Everything is 0s and 1s to a computer. What a pattern of 0s and 1s encodes is decided by people–often arbitrarily. Over the years there have been attempts to standardize encodings but, for legacy reasons, older encodings are still valid.
The 0s and 1s that encode ’ in UTF-8 (a standardized encoding) are the same 0s and 1s that encode ’ in CP-1252 (a legacy encoding).
The � symbol is shown when the 0s and 1s don’t encode anything of meaning.
I don’t think Lemmy supports media fields in comments (though I’ve only skimmed the API, I could be wrong) just on posts. I usually use Postimages for hosting images for comments.
At least where I work, the developers who actually wrote the code would probably never see it. We have service staff that deal with an initial problems reported from customers. They’d likely figure out someone actually entered those values.
They are quite well seasoned. But it’s also worth noting they are developers as well because the job usually has you debugging things or writing code that needs to be run for the specific customer. Not a large amount of code, but just things that end up being specific to a customer.
And if they have to come to a developer that actually works on the product, it’s usually a pain to try and figure out what is going on. Thankfully, this is very uncommon.