RE: Line termination philosophy [Re: I-D ACTION:draft-ietf-secsh-publickeyfile-08.txt]

["denis bider" <ietf-ssh%denisbider.com@localhost>]

It is fine for the public key spec to explain that the ability to parse any
CR/LF line termination is desired. However, the draft does so clumsily and
without explaining the problem. The strength of this requirement needs to be
reduced to make it a recommendation, and it needs to be explained that the
requirement is there so that no human intervention is required to make
public key files work after being transferred in binary form, e.g. using
SFTP version 3 (the most common mechanism for transferring public keys).