This headline is unfairly editorialized. What the Firefox dev said was that the automatic encoding detecter confusing some ISO/IEC 8859 encodings for each other is an intractable problem, and fixing one Latin text encoding being auto-detected for a different Latin text encoding is a very low-priority use of time. Also, removing the encoding override is completely separate from that, since it's (apparently) a very complicated feature to keep around that 1) breaks webpages (e.g. etrade) and 2) keeps confusing users and leading them think Firefox is worse at text encoding then Chrome.
Regardless of your opinion of the change in question, these two statements are completely different from what the headline implies.
He straight-up said that it’s not as much of an issue because it’s still legible because not all of the letters are borked. I don’t see how this is a misrepresentation of what he said.
> Also, removing the encoding override is completely separate from that,
It is not separate at all. If the detector cannot do it, and the ability for the user to do it is removed, these issues are intrinsically linked. Sure, there are the other issues you mentioned around keeping it around, but you can’t act like these issues can be considered separately.
The etrade breakage was a failure of the previous detector, which expected an encoding declaration within the first few page KB and introduced a page reload otherwise. Switching the current encoding manually does not reload the page per se.
Regardless of your opinion of the change in question, these two statements are completely different from what the headline implies.