"With the Linux 3.3 release, it became possible to boot Android on a vanilla kernel."
Does "boot Android" simply mean, "run a Dalvik-like JVM"? I'm sure I'm not alone in wondering, "How many of the Android features in staging will be usable with glibc? Or must bionic features be merged into glibc as well?" Undoubtedly many of the merged Android features are valuable in their own right, especially for embedded use cases, and there's merit in merging the changes just to reduce the size of the mainline-Android kernel fork. Nonetheless, I wonder if the effort necessary to permit Android apps to run natively on Linux is an important goal given how well lxc works as a host.