search

SanKumar2015 / EST-coaps

EST over CoAPs IETF draft
1 stars 1 forks source link

ArbitraryLabel of shortest length possible (Klaus WGLC review 2/27/2019) #139

Closed csosto-pk closed 5 years ago

csosto-pk commented 5 years ago

The text says

The ArbitraryLabel path-segment, if used, SHOULD be of the shortest length possible

And Klaus pointed out that

if someone decides to use an ArbitraryLabel path-segment consisting of more than 0 characters (which is the shortest length possible), then it's a protocol violation? This seems to be a matter of implementation quality to me, not a requirement for interoperability.

csosto-pk commented 5 years ago

Need to remove this normative SHOULD. We could just mention that it is worth considering or completely remove.

csosto-pk commented 5 years ago

Removed the normative language and instead put in

Implementers should consider using short such labels to minimize transmission overhead.