[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#426491: ucf: [debconf_rewrite] Debconf templates review



Manoj Srivastava wrote:
> Hi,
> 
>         Thanks for your efforts for the translations for the ucf
>  package.  This is much appreciated.  Unfortunately, I will not be
>  applying the patches unmodified, since I  do not agree with some of the
>  changes suggested. (I am also not fully in agreement with the
>  antiseptic tone being advocated wrt user interaction; but that is a
>  discussion for another time)

That's fine by me.

> [...]
>         Next, cannot and can not are both correct -- in different
>  contexts.  If the usage is opposite can -- if I am unable to perform
>  some task, then I cannot do it.  This is not the case here -- I
>  obviously _can_ label the file a conffile, as long as I am wiling to
>  forego some desirable aspects of the situation.  I also can _not_ make
>  it a conffile.  

Sure you can.  But that isn't what "can not" means here:
# This script attempts to provide conffile like handling for files that
# can not be labelled conffiles, are not shipped in a Debian package,
# but handled by the postinst instead.

This is talking about files that "want" to work as conffiles in the
normal fashion, but can't (a problem for which ucf is the solution).

It is *not* talking about files that can (are free to) be unlabelled
as conffiles - a category which would include man pages, binaries
and so on.

If the structural ambiguity in the scope of the negation ends up
causing confusion even for the writer, that's a particularly good
reason for a policy of avoiding "can not" in favour of "cannot".

>         When written as two words, one may imagine an emphasis being
>  placed on the word not.  In this case, writing it out as two words
>  correctly conveys the nuances it was meant to convey.
>
>         However, since this is obviously creating some distress, how about:
>    This script attempts to provide conffile-like handling for files that
>    may not be labelled as conffiles.

I would advise finding a wording that resolves the ambiguity rather
than preserving it, and resolves it in the direction of the sense
"cannot be labelled".

Or if I'm wrong, and you really did originally intend the other
sense (ucf "attempts to provide conffile-like handling for files
that don't need it") then it's going to take a lot more rewriting to
make that clear. 
-- 
JBR
Ankh kak! (Ancient Egyptian blessing)



Reply to: