problems programmatically downloading http://www.w3.org/2004/02/skos/core

Hi

we have a regular maintenance job written in Java where we cache various 
rdf files off the web including
             "http://www.w3.org/2006/vcard/ns", ok
             "http://www.w3.org/2000/01/rdf-schema",  ok
             "http://www.w3.org/2004/02/skos/core",   *500
             "http://www.w3.org/2008/05/skos-xl", ?
             "http://www.w3.org/TR/owl-guide/wine.rdf", *500
             "http://www.w3.org/TR/owl-guide/food.rdf", *500

The marked files have recently started giving us 500 responses - is this 
intended behavior? (Like it is with DTDs)
http://www.w3.org/blog/systeam/2008/02/08/w3c_s_excessive_dtd_traffic
Are other people impacted?





Do we have a workaround other than to manually download and cache ourselves
How does it know that we are running a java client - how many headers 
would I need to fake before it will give me the file?

Reading the above link - actually I will try setting the UserAgent to be 
"TopQuadrant's cache builder for TopBraid Suite"
it is only run internally by us.
I will update here in a bit as to whether that helps

Jeremy