But I'm comfortable using map, and I don't see anything wrong with using it in void context.

Here's a node I would refer you to:  is the use of map in a void context deprecated ?, which elaborates further.  In it, liz says:

...I seem to recall that map was just rewritten to check for void
context and avoid the extra work if possible.

    That is correct. And it's in Perl 5.8.1!
    From the 5.8.1. perldelta:

        "map" in void context is no longer expensive. "map" is now
        context aware, and will not construct a list if called in
        void context.
[download]

And I particularly like Abigail-II's reply:  "Think for yourself.", as well as Abigail-II's many successive comments in the thread.

Update::  I also have to mention BrowserUk's final word in the thread, which links to this:

Larry Wall (<larry {at} wall.org>)
Re: grep/map in void context
perl.porters-gw

The argument against using an operator for other than its primary
purpose strikes me the same as the old argument that you shouldn't
have sex for other than procreational purposes.  Sometimes side
effects are more enjoyable than the originally intended effect.
[download]

Thanks for the references. I understand Abigail-II's arguments, but I find that I agree with tilly's reply and demerphq's reply to the reply. When I see map in void context, it makes me wonder if the programmer meant something different than what's written. Was that line an assignment before? Was the returned list being used for something? Did the programmer forget something?

Not long ago, I encountered "(my $x, $y) = f()", which is equivalent to "my $x;($x, $y) = f()". As it turned out, that is what the programmer intended, but I spent quite a while finding that out, because at first glance I wondered if it was supposed to be "my ($x, $y) = f()".

(Back to map.) Given that the equivalent code (with for) is just as easy to write, the map stands out. If you're using it to get a list context (as mentioned in Re: Think for yourself.), then I think that would make an excellent comment right above.

(As a semi-side remark, in Re: Think for yourself., Abigail-II says that a familiarity with map (from experience in other languages) that predates Perl's map is some part of the reason for seeing it as just another looping construct. I, too, was using maps in Scheme before I ever saw it in Perl, so I find it interesting that I don't share that view of it.)

Thanks again for the references. It really is fascinating reading, and I recommend it to any other monk reading this.