Re: xml:lang in JSON results format

Alistair Miles wrote:
 >
 > Hi all,
 >
 > I'm sure you must have talked about this already, but I searched the
 > groups lists and couldn't find anything.
 >
 > Comment on:
 > 
http://www.w3.org/TR/2007/NOTE-rdf-sparql-json-res-20070618/#variable-binding-results 

 >
 >
 > """
 > RDF Literal S with language L
 >
 >     XML: <binding><literal xml:lang=" L "> S </literal><binding>
 >
 >     JSON: "name" : {"type":"literal", "xml:lang":" L ", "value":" S"}
 > """
 >
 > Please can it be...
 >
 >     JSON: "name" : {"type":"literal", "lang":" L ", "value":" S"}
 >
 > .. instead. It's a matter of convenience, but I'd like to be able to use
 > the dot operator to access the language tag in a plain literal variable
 > binding. Having to do binding["xml:lang"] is a real pain.

Hi Alistair,

I (very much) sympathize with your comment but at this point it's pretty 
much too late to consider changes. The working group last published the 
Note on the JSON results format half a year ago, and that only removed 
the "distinct" and "ordered" attributes which had been removed in the 
XML results format. The rest of the JSON format has been stable and 
published since October of 2006.

The Working Group is largely in hibernation mode right now and given 
that and the relatively widespread deployment of the current JSON result 
format, I don't see it likely to change.

I tried to look back through some emails to see a history of "xml:lang" 
as the JSON key, but couldn't find anything conclusive. My best guess is 
that it was done as an effort at consistency with the XML attribute, 
which was a motivation that drove a lot of the JSON result format design 
(even at the expense of making the JSON format somewhat more verbose 
than it otherwise need be).

Lee

 > Otherwise, thanks for the great work.
 >
 > Alistair.
 >

Received on Friday, 14 March 2008 03:40:29 UTC