As I said in the article, the lack of green threads, lack of value types, and lack of primitive boxing are all things that have gotten in the way of optimizing Bazel to the levels shown by the prototype Go reimplementation. These are all things that you'd expect a systems language to have, and Java did not have them. These are finally materializing now, but it's a little too late.

Then you also have the slow startup time of the JVM plus Bazel's initialization, which leads to a possibly-unnecessary client/server process design. This is more wishy-washy though, as this might be optimizable to levels where the startup time is unnoticeable -- but the fact that it hasn't happened in years means that it might not be an easy thing to do.