RE: Is There a Fix For Hive's Downvote Problems?

[allentaylor]

I'm talking about perceptions. If you have a lot of former Hivers with the same complaint, there's probably something to it. It takes all of a few days to make a reputation of 30. That's not enough time to prove oneself a "model citizen". As people join a new platform, they need some time to learn about its culture. If their first few experiences are slaps in the face for violating some unspoken or difficult-to-find rules, then they're likely to leave and not come back.

Comments

[themarkymark]

Again, please show me rep 30's getting hit with $60 downvotes. All I am hearing is he said she said.

[allentaylor]

If that's what you're hearing, you aren't listening.

[themarkymark]

My bad, show me examples.

[allentaylor]

I'd love to. You tell me a tool that measures downvotes. I've looked and can't find one. Here's an interesting post from a month ago that attempts to answer the question. I disagree with @demotruk's conclusion, however. Here's why:

1. First, the data used for his analysis is aggregated data and not all granular enough for a proper analysis. He's looking at total blockchain activity versus the number of downvotes. While that data is interesting, it doesn't provide an up-close view of individual cases, which is what is necessary for this type of determination.
2. Under Point #2, he makes this disclosure:

HiveSQL does not have any table which tracks the full monetary value of a vote, be it an upvote or downvote.

Well, if there's no way to determine the full monetary value of an upvote or downvote, then the discussion we're having is moot. Neither you nor I can show any evidence to prove any claim we make regarding the question. I think you knew that already.

He goes on to say:

To get the full value, you have to recalculate the state of the chain, which is too complex for my purposes here.

Since this is a complex issue, it's necessary to get into that level of detail. Has anyone done that? If they have, I'd like to see it.

In his "hack," he defines Heavy Downvotes as:

I define Heavy Downvotes as those downvotes with weight of at least 50% by accounts whom are earning an average curation reward of 1000 Vests (typically about 0.5 Hive) 7 days later from their upvotes.

Again, while interesting, that's not really helpful. Hive Stats shows that your own current voting power at 30% is worth just $23 and some change. While that isn't a lot, it can devastate a new user who introduces himself and earns $25 on an introduction post only see that gone in a microsecond with a single vote. A person with thin skin, a low level of self-confidence, and an imperfect understanding of the blockchain incentives system may just feel discouraged and reticent to give it another go.

He concludes that there is no "statistical" relationship between heavy and whale downvotes to user retention, however, I would conclude the relationship is indeterminable. He goes on to say:

I define Heavy Downvotes as those downvotes with weight of at least 50% by accounts whom are earning an average curation reward of 1000 Vests (typically about 0.5 Hive) 7 days later from their upvotes.

And to that, I agree. Have you seen such an analysis. Are there blockchain tools available that would allow anyone an opportunity to make such an analysis? I'm asking, and if you're an honest man, answer me honestly. If I conduct such an analysis and it proves that downvotes do not lead to a user retention problem, I'll make the post public and allow everyone to see for themselves.

[themarkymark]

I've looked and can't find one. Here's an interesting post from a month ago that attempts to answer the question.

I didn't see any large downvotes, or any downvotes on that post.

Again, while interesting, that's not really helpful. Hive Stats shows that your own current voting power at 30% is worth just $23 and some change. While that isn't a lot, it can devastate a new user who introduces himself and earns $25 on an introduction post only see that gone in a microsecond with a single vote. A person with thin skin, a low level of self-confidence, and an imperfect understanding of the blockchain incentives system may just feel discouraged and reticent to give it another go.

Again, just more stories and could bes maybes. Not actual evidence, or facts.

Imagine if someone with a broom ran around super markets and slapped people in the face? Oh man, this could be devestating.

[allentaylor]

I didn't see any large downvotes, or any downvotes on that post.

See, this is a part of the problem. I didn't say there were any downvotes on the post to which I linked. In fact, the post is ABOUT downvotes and whether they are causing a user retention problem on Hive, the very conversation we're having now. If you had read the post, you'd have known that. If you had read the title of the post, you've had known that. Now we're talking about willful ignorance. And just in case there might be some confusion, I'll spell it out: I'm talking about YOU.

Again, just more stories and could bes maybes. Not actual evidence, or facts.

It's called logical reasoning. Analysis. That IS a form of evidence. Sure, it's 100% anecdotal, but anecdotal evidence is still evidence. Is it weak evidence? Yes, yes it is. But it's still evidence. And all you have to counter that small amount of weak evidence is some imaginary scenario that makes sense only in your head.

I've actually invited you, and now I'm imploring you, to point me to some tools that will allow me to gather evidence to prove you RIGHT. But instead of helping me help you be right about something that is obviously important to you, you lash out with non-sequiturs and fallacious misdirections. I want to be your friend, but you seem intent on making an enemy. If there's no evidence that downvotes has created a user-retention problem for Hive, a proper analysis should reveal that. Where is it?

[themarkymark]

See, this is a part of the problem. I didn't say there were any downvotes on the post to which I linked. In fact, the post is ABOUT downvotes and whether they are causing a user retention problem on Hive, the very conversation we're having now. If you had read the post, you'd have known that. If you had read the title of the post, you've had known that. Now we're talking about willful ignorance. And just in case there might be some confusion, I'll spell it out: I'm talking about YOU.

Again my point, just a bunch of people talking. No actual downvotes just would be could be.

I don’t need or want to read the post, I’ve seen enough people crying about downvotes that haven’t happened.

100% anecdotal

Exactly, just a bunch of people crying about things that haven’t happened.

I've actually invited you, and now I'm imploring you, to point me to some tools that will allow me to gather evidence to prove you RIGHT.

It’s a public blockchain, it’s all right here. There are no tools to providing downvotes. You can use any programming language to query the blockchain or even use SQL but unless you’re a developer you are not going to get far. I mean with all these people crying about downvotes, surely one of them must have seen them. In almost every instance the downvotes are justified. Not 100% of the time but it isn’t even remotely an epidemic, but what is crying about it without evidence.

If there's no evidence that downvotes has created a user-retention problem for Hive, a proper analysis should reveal that. Where is it?

You are the one making the argument. Surely you have evidence on your side rather than just repeating other people’s nonsense? You are asking someone who could care less about this crying wolf to prove a negative.

All I am asking for is one simple thing. Evidence. You are providing an argument but it’s all just based on hearsay.