Why is log2NearDistance set to log2(nearDistance) instead of log2(nearDistance+1)?

Most of Cesium’s logarithmic-depth values seem to be computed as log2(linearDepth+1), but log2NearDistance is set to log2(frustum.near). Why isn’t it log2(frustum.near**+1**)?

I’m (very slowly) thinking through an issue with point primitives flickering/disappearing when the logarithmic depth buffer is enabled. The above question probably doesn’t have any direct bearing on that issue … but I’m trying to understand the relevant code, and log2NearDistance has me scratching my head.