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

Bug#212034: Debian Perl Policy manual uses "dependency" backwards



Scott:


Is ambiguity in the quoted part of the manual causing confusion?  

First, does the second occurrence of "that package" in:

  Only one package may contain the /usr/bin/perl binary and that package
  must either be perl or a dependency of that package.

refer to perl?  That is, do you agree that this is equivalent?:

  Only one package may contain the /usr/bin/perl binary and that package
  must either be perl or a dependency of perl.

Second, is this statement correct?:

  The binary must be contained in the perl package or a package on which 
  perl depends.

That is, do you agree that that statement is what the author of the 
manual excerpt intended to say?

If any of your answers are "no", then the problem may not be that 
"dependency" is used wrong (and either the wording is _really_ screwed
up and/or I'm misinterpreting what the manual is trying to say in the
first place).

If all of your answers are "yes":

Then you consider "a dependency of perl" (as used above) to mean
"a package on which perl depends," right?

If so (sorry for all the conditionals), then:


As you quoted, your dictionary says that "that which is dependent" 
is a dependency.

So let's consider perl and a package on which perl depends.

The other package does not depend on perl, so the other package is 
not dependent.  (Agreed?)

Since the other package is not dependent on perl, then by your own
dictionary's definition, the other package is not a dependency of 
perl.  (Any divergence between us yet?)

Therefore, since the other package is not a dependency of perl, the 
phrase "dependency of perl" does not refer to the other package.  
(Fairly trivially.)  (Still agreed?)

Since the manual is trying to refer to a package on which perl 
depends, but uses the phrase "dependency of perl," which does 
not refer to a package on which perl depends, the manual's wording 
is wrong.

If you don't still agree, then where do we diverge?



Daniel
-- 
Daniel Barclay
dsb@smart.net



Reply to: