Re: I-D ACTION:draft-ietf-secsh-publickeyfile-08.txt

[der Mouse <mouse%Rodents.Montreal.QC.CA@localhost>]

> This is just getting silly... no, more than that, it's ridiculously
> pedantic.  Yes, it is in theory possible to find a record-oriented
> system that doesn't support flat text files.

However, that's not what I'm talking about.

I'm talking about systems that support "flat text files", but in ways
other than as an octet stream with line boundaries marked by
distinctive octet sequences.  Some text file types on VMS, for example.

> However, if anyone were ever to port SSH to CICS, I'm sure they could
> write a special profile for it that would allow it to interoperate
> with all the other CICS implementations of SSH.

Quite possibly, but, as written, they would not be able to use
publickeyfile (assuming, that is, that CICS text files cannot use
line-termination octets).  They would at best use a similar format that
did not require line-termination octets to represent line breaks -
basically, publickeyfile without the folderol about line termination
characters.

> In the meantime since MIME, PGP, XML, and a million other flat-text
> formats have somehow managed to struggle by without worrying about
> this, why is it a concern here?

Quite so.  I would prefer that all the language about line-termination
characters be ripped out and the spec be written purely in terms of
lines, without specifying anything about how those lines are
represented.

> The coverage of the flat-text format in the spec seems perfectly
> clear and completely adequate to me, I'd suggest that if anyone feels
> the need to add 20 pages of additional wording explaining to readers
> what a text file is, they may prefer another line of work.

(1) I point out to you the language about line termination characters,
which is unimplementable nonsense on OSes that don't use line
termination characters to delimit lines; (2) 20 pages is a gross
exaggeration; (3) I am not suggesting adding any text, but rather
removing text (unless someone is steadfastly opposed to removing the
line termination language, and even then only the occasional "if any"
note or equivalent); and (4) this (proposed) change is not an attempt
to explain or specify what a text file is, but to avoid constraining
what a text file is in ways that are incompatible with some
environments.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse%rodents.montreal.qc.ca@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B