't *forced* to use server side querying, so that's okay. > In Joseki [...], the query string identifies the RDF query > language (so is not specific to a fixed language or > predetermined set of languages) Cool. I see from [1] that queries are of the type "GET http://josekihost:2020/model?query=...&lang=RDQL HTTP/1.1"; whilst I think that it'd be in some ways better to use URIs for the query language, it's great that the language is being at least identified. Hmm, perhaps a MIME type would work, though I doubt that anyone's thought about a MIME type for RDQL/SquishQL? BTW, these three sentences in [1] are a joy to read: "The configuration file is in RDF. This document walks through an example below, discussing each of the features. It has used N3 because it can give a more readable layout." :-) [I wrote...] > > One could hack around this by using "uri" as the > > (other-)range-unit, and then using the first token > > of its value as the URI for the type of the rest of > > the contents. I forgot to mention that that is perhaps not such a good idea, since you can't use URIs in the Accept-Ranges header... [1] http://www.joseki.org/configuration.html -- Sean B. Palmer, <http://purl.org/net/sbp/> "phenomicity by the bucketful" - http://miscoranda.com/
Received on Tuesday, 3 June 2003 12:57:21 UTC