$haha =~ s/tyre/tire/g;
[download]

However, I'll also give the direct answer to your question: Perl and shell have a great deal in common, for example the $ sigil on variables, the interpolation of variables inside of double quotes ("...") and backticks (`...`), and the (shell) execution of a string in backticks. The important differences (for this example) are that

• in perl, you need to use the $ sigil on your variables, even when they are on the left-hand side of an assignment (unlinke in shell)
• shell will automatically handle whitespace* from the results of something executed by backticks (which perl will not do... perl will just give you the output of the subshell exactly, so, in this case, including the trailing newline from the echo | sed)

$haha=`echo -n $haha | sed "/tyre/tire/g"`;
[download]

• changed haha=... to $haha=...
• adding a -n option to echo. This makes echo not include a trailing newline. If echo had included that newline, then sed, too, would have passed that trailing newline, and perl would have not trimmed it off, so you've had ended up (inadvertantly) adding a newline to the end of $haha.

NOTE Regarding the shell and it's "handling" of whitespace: in general, the shell is actually going to be doing a lot more than just trimming the trailing newline... what the shell does is basically split all the output on any whitespace (including spaces, tabs, newlines), ignore any leading or trailing whitespace. It will use these "words" as a list, if the backtick is used in a list context, and rejoin the split-apart "words" with single spaces if it is using the results of the backtick as a string.

------------
:Wq
Not an editor command: Wq
[download]