I was one that said “no” the second time.  Thank you for the extra clarity, 
Ahmed.  While I’m a soft “no”, the reason I said it was based on Ahmed’s 
description of the “why”.  He said, there may be a bug where the device sends 
YANG-looking garbage.  That feels more like a corner case and one that 
consumption code would likely catch and propagate anyway.  Moreover, it could 
be that the scheme is technically invalid, but still workable by the consumer 
code.  In this case, rote validation might end up producing suboptimal results.

That said, I think the notion to ensure integrity of the data is useful, so I 
can fully understand the WG wanting to pursue this.

Joe

_______________________________________________
netmod mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to