And yet TSR/WotC keeps making new editions. And somehow I doubt they do it just to throw a few more game designers a job for three or so years.
There’s a balance.
You need release product to justify your existing staff and keep the product in the public eye (and sell more core books). If you’re just selling four books, you don’t need any staff to make new books. But eventually sales of accessories and expansions dip and your run out of new books to release. At that point, a new edition is useful for sustaining the game.
But making new editions too often results in diminishing returns as fewer people upgrade each time, and if you release too fast you’re losing players faster than you’re earning new ones. And each successive edition has to be better than the last to encourage upgrading. If an edition is not an undeniable improvement in almost every way people won’t switch.
Game makers could also make this easier on themselves with smaller new editions, ones that can be handled in a year of downtime rather than three, reducing the development cost by 2/3rds.
D&D is really an anomaly with its massive sweeping changes between the last three editions. Many other game systems release new editions that have much smaller revisions, closer to the changes between 1e and 2e or 3.0 and 3.5.
Really, this is not unique to RPGs. You can see a similar thing in a market as different as computer Operating Systems. Microsoft needs to keep releasing new versions of Windows to justify its existence, when it could just as easily just patch an existing OS indefinitely. But if they release OS too often few people want to willingly upgrade.
MS releases new OS twice as often as they should, as every other OS seems to be ignored.
But hey, like I said, the value of long-term support is debatable. I certainly wouldn't complain if WotC were to announce "We've recently realized that 5e just isn't worth it, so we've decided to continue supporting 4e for the foreseeable future." But I don't see that happening, or anything similar happening after 5ish years of 5e.
And I wouldn’t mind them deciding to go back to 3e or make a Revised 3e that was 3e with fixed math and better balance. But they’re unlikely to do that either.
Really, they didn’t decide to end 4e early for gits and shiggles. If they could have continued to release the edition and make money, they would have. You don’t decide to tank profits for 2 years and gamble your jobs and the future of the game if not absolutely necessary.
Honestly, I like their approach of “back to basics” with 5e, looking at the simplicity of earlier editions such as OD&D and BECMI. I think they’re taking the right approach with simplicity that you can layer complexity atop. I just wish they’d actually
show us the modularity.