I still remember running the self-extracting pkz204g.exe on MS-DOS, and never heard of SEA or encountered their software. That's really unfortunate.
I don't think what Katz did in SEA saga is worth defending, but I think we have to delineate the two formats. As Phil Becker of eSoft says in the clip, the reason that Katz's software ultimately won out was that he was more adept as a businessman—and the open nature of the ZIP format was a good example of that.
Also, remember that PKarc was popular because it was faster than Arc. The feeling in the BBS community at the time were feeling a bit of rage, that a superior product was being killed through the courts. So when PKzip came out, which was not only faster, but had a better compression ratio, combined with the general sentiment in the community that were fed up with look-and-feel lawsuits, doing a mass conversion from arc to zip was also a form of public protest.
Low barriers to entry make software easy to get into... and easy for your replacement to get into. It seems software businesses aren't common any more for that reason (excepting MS).
And games, games developers are still software companies, albeit in the entertainment space.
The difficult programming in games seems mostly deferred to engines (which are becoming cheaper and nore commoditized), and the game is mostly "content" - more akin to a movie than software. But you're right, games are software, and are thriving.
I got two games as a gift recently, and at least one of them require a big download to even function. And both require that i register them with steam before i can play them at all.
Effectively even "offline" games have just become a client to a "service" running in the clouds.
https://www.youtube.com/watch?v=YiaM5MT1Ok8
I admit that I've always wondered what a dramatic film based on this story would look like. It has all of the highs and lows of a movie, and it would be worth retelling for both the lessons it has to teach but also the important role in computer history Katz held.