[jasode]

>I also don't think it's terribly relevant; 1995 was a quarter century ago.

It's relevant because the context of the conversation was this gp's quote you responded to: "Back then people still thought Java would take off on the desktop."

The "back then" is referencing XUL circa ~1997. And the "Java would take off" was the ambitious idea of most desktop apps being written in Java to weaken the MS Windows ecosystem. Not only was Java hyped to be a Microsoft killer, it was also touted to be a C/C++ killer. (E.g. the idea was that computer desktops have gotten so powerful with so many wasted cpu cycles that manual memory of C/C++ is obsolete and letting GC use the excess cpu to automatically manage memory is the future.) History has now shown us that prediction didn't happen either. C/C++ is still heavily used for new desktop apps. That's a different idea than today's 2019 landscape with some niche Java apps like Jetbrains IDEs.

I do understand your point. Yes, you can also have an alternative definition of "not a failure on desktop " because you can count some current Java apps today. That's also a valid perspective. However, for the sake of not confusing the conversation... that's not what the gp was originally talking about. I don't think there's any misconception about what "Java failed on the desktop" means -- especially among the HN audience. I also regularly use Jetbrains IDEA for Android deveopment and Webstorm for Javascript but my usage of those Java apps doesn't change what "Java failed on the desktop" means to other people.

[topspin]

Yes, I moved the goal posts making my point. I believe the storied history of Java clouds the contemporary reality of Java too much, to the point where one can argue Java "failed on the desktop" specifically among people that spend a good fraction of their waking hours using Java desktop software, and I appear to be one of the few that notices this irony.