Re: Review of SPARQL Protocol

On 11/21/2011 5:37 AM, Steve Harris wrote:
>  From Nick Lamb of Garlik, who implemented most of our HTTP Server.

Thanks, Steve, and please pass along our thanks to Nick as well. 
Responses inline.

> --
>
> The removal of WSDl definitions throughout has improved clarity.
>
> --
>
> ...However 3.1.9 still refers to MalformedQuery, which no longer appears
> in this way elsewhere in the document. The same for QueryRequestRefused
> in 3.1.10

Removed these.

> --
>
> Some internal links such as:
>
> http://www.w3.org/2009/sparql/docs/protocol-1.1/Overview.xml#SparqlProtocol
>
> ... don't function in the document.

think I got these, but we'll get the rest when we run linkchecker..

> --
>
> Some guidance for how a SPARQL server can report diagnostics in 2XX
> responses to UPDATE operations would be valuable. Right now the standard
> ducks the issue. If it's too late to fix in this document, a future WG
> might be tasked with addressing it?

I think it's too late - we explicitly punt on this in the document right 
now. I'll add it to the future work wiki page.

> --
>
> Section 4.1 seems a bit thin. Perhaps someone from the security
> community could be invited to help write more here? Existing best
> practice documents for query languages such as SQL might offer some
> insight too.

Will ask Sandro for advice here on whom we might ask.

> --
>
> A.2 lists IETF RFC 2781 (UTF-16) but nowhere else in the standard is
> this encoding mentioned or implicitly used. Yet in contrast UTF-8 is
> used throughout, but RFC 3629 (the current Internet standard for UTF-8)
> is not referenced in A.2
>

thanks, we'll clean up the references.

Lee

Received on Monday, 21 November 2011 19:45:16 UTC