A few weeks ago I posted to debian-devel some stats gathered from data
mining the "System Information" part reportbug adds to bug reports.
Thus, unlike popcon which is biased towards users who install the
straightforward way, these stats are biased towards contributors.

If you're interested in other bits, the thread is here:
https://lists.debian.org/debian-devel/2017/02/msg00551.html

What's relevant for DNG, is the systemd:sysvinit ratio. Obviously, as
wheezy and lower get replaced, it keeps getting worse, but the interesting
thing is unstable:

max=44%; X scale: week
⠀⣇⠀⢰⠀⡆⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⣿⠀⢸⢸⡇⠀⠀⠀⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⣿⢰⣾⣸⣷⠀⠀⠀⡇⠀⡆⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⠀⣿⢸⣿⣿⣿⠀⡆⠀⡇⠀⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⢠⣿⣿⣿⣿⣿⠀⡇⢀⡇⡀⡇⠀⠀⠀⠀⠀⠀⡄⠀⡇⠀⠀⢠⠀⠀⠀⠀⠀⠀⠀⡇⢸⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀
⣼⣿⣿⣿⣿⣿⠀⣿⣼⡇⣧⣧⡀⡀⠀⠀⡀⠀⡇⡆⣧⠀⡄⢸⠀⢸⠀⠀⢀⠀⠀⡇⢸⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⡆⠀⠀⡀⠀⠀⠀⢀⠀⠀⠀⠀
⣿⣿⣿⣿⣿⣿⣦⣿⣿⣿⣿⣿⣇⡇⡇⡄⣿⠀⣧⡇⣿⡇⡇⢸⡇⢸⡀⠀⣾⡆⠀⡇⢸⣠⢸⠀⠀⣤⠀⠀⠀⡀⠀⠀⡇⠀⠀⡇⠀⠀⡄⣾⡀⡄⠀⡆
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⣿⣷⣿⣤⣿⣧⣿⣷⡇⢸⣿⣼⡇⡀⣿⣿⣤⣷⣾⣿⢸⡇⡆⣿⣠⡇⡄⣿⡄⢠⡇⡆⠀⣧⣠⠀⡇⣿⡇⡇⣤⡇
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⣿⣿⣿⣿⣿⣿⣿⣿⡇⣿⣿⣿⣧⣷⣿⣿⣿⣿⣿⣿⣼⡇⣷⣿⣿⣿⣷⣿⣧⣼⣇⣧⡆⣿⣿⣴⡇⣿⡇⡇⣿⡇
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣇⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣇⣿⣿⣿⣧⣿⣇⣷⣿⡇
⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇
Dec 2014                     ↑                  Jan 2017
Avg for 2016+: 14%.

Looks pretty flat (save for noise) for the whole of 2016!

--
⢀⣴⠾⠻⢶⣦⠀ Meow!
⣾⠁⢠⠒⠀⣿⡁
⢿⡄⠘⠷⠚⠋⠀ Collisions shmolisions, let's see them find a collision or second
⠈⠳⣄⠀⠀⠀⠀ preimage for double rot13!