On 4/28/22 10:07, Stephane Bortzmeyer wrote:
It should say that the IETF shamefully abandoned its own protocol, despite it being valuable and providing features that don't exist in any of the supposed protocols IETF replaced it with.Listing of remote files in a (may be) structured format? (It would be a nice addition to HTTP; I'm sure there is at least a draft with such an extension.) Difficulty with firewalls? Separation of signaling and retrieval? (This is less important with HTTP/2 an 3 streams.)
In spite of attempts to conflate them, FTP and HTTP have fundamentally different models. Of course you can extend HTTP to provide the functionality that FTP has, but people will still think about them differently, provision them differently, and use them differently, and have different assumptions about them.
Or maybe what bugs me is not so much that IETF has abandoned FTP
(though that does bug me) but rather that the Internet still has
no standard, platform-independent, file system access protocol.
FTP has obvious shortcomings for such use, but so do CIFS and NFS
and sshfs, and FTP is arguably still the best of several poor
solutions when you don't need to require authentication.
But rather than try to address this shortcoming, IETF decided to
just give up. Shameful.
Keith
