On Thu, Aug 15, 2019 at 06:50:38PM +0100, Jason McIntyre wrote:
> try as i can, i can;t really see how this wording avoids the ambiguity.
> nor how the current wording contains it exactly.
s/then the/all/ is an attempt at making it clearer the prefix effects
not just the immediately following cipher but all of them, basically
accentuating the plural in "ciphers" - this seems to be the only detail
in the paragraph that clearly states how on prefix is applied to
multiple ciphers.
> what about
>
> If the cipher list begins with a
I like it, "list" is quite obvious.
> i think it's one of those things that will be clear if you try to
> specify it (like, how else would it work?)
Yes and no: If you play around with various incarnations like
a,b
-a,b
-a,-b
-a,+b
you'll notice values (lists) starting with a dash will generally pass
the parser, whether the make any sense or not. Those starting with a
plus may fail hard.
In fact, while exercising `-o Ciphers=...' I found a new odd cases: I
shall write a separate mail to bugs@ with proper details soon.
> why do you want to remove the list of ciphers but not kex algorithms?
> both contain info on displaying the default using -Q.
Oops, error on my side; I'd zap the kex ones also.
> having said that, i think the idea to trim these may be good. it does
> add text to an already large page, and it would save people having to
> bump this every time it changes.
>
> so maybe we could do this for all places where -Q works? but maybe there
> was a solid reason for listing them in the first place?
I concur; let's see if someone speaks up.
