i suggest you, to read something about the "ninja mining" of steem.
https://bitcointalk.org/index.php?topic=1410943.0
https://bitcointalk.org/index.php?topic=1410943.60
https://bitcointalk.org/index.php?topic=1410943.240
if you use the stake as only parameter to measure vote weight, the ("initial") distribution has to be faire
perhaps mind changing for you:
As illustrated by the results, the distribution of weight of votes in witness election is heavily skewed, which
suggests that the election of 21 witnesses may be significantly impacted by a few big shareholders, This phenomenon may not be a
good indication for a decentralized social media platform.
source: https://arxiv.org/pdf/1904.07310.pdf
for me this was kind of mind changing. i dont like dpos anymore
I don't need to read on this, and it won't change my mind, as I was there nearly from the start ;) Which is a proof that it was not as "ninja" as some people like to make it seem. Yes, there were only a few people interested in getting into it, and I do agree with the distribution being one of the big problems we had. It's been improved on a lot though, there are many big holders who got in later.
It's still hard to get into/stay in the top witnesses without massive whale support, and especially one miner account has an unproportionately high say in who's up there. It needs more time and more people interested in investing significantly to fix these issues.