Re: Query forms should be resources, not operations

On Thu, Jul 28, 2005 at 03:00:12PM -0400, Mark Baker wrote:

> the same arguments with different operations).  Instead of POSTing a
> SPARQL query that consists of an embedded operation, the query (without
> the operation) would be POSTed to one of the aforementioned four query
> processors.

Mark,

I'm not entirely sure what the import of yr comment is. Can I ask for
clarification?

For example, the present design motivates an implementation like this:

GET /sparql/?[query & dataset]

or

POST [query & dataset] /sparql

Where "query" is "SELECT ?a ?b ..." or  "CONSTRUCT ..." or "DESCRIBE
<uri>..." or "ASK ..."

Yr suggested design, insofar as I understand it, motivates an implementation
like this:

GET /construct?[query & dataset]
GET /select?[query & dataset]
GET /ask?[query & dataset]
GET /describe?[query & dataset]

or

POST [query & dataset] /construct
POST [query & dataset] /select
POST [query & dataset] /ask
POST [query & dataset] /describe

Where "query" is similar to the pseudo-queries above, but with the query
form productions removed. Thus, for example, instead of "SELECT ?a ?b WHERE
{<foo> ?a ?b}", the on-the-wire query would be "?a ?b WHERE {<foo> ?a ?b}".

Is that what yr suggesting?

Cheers, 
Kendall Clark

Received on Tuesday, 2 August 2005 16:32:27 UTC