[[ Mailed copy of news article posted to lists.emacs.rcp ]]
>>>>> "Kai" == Kai Grossjohann <[EMAIL PROTECTED]> writes:
> Btw, I find that using the "rm" or "sm" method is not as slow as one
> might think: in the Emacs 20.4 pretest, base64 encoding and decoding
> is implemented in C, so at least the local part is fast. And invoking
> mimencode on the remote side seems to be reasonably quick.
> Subjectively, this method feels faster than using scp. What do you
> feel?
I still think that a zero-encoding would preferable (using a "randomly"
generated (with host+time+pid, typically) end-string).
The risk of having that specific string appear in the downloaded file
can be made much smaller than most other risks (an emacs bug, a packet
corruption which still checksums, ...) so it's not a real issue.
Stefan
