Reports of other software incompatibilities due to the ROM changes would start to come out once the 1200XL was actually released and got into user’s hands, hurting its reputation.

That wasn’t as big a deal in the 80s as it is now. Reputation was limited to real life friends and maybe a few homegrown newsletters or computer clubs.

Very few people were using the Internet to share opinions in the early 1980s, so “reputation“ could be very effectively managed by Atari and other companies through advertising and leaning on trade media to suppress negative reviews and angry letters to the editor.

That is, unless the problems were too big to ignore and customer anger became too great, as was the case with many late era Atari 2600 games.

A bigger issue for the 1200XL was price as well as something not addressed in the article: competition. By this point there were other platforms to consider, often at better price points with attractive features and software.

>That wasn’t as big a deal in the 80s as it is now.

It was a big deal for me. Software expenses were a huge portion of the cost of owning a computer.

Almost always the price of the computer was less than the cost of buying software to run on the thing.

Letter Perfect was around $300. If it didn't run on the 1200XL I'm not shelling out $800 for the computer and another $300 for a compatible word processor.

I am convinced that cross-vendor incompatibility was THE reason for CP/M's failure. Not anti-competitive behavior, not shenanigans, but the fact that if you spent $495 on the Kaypro version of Wordstar and then bought an Osborne, it wouldn't work. Same Z80, same CP/M, wouldn't work.

Even today PC manufacturers are only starting to remove the BIOS compatibility layers that allow you to boot >30-year-old versions of DOS on a modern hardware, and Apple has provided binary translators since the 1994 PowerPC transition and supported them for years after breaking native compatibility.

BBSes were a thing back then, and while it wasn't the interest you did have large discussion. If you could afford compuserve (which charged by the minute!) you had a nationwide audience on a platform that was bigger than the internet of the time. A few people also had access to the internet (via their university), or at least usenet (via their work or the internet) and so there was discussion that way - but compuserve was where it was at.