I tried building something with xulrunner a while back and the experience was awful, IMO. I was up and running with Electron within an hour.

xulrunner may be a superior technology, but if it's a PITA to work with, then nobody is going to want to use it.

I haven't used electron, but in 2007 xulrunner was such a pain in the ass to write code for it wasn't worth the effort for anything where you didn't absolutely need an embedded browser, and even then there were better options.

Documentation was poor or non-existent and IIRC the API was based on the half-baked XPCOM because COM and CORBA were all the rage at the time it was created.

Wasn't xulrunner based on XPCOM which Mozilla deprecated? I think this is one of the technical reasons for its decline, but I think the biggest reason was timing.

When xulrunner was being developed there wasn't the same interest in desktop software. It seems like Chromium reached a point where you could build desktop-class software, but the distribution model wasn't right. Someone then decided to adapt Chromium to solve the distribution problem and you have Electron.

If Firefox had evolved to allow the tipping point experience that Chrome did, then you might well have seen xulrunner be that thing. It was years ahead of itself, but perhaps you could question the focus of the folks on the Mozilla side who were spread across so many concepts/ areas compared to the folks at Google - but I think that's for a different thread.

JS had nowhere the popularity it had today. At the time, creating a desktop app in Javascript was not regarded as serious. The reason electron became popular is because so many people from other fields (designers, integraters, students...) learn programming through copy / pasting JS and wanted JS for everything. Hence node. And now electron.