On 11/18/20 5:34 PM, John E Drake wrote:
Is there any way you could summarize your perspective in a single email and then stop?
I can try.
1. Discontinuing FTP service will be disruptive no matter what. Broken
scripts, broken links, tools that people use will no longer work.
2. Discontinuing FTP without replacing it with something equivalent
(e.g. a file-access service that can list directories, permit tree
walking, distinguish ordinary files from directories from symlinks in
addition to permitting downloading) will rob IETF document users of
valuable functionality that is not provided by either web browsers or rsync.
3. WebDAV appears to be a functional equivalent to FTP for this
purpose. However, replacing FTP with WebDAV looks like replacing a
mature and well-established protocol that enjoys very widespread client
support, with a less mature, less widely-supported, and even more
baroque protocol than FTP, in addition to being disruptive. Though it
would permit encryption, so that's a plus for WebDAV.
(In a brief search I did not find evidence that some countries are
[still?] blocking https, only that some countries are blocking TLS 1.3 +
ESNI. )
4. I would be surprised if the opex associated with WebDAV were actually
less than running an FTP server, but it's possible.
IMO continuing to support FTP is a better decision than migrating to
WebDAV, but either of those alternatives is better than simply pulling
the plug on FTP.
Keith
