Next time you want to do an experiment...

in #experiment7 years ago


You need to decide on methodology. Everyone needs to follow it. If you cannot cancel out 1:1 with your flags to offset the people not participating then the data is going to be skewed. Also having to flag has unintended impact on the platform/community in other ways that will alter behavior, and thus corrupt the results.

If you want to do a test to see what it would be like with no whales voting. Get Dan/Ned involved and have them remove the ability for people over a certain steem power to vote (up OR down). Agree to a time period for the test, and apply it.

Then there is no guesswork. You would only be measuring the impact of no whale votes.

If you cannot guarantee a controlled environment then your results are going to be skewed and people's interpretation of that data will be opinion and speculation.

The platform is in beta as I understand it though I saw another person refer to it as being in alpha. We can do tests, but PLEASE communicate with the community about it. Don't just start doing it, and don't do it in a completely disorganized manner.

Right now with this short time for the experiment we can tell there is definitely something that would likely happen with this experiment. However, the results are so muddied by people not participating, and thus being countered by whales who are participating by them DOWN VOTING/Flagging those posts. Flagging is not psychologically the inverse of an up vote. That should be obvious by the reactions it gets. As soon as you had to start doing that you might as well have called off the experiment as the data collected will be meaningless and based purely on opinion and speculation. There are now too many different variables besides simply "what if no whales voted".

I'm all for experiments. Yet you need to state methodology and get EVERYONE that is part of the group you are testing to agree to them. If you can't do that then you need Dan/Ned to enforce it via code for a period of time.

If everyone is not participating your attempts to correct it just further corrupt the results.

At the moment the experiment may actually be harming the platform. That isn't to say a controlled experiment would harm the platform. This is far from controlled so we cannot know.

I will be limiting my effort on my posts until this "Experiment" is concluded. I don't mind the no whale votes, and I like experimenting. It is not about payout. I do not want to put forth effort to have it flagged at a higher weight than any whales that may have up voted it, or worse have it flagged when a whale didn't even up vote it.

Prior to today I'd really only ever had a post flagged by the @asshole bot. So if you find a way to do an actual controlled experiment you have my support. Until then I'll likely refrain from serious effort on content.

I do write when I am inspired so it is also possible I'll write something anyway because it is like a living thing that if I do not breathe life to it by writing about it, it'll drive me insane, or I'll go insane thinking I might forget about it if I don't write about it.

I also converted some steem to SBD yesterday so I'd have some SBD to promote posts. I will not be promoting posts while this flagging is occurring.


Steem On!




Sort:  

LOL.

TL;DR - version pretty close.

Only other valuable things is I did suggest how they could do a CONTROLLED EXPERIMENT in the future that would actually not allow the bullying. :)

Precisely. That would be actual science.

I also think it IS an experiment worth doing. As long as it is controlled.

@dwinblood I agree with this ..

Get Dan/Ned involved

and the community please... do ask if we're also okay with it
and how long will this go on
oh btw.. I like the fact that my voting influence is really doing something so please make it last long - like a year or two :D

as for the flag - what ? there's a flag involved?
oh that flag I hate it

Dan saw my post and up voted it and thus likely thinks if they want to do an experiment they should talk to them and he'll work with them.

@dwinblood they're the founder they shouldn't be left out just my IMHO

Correct, and Dan and Ned would not be able to vote during such an experiment either as they are whales. I think they'd both be fine with this idea. It'd need to be a short period (no longer than a week or two due to 7 day payout change coming) but if it was enforced across the board any data collected should at least be useful.

@dwinblood hmmm.. what's the point of doing it if its just for a short time then
with all the flagging, too sighs
oh well.. let's see how this turns out .. hopefully for the better?

EDIT : nested comment - I see..

The data is useful for understanding trends while this is in BETA so we can come up with fixes.

It is not about making people happy now that they are getting paid more, before we've fixed it.

With useful data we can try to come up with fixes that don't screw anyone over... including those with power.

that clarifies it .. thank you for explaining

The short time is to find out the effect. This is not about socialism or communism. We don't want to make it so people once they have reached a certain power are no longer able to vote. Not without making some other incentive. They'll simply game the system if we do that and make multiple accounts and make sure they never reach that limit.

The data is useful for understanding trends while this is in BETA so we can come up with fixes.

It is not about making people happy now that they are getting paid more, before we've fixed it.

With useful data we can try to come up with fixes that don't screw anyone over... including those with power.

I wrote a community encouragement post today, got 5 or more flags, killed the rep and payout, and I had a lot of time and thought into it but who cares, about encouraging content right that is then worth $1?

It was a failed experiment and these flags, with no start or end date, or announcement, are driving good people off here daily.

https://steemit.com/steemit/@barrydutton/steemit-people-i-m-thankful-for-mindful-community-engagement-encouragement-e-1-march-12-2017

Did it actually ding your reputation? Usually unless they are higher reputation than you it won't do much and your reputation is high enough that it'd take more than a single post to even drop it by 1 point I'd think.

I could be wrong.

Don't be fooled by the half information we read on here about all kinds of stuff like that. Anyone with THAT much power over anything in life, and in this case that much SP wields power across the board on things.

It is hard to read in my screenshot, see for yourself, look at the 5 grey areas.

https://steemdb.com/steemit/@barrydutton/steemit-people-i-m-thankful-for-mindful-community-engagement-encouragement-e-1-march-12-2017/votes

I am for testing rather than guess implementation as I posted a bit earlier. This is far from good testing methodology and has created influencing factors that were not considered. I am not sure if anything useful will come from it. On the upside though, perhaps a failure or three will be the impetus needed to create a clear and sound testing approach for future experiments. A starting point would be communication.

Silver lining (aka your upside) is that maybe they'll learn more about how NOT to do an experiment for future efforts.

Maybe they should call in the freakonomics team.

Or just talk to people like @lemouth in our community that are actual scientists. He works on the Large Hadron Collider.

I mean I know the scientific method quite well too and could have explained how it works, but if they want someone who might be better for an appeal to authority then @lemouth is a good example. Great guy too.

Only the large one? Pffff

;)

I can make use of more superlatives if you want ;)

A downvote was applied to partially counter earlier whale votes as an experiment to reduce whale domination of voting influence. Not intended to express an opinion on the content nor result in a net reduction of rewards or reputation (automated notice)

I am afraid that much as I understand and support the cause you are crusading - it is having a negative effect unless someone can explain why we cannot see the purpose of the experiment. The elephant in the room is being ignored - flicking its ear is not going to make a meaningful shift to the structure of steemit, I am afraid.
The longer the elephant is ignored, the greater the pain will be.

there is no censorship on the blockchain..

yeah...but..

Well this is a common argument. I agree with you. The argument "It is still on the blockchain" is for some people the same as saying "It is in a file cabinet at the Pentagon". "But they censored it...", "no they didn't you can file a FOIA request".

"Why did my favorite magazine stop publishing?"

"They had people yank their funding and couldn't afford to put that much effort to it any longer"

"Isn't that censorship?"

"No, just pulling their funding. They still could have done it for free."

As long as there are multiple accessible clients controlled by different people, the argument holds. Currently we have steemit.com, busy.org, steemdb.com, steemd.com as well as clients you can run directly on your machine. It's often literally just a case of changing two letters in the url. For some that may be too much, but we should hopefully get even better than that in the future.

Yes, I get this. Yet that is like a FOIA request for some people.

Different people have different tech levels. Just because it isn't censored to you or I, doesn't mean it won't be censored for someone who cannot do those things.

Furthermore, if I don't know it is there, will I bother to go look for it?

True, but it's also as good as we can do on a technical level. Removing censorship entirely is technically impossible, unless we remove all ability to filter out noise.

Furthermore, if I don't know it is there, will I bother to go look for it?

This is a cultural thing we need to encourage. Just like in Bitcoin we tell everyone over and over "if you don't control the keys it's not your Bitcoin", people should be encouraged to use interfaces which they're genuinely in control of. Whether that's entirely desktop-based stuff for the really hardcore anti-censorship user, or browser extensions which highlight potential censorship while using a website, it's a matter of getting people to acknowledge the possibility of censorship and use the tools that keep themselves in control.

I promote Steemit as a censorship-resistant platform on some of the subreddits which are interested in that. I always say "censorship resistant" not "censorship free", because all we can ultimately do is give people the tools to resist censorship. The expectation of "totally no censorship" is impossible and not even desirable (in the sense that there is value to filtering out noise that you don't want to see, and sometimes you trust others to determine what is noise, but it should be your decision how you go about that).

True. My point mainly is that CENSORSHIP has an obvious true technical thing.

Yet barriers can also have the same effect as censorship whether it is no funding, or extra steps to get at it.

