Re: A future for the SSH File Transfer Protocol?

[Anthony DeRobertis <asd%suespammers.org@localhost>]

On Sunday, November 11, 2001, at 08:07 , Damien Miller wrote:
> - Inability to do basic filesystem operations (e.g. rename)
> - Inability to access parts of files, or continue aborted transfers

Those are issues, especially the second. But that's not rcp anymore.
> You really want to have a protocol which forces clients to include code
> to protect from a hostile server?

I think, by definition, cryptographic implementations must. And 
we've certainly seen a bunch of mistakes (ssh1, for example). I 
think that's a lot harder than making sure returned file names 
match the glob.

But, sure, if we didn't have to do that, that'd be great. I just 
wanted to point out that is not a fatal flaw.

> If we are going to change a 10+ y/o protocol, why not just 
> write a new one
> without the (many) inherent flaws. Even better, why not just 
> tweak the one
> that has _already been written and implemented_.

No problem with either of those.

> Why are we discussing requirements on somethings that is already
> specified and widely deployed?

Because we're talking about extending it? I'd think would be a 
good idea to specify what is being accomplished by the protocol 
before extending it.