I agree with the title but almost nobody uses regex the way he described it. If you're going to work with other people's code you'll have to learn \w \d and \s
I wrote a JS bookmark that lets me jump to the next top-level comment on HN. I use it for finding interesting conversions without having to scroll and look for the next top-level comment.
It works by finding all of the comments with indent=="0". Then, each time the bookmark is clicked, it creates and follows a link. Each comment has a unique id so I just have to concatenate current_url + "#" + unique_id and follow that link.
Once the script reaches the bottom, it follows the "more comments" link and repeats the same process.