Lojban has succeeded as a language that people use to communicate, but as a project, it fails to meet the requirements as it stands now. From a bird’s-eye view, this might not be so apparent, but then you realize that so many little things are left undecided, like {ce’u} filling, exact semantics of FIhO, some grammatical quirks like {.y bu} and {zo .y}, and the ever-present issue of overengineering…
You could say that this is not a problem for Lojban, and I’d agree: not for Lojban as a language (or languages, since Lojban dialects tend to solve these matters more elegantly and therefore amount to an advancement, not a hindrance). But the overall project’s criteria have not yet been met, and thus it is still ongoing, which raises the question of why one should care to believe in its completion provided that it’s over 30 years since it was founded. I, for one, prefer to stake my bets on a loglang project whose development history — track record — and leadership do not foretell its eventual abandonment.