Witnesses, please make voting simpler, not more complex. Don't approve HF 20 without some changes.

Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes. - Charles Goodhart

Apologies for using the witness category tag when I am not a witness.

I am going to post this because on cursory review of the changes they want to implement I just had to give my own opinion. Now I know that it's rather routine in this community to just take what Steemit Inc. hands the witnesses and push the update, but let's think about this for a second.

If there isn't witness push-back now, then I doubt there will be any when the patch drops. I know some of these won't be changed, but these are just my opinions and I'd like to spark a debate around this, and you can come at me bro if you think you have good counter-arguments.

My list of code-related suggestions.

• Keep n=1. Linear is good. You're just greedy for thinking otherwise.

• Minute 0 voting should be allowed. Arbitrarily soft-preventing someone from voting on a new post or comment they enjoy is terrible, period.

• Curation rewards should always be a fixed percentage and independent of the time the vote is cast.

• Curation rewards should be higher, possibly around 50% or so. Don't encourage or force the largest vested accounts to also spam us with posts. Let them simply vote to generate ROI, to punish human curation less as opposed to self-voting or vote-selling. In fact you could posit that pushing curation up as high as 99% would theoretically price bid-bots out of the market completely and create an environment where monetary rewards for authors are simply small tips instead of the bulk of the rewards pool given out, but this would lock-in the inertia of current voting power rather tremendously. Personally I do see other advantages to this approach, especially as far as making potential post rewards a secondary, milder effect of posting as opposed to being a primary focus of posters.

• Curation rewards should have identical payout options as author rewards. Do not advantage one over the other with SBD/SP/STEEM payout options unavailable to the other. This is a no-brainer.

• No increase or decrease in curation rewards based on the voting order. Coming late to the party should not be so terrible for voters if they want to reward an older post.

• The new sub-15 minute curation pool is garbage. Remove the option to auto-vote on post creation if you're really going to do this, as it's disingenuous to leave it as-is without a huge blinking red disclaimer attached.

Adding burden of knowledge to use a platform is anti-user and anti-investor.

You might be someone who enjoys reading changelogs all day, you might have once upon a time come to Steemit and read for hours a day to get up to speed on how voting mechanics work, you might have done the math yourself on how to maximize curation rewards or adjust your auto-voters to get the most curation back for your SP.

But ask yourself this, nerd, does the average individual (no, not the person aiming to be an influencer or an account-seller on certain sites, an actual average, normal person) have to do any of that to understand what a vote or a like or an upvote does on ANY OTHER SOCIAL MEDIA?

So why the hell should they have to do that here?

Furthermore, what makes you think having elaborate hoops of game-theory that you have to jump through in order to maximize ROI as an investor make the platform attractive to anyone but the nerdiest of nerds? Investors want no-bullshit, hands-off ROI. Sure, maybe there could potentially be something to be said for forcing them to learn at least the basics of the platform in order to profit from it in order to dissuade some from investing, but without strong hands the platform is dead as no one will be there to catch the inflation year after year.

Literal anti-social mechanics built-in to the platform are not good for the platform.

Why does this even need to be said? Do you even want people to use steemit or other dapps, or do you just want to give off the appearance that lots of people use it so you can take a nice healthy dump on new batches of imbeciles whenever you want to sell?

Okay, so I get that Steemit Inc. wants this all to be some freemium bullshit phone game where you have to pay a certain amount to get a vote slider, pay a certain amount to get bandwidth on heavy use days, pay a certain amount to get exposure, and so on. That was the site's intention from its inception, and through that and a million other screw-ups we have the tiny userbase that we have now.

The early distribution process was awful, and so were tons of other things that happened, which is why our current landscape resembles Clash of Clans more than it does Reddit or any other social media it's trying to ape. Except at least in Clash of Clans the devs didn't just gift a few accounts with an insurmountable amount of troops and power that would be carried in perpetuity.

Voting should be simple, it should be a way to reward content the user enjoys. They should not be punished for failing to take on the extra computational load that the code forces onto them before making what should be a simple decision.

"Do I enjoy this content or not?" should be the only thing someone should have to ask themselves while reading a post. Analyzing the content should be the first and foremost burden placed on the user, not analyzing the timing of the vote and potential hindrances to curation, who has or hasn't voted yet, the potential follow-on votes, the user's popularity and average past voters and vote times, and so on.

A user should know beforehand based on their own vote value what their curation amount should be, and how much the author should get. HF 25 better not force us to all brush up on our multi-variable calculus to get the answer.

TL;DR: Want to keep Steemit anti-user? Then push the patch as-is.

The voting system is bad now. The voting system will still be bad after HF 20.

Reducing complexity means just that. It doesn't mean increase the complexity, it doesn't mean change it around so it's about as complex as before, it means lowering it. Making it smaller. Decreasing it.

Requiring a delay on voting to prevent bot votes arguably only matters because you have the already stupid decision to increase curation rewards the earlier you vote on something and based on the size of other votes that come after you. It makes the entire thing a game of finding authors with average post rewards where you can sneak your vote in before other people.

I'm not going to get into why having a forward-facing system that entangles the concepts of "popularity" with money is a terrible idea for social media in the first place, but if we're going to keep it we might as well make the process a bit less onerous to use.

Steemit is already the land of the bots, vote bots, comment bots, me. Make Steemit human again by appealing to humans and how they naturally want to interact. Thank you for your time.