Well, elsewhere I posted a photo of the one house left standing in a neighborhood after the Pacific Palisades fire.
That guy didn't settle for good enough. The others are out millions.
I can't argue with the logic for videos, they aren't something that people have to use dozens or hundreds of times each day. The occasional error can be handled when a viewer asks for clarification.
The same isn't true of software, at least when it comes to functionality and keeping totals.
I got lots of clients because the company I previously worked at was happy with a good enough total that was plus or minus $20,000 each month.
My first client (for my claim processing business), was very happy to get a new service (based upon software) that balanced to the penny each month.
I had to sweat a lot of details to make that happen. Good enough wouldn't cut it.

I think the distinction that I'm trying to make is that your software can be good enough that it's functional and that it works. It obviously has to work correctly. I'm not talking about leaving bugs in your software and not testing it, but as long as it works and it's functional and it does the job, you don't have to waste tons of time adding all kinds of bells and whistles that most people won't appreciate anyways. That was the point I was trying to get across.

@Richard Got it! The part about most people not appreciating it is so true.
I'm working on an address form right now. It turns on and off a lot of bells and whistles based on entry.
Without that though, I fear a lot of users won't understand how to use it properly.
I think testing takes about 85% of my time. It's amazing the questionable things I find after thinking I thought through it all.