Update: Guides for using Microsoft PivotViewer with SPARQL endpoints


An update to what I sent out yesterday. There are also two FAQs feeding 
off comments and discussions I've had with people offline. For example, 
why do some of my demos take so long to run etc.. The answer lies in 
understanding the degree to which SPARQL and PivotViewer have been 
integrated re. dynamic CXML generation.


I have a demo that uses the SPARQL endpoint at RPI as part of a 7-day 
earthquake tracking meshup [1]. The SPARQL in question is SPARQL-FED, 
and I've set no LIMIT on the resultset -- bar filters in the original 
query by the RPI team. As you can imagine, SPARQL-FED on its own is 
inherently challenging let alone incorporating that into a query 
pipeline that includes image process en route to final CXML generation 
and presentation in the browser.

We address these uncontrollable variables by offering an "Edit" URL that 
places you inside a SPARQL query editor. Once you have the editor in 
scope you can alter the query to suit your needs. Of course you can also 
alter a variety of options that ultimately control the behavior of the 
Pivot page.

Note: the LOD cloud cache will be offline for HTTP access for most of 
tomorrow. We are adding an additional 5 - 11 Billion triples (from 
various sources) to the 15 Billion currently in place.


1. http://uriburner.com/c/DVMO43 -- 7-day earthquake tracking meshup

-- SPARQL query editor within PivotViewer page

-- PivotViewer and Linked Data guide

-- FAQs

5. http://virtuoso.openlinksw.com/dataspace/dav/wiki/Main/VirtSparqlCxml 
-- SPARQL and dynamic CXML guide

-- FAQs



Kingsley Idehen	
President&  CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen

Received on Friday, 25 March 2011 19:56:14 UTC