Re: Publickey subsystem draft posted

[Jon Bright <jon%siliconcircus.com@localhost>]

Bill Sommerfeld wrote:

> > If I should avoid wholesale editing of other people's documents in the 
> > future, please let me know :-)
>
>
> One common fix to this problem is to add the new editor's name to the
> document ;-)

I wouldn't want to go treading on other people's toes :-)

> other meta-comments:  [not speaking as WG chair..]
>
> it would be helpful to also add a list of some well-known
> restrictions, while making it clear that precise interpretation is a
> local matter on the server.

I'm not sure what you mean by this - we're already listing x11, shell, 
exec, agent, env and subsystem?

> Decide whether to call them "attributes" or "restrictions"

At the moment, they're two different things (and I think I used the 
names consistently to refer to one or the other).  I'm increasingly 
inclined to the view that the restrictions should become attributes.

> Having a "critical bit" for restrictions may be important.  (this
> means "fail the request if I don't know what to do with it")
>
> Maybe the way to go is to define restrictions as attributes with the
> critical bit set?

I think so.  If everyone else also thinks so, I'd be happy to produce 
another version of my edits with this changed.

--
Jon Bright
Lead Programmer, Silicon Circus Ltd.
http://www.siliconcircus.com