On 25 Oct 1999, Kai Gro�johann <[EMAIL PROTECTED]>
wrote:

> Pete Forman <[EMAIL PROTECTED]> writes:
> 
>> My question was somewhat rhetorical, I think that making the remote
>> shell interactive will lead to other problems. My .profile is just an
>> example.
> 
> So, are you saying that my change was a bad one and that it should be
> reversed?  Hm.  Even with the old mechanism, one used to get
> interactive shells.  But since nobody noticed, probably nobody really
> used all that.  Hm.

I don't think that this is a serious issue. Given that the first thing,
pretty much, that is done by rcp.el is 'exec /bin/sh', setting a remote
variable for that is probably not much of an issue. 

The connection functions do a fairly good job of sanitizing the
connection I think. If needs be, that can be improved. IMHO, of course.

> Do you think it would be a wrong approach to fix the `ls' problem now
> and to fix other problems as they appear?

I think that the new change is probably a good thing. The only real
problem with it is that I am thinking, vaguely, of trying to get a
slightly faster inline method working (using zmodem transfers) and that
will be irritated with a less than eight bit clean connection.

Still, it's not that much of an issue.

        Daniel

-- 
Of course I lie to people. But I lie altruistically - for our mutual
good. The lie is the basic building block of good manners. That may
seem mildly shocking to a moralist - but then what isn't?
        -- Quentin Crisp

Reply via email to