"Generating every name consumes about 16 random bytes. This can exhaust the system entropy and slowdown other applications."

Crys and Alex_Gaynor confirmed me on IRC that these two assumptions are both wrong.

See for example https://www.2uo.de/myths-about-urandom/

Q: But that's good! /dev/random gives out exactly as much randomness as it has entropy in its pool. /dev/urandom will give you insecure random numbers, even though it has long run out of entropy.

A:  Fact: No. Even disregarding issues like availability and subsequent manipulation by users, the issue of entropy “running low” is a straw man. About 256 bits of entropy are enough to get computationally secure numbers for a long, long time. 

--

About performance, well, it's not exactly "wrong" but "inaccurate". Abusing /dev/urandom only hurt other applications which also abuse /dev/urandom. Such use case is very unlikely.

* The bad performance of concurrent /dev/urandom reader was analyzed by an old article of 2014, but see comments:
  http://drsnyder.us/2014/04/16/linux-dev-urandom-and-concurrency.html
* The performance issue was fixed in Linux 4.8, https://github.com/torvalds/linux/commit/1e7f583af67be4ff091d0aeb863c649efd7a9112