It generates an answer that looks correct. Actual correctness is accidental. That’s how you wind up with documents with references that don’t exist, it just knows what references look like.

You couldn’t crank your CPU in the olden days, it’d make games run in fast forward.

Nobody wants to deliberately use the wrong compression type when extracting, so modern tar will figure out the compression itself if you just point it at a file. So tar -xf filename works on almost anything. You don’t need to remember which flag to use on a .tar.bz2 file and which one for a .tar.xz file.

“Hey, here’s a useful thing that I recommend to people: <your work>”

It’s basically a compliment

When my phone’s barcode reader app sees a web link, it fetches the page’s title to display next to the actual link. So it is going to that web server and fetching resources by itself. Even though it isn’t actually rendering the page and running javascript, it might be exploitable.