On Fri, 02 Jan 2015, Kees van Eeten wrote to Benny Pedersen:

 KvE>  The general reaction to inclusion is, that a higher level C
 KvE> should not make  any changes to the segments offered for
 KvE> inclusion.

 KvE>  In a way this makes sense. The lower level C knows, that he
 KvE> submits a segment  with errors, or ignores the results of the
 KvE> testrun, that is available to him. 

errrmmm... there are/were many *Cs that manually maintained their segments in a
plain old text editor with no assistance from any software... there is no
guarrantee that they had any sort of ""testrun"" or "validation" of their
segment other than that returned by the next *C up the chain... those *Cs
should be able to edit the submitted segment to fix it so as to ensure that the
net or region covered by the broken segment is not left out of their segment
being passed on upstream...

let's not fall into the fallacy that everyone can even run makenl or any other
nodelist tool or that they even have the desire to do so...

)\/(ark