I don't get what you mean here. qw/ac qw/ and qw/ca qw/ are treated identically by the flip-flop operator:

Yes, I know that. Also the description on perldoc seems clear if you already know what to expect. My whole point is that the result of perl -we'my @l = qw/ca qw/; for (@l) { print if /a/../c/ }' is not logical.

Suppose that I want to process html file where I have tags like this:

<tag>
a
</tag><tag>
b
</tag><tag>
c
[download]

If my flip-flop condition is /<tag>/../<\/tag>/ it would print

<tag>
a
</tag><tag>
</tag><tag>
[download]

which is not what you want. With three dots you get

<tag>
a
</tag><tag>
</tag><tag>
c
</tag>
[download]

still not good :(

Basically, what I originally (and incorrectly as I see now) expected from this operator was that it processes each element char by char (as a real flip-flop) and not as a whole. Of course, in my html example I can first format the file but that is another story.

BTW: I am new here and have a question about writeup formatting. How do you cite someone's text if it contains a code? Copy-pasting and than manually putting those code tags back (which I did) seems pretty tedious.