Re: Formal consultation prior to closing the secsh working group

[Nicolas Williams <Nicolas.Williams%sun.com@localhost>]

On Mon, Jul 31, 2006 at 09:16:38AM -0400, Richard Whalen wrote:
> The Filexfer draft was never a ftp-like protocol.  It was originally a
> file access protocol that was missing a text access method.  After the
> text access method got added it has moved towards becoming a file system
> protocol.

IMO, publish the parts of the filexfer protocol that are commonly
implemented as an Informational document (and then only as Informational
and not targetting Standards-Track if it fails to meet general IETF
requirements, e.g., I18N).

I don't see why a text access method should be required for a
Standards-Track filexfer protocol for the Secure Shell protoco, but hey,
I'll take an Informational document over no document at all any day.

> I feel that the filexfer draft has grown so much that many don't have
> the resources to implement the current drafts, hence many implementations
> are still a few versions back.

Agreed.

> I feel that there is a significant need for the technology though and
> would like to see a protocol for the transfer of file data over SSH
> standardized.  Secsh may not be the right place for this.  SSH may be
> well enough established that people can think of it as a transport and
> that the proper protocol can be developed in the applications area.

Agreed.  The protocol should be reviewed in the apps area, by apps area
experts and ADs.  And since the protocol borrows from NFSv4 it should
also be reviewed by NFSv4 WG participants, at least with respect to the
parts of NFSv4 that filexfer does borrow.

Nico
--