Re: IUTF8 pseudo-terminal mode

[Jacob Nevins <jacobn+secsh%chiark.greenend.org.uk@localhost>]

der Mouse writes:
> [Colin Watson:]
>>           42    IUTF8       Assume input characters are UTF-8 encoded.
> 
> Actually, if you're going to be adding such things, how about adding
> bits for ECHOPRT, ALTWERASE, NOKERNINFO, and character size (CS5..CS8)?
> In my own implementation, I created a private channel request to carry
> those bits....

(You have CS7 and CS8 already, although you have to be sensible about
which combinations you enable.)

Well, the IETF seems to care more about UTF-8 than those other
things. :)

Given that the SSH protocol doesn't assign any semantics to these
terminal modes -- they are just tokens passed from one side to the other
-- it's a shame that they don't have extensibility and private-use
facilities, unlike most of the rest of the protocol.

If it's too late to get IUTF8 (etc) into the core drafts, and consensus
is that it's worth a new document, it's probably best to define a
channel extension like yours with textual names, private-use provisions,
etc, rather than just assign some new code points in the existing
mechanism (given how long it looks like it will take to get _any_
document through).