Comments on: Typical Microsoft Bull***t http://www.mkbergman.com/402/typical-microsoft-bullt/ Mike Bergman on the semantic Web and structured Web Thu, 11 Mar 2010 14:03:35 +0000 http://wordpress.org/?v=2.8.4 hourly 1 By: c http://www.mkbergman.com/402/typical-microsoft-bullt/comment-page-1/#comment-46215 c Thu, 13 Sep 2007 08:56:08 +0000 http://www.mkbergman.com/?p=402#comment-46215 why are you complaining about microsoft reinventing and not following standards? i dont complain about an apple not being an orange.. microsoft.. post about them when they do something shocking :) why are you complaining about microsoft reinventing and not following standards?

i dont complain about an apple not being an orange..

microsoft.. post about them when they do something shocking :)

]]> By: Mike Pittaro http://www.mkbergman.com/402/typical-microsoft-bullt/comment-page-1/#comment-46210 Mike Pittaro Wed, 12 Sep 2007 16:10:27 +0000 http://www.mkbergman.com/?p=402#comment-46210 Here at SnapLogic, we haven't had an requests for RDF as an output or syndication format yet, although it could be added. Right now, the main interest we've seen has been in understanding the data processing within a pipeline, and RDF is essential for solving that problem. We use RDF as the basis for our metadata model, where every REST resource has an RDF description of the output available. RDF is a very natural fit for this. Whenever a resource is defined, the system automatically maintains the detailed description, which enables some interesting capabilities that we are just beginning to tap. For starters, the RDF description is available with a simple GET, so clients can determine what's available at a particular URI, and what the record set looks like. Within the system, its also possible to query the metadata store, to analyze the defined resources. This makes it possible to follow the flow of data through pipelines, even across servers. It's not a case of linked data per se, but more a case of 'where did this linked data come from?'. Here at SnapLogic, we haven’t had an requests for RDF as an output or syndication format yet, although it could be added. Right now, the main interest we’ve seen has been in understanding the data processing within a pipeline, and RDF is essential for solving that problem.

We use RDF as the basis for our metadata model, where every REST resource has an RDF description of the output available. RDF is a very natural fit for this. Whenever a resource is defined, the system automatically maintains the detailed description, which enables some interesting capabilities that we are just beginning to tap.

For starters, the RDF description is available with a simple GET, so clients can determine what’s available at a particular URI, and what the record set looks like.

Within the system, its also possible to query the metadata store, to analyze the defined resources. This makes it possible to follow the flow of data through pipelines, even across servers. It’s not a case of linked data per se, but more a case of ‘where did this linked data come from?’.

]]>