So while technically there is no ACTUAL censorship as it is still on the blockchain. There are barriers which I think I've heard other sites/articles not on steemit refer to as Soft Censorship. I don't know that I really like that wording/phrase, but I share it here only so you might see what I am getting at.

I realize there is no actual censorship since it is available on the blockchain and it isn't as if the book was burned.

It was simply moved to a corner room on the third floor down a little used hallway. :) (yes, it may not be this bad... but I was enjoying the imagery)

So people can hide behind "it was not censored" not just here but, many places. When depending upon the barriers involved the end result can be the same as if it were censored.

If no one wants to put forth the effort now required to see it, then it can become as if it was not written. It is there, but it may not be witnessed.

Likewise the book may not be burned, but people voting for a person to write a book, and then having one or two other people come along and say "don't write it" and thus have no money to write it could actually be WORSE than burning the book.

Why?

A Burned book at least existed for awhile. Someone may have read it. It was a creation.

If funding is attacked for subjective reasons cancelling out the funding other people would give it then that can kill future books before they are even written.

So is it technically censorship? No. Yet it can have the same end result, and in some cases may be worse.

I actually think that is technically censorship. Not just soft censorship, but censorship. Like I said, there's no such thing as a censorship-free platform.

The best we can do is:

  1. Ensure the platform is censorship resistant (public blockchain where all data can be accessed)
  2. Teach people to use clients which they control, so that they can actively resist censorship
  3. Encourage an anti-censorship culture, where attempts to censor and justifications for them are frowned upon and seen as shameful.

I think we are making good progress on all three of these on Steem.

I think I can agree with that. This is also why we talk about it. We can't fix it by ignoring it.

Yeah... but I experienced it before the experiment and if this little minnow who barely gets enough rewards to reach a dollar can get censored . I know others can/have too. Yet, the community hardly heard our cries or kick up any dust for the bad actions of the same bad whales controling steemit and the rewards pool.
Since this experiment, now it's trending because whales are effected and the people they support? Manybe this is how minnows who have been crushed by bad whales can get a voice. Go figure.

I believe we are stuck in a loop.post-22106-Cup-and-Ball-perfect-loop-gif-pOdK.gif

Well said, even over the past few hours @abit has been sending steem to other whales to try to get them to stop bots and vote with low weight... this should have been done days ago, in my opinion.

Suppose we mute everyone who downvotes?

Some of them don't say much. They just down vote. So yeah it is an option, kind of like shunning. It may need to happen.

Yet really since it is in beta I'd like us to learn from things and try to find positive solutions before we try to open the floodgates.

I've always said let us mute whale rewards if we so please.

If I get flagged I mute the account that did so.

Unfortunately the high powered flaggers earn a lot on curation, so they likely are unphased by that. They are not posting things that you are not seeing and thus losing out on your audience. If they were then that would be an effective tactic.

I'm doing it for me...not for them. If I don't see what they post then I won't feel the urge to comment on it. Let them talk to themselves.

Bypassing other comments, for the moment... I agree that there's a lack of "scientific methodology" involved here, which one would typically hope for in an experiment.

For me, it raises a question mark on the whole "decentralized" vs. "community building" aspect of Steemit. It's a great concept, but let's not completely throw out the baby (centralized organization) with the bathwater. I've only been here six weeks, but a lot of stuff around here seem like unorganized chaos, with features and tweaks coming in chaotic fits and starts. Methink there might (at least) be some kind of "steering committee" behind moving the community forward.

And if there actually IS one, they are doing a piss-poor job of communicating with the community.

Not sure where the steering committee would be. I'd expect it would be steemit inc employees, and the top 19 witnesses. This is why voting on witnesses you believe are representing how you want steem/it approached is important.

If anyone has input it is those people.

Then there are of course the whales who through their sheer accumulated steem power and the n^2 curve can completely dominate steemit unless other whales WANT to challenge them.

One thing about this down vote I've been trying to get to others I explained in a reply on another post... here are my examples I gave @merej99 ...

1% is not the same unless the steem power of the whales is identical.
@wang is 1,540Million Vests
@smooth is 5,052Million vests...
In other words Wang's 1% vote was worth 15.4Million Vests applied towards the post. Smooths is 50.52 Million vests. 15.4 - 50.52 = -35.12 Million Vests
In reality @smooth's downvote at 1% was worth 3.28 times as much as @wang.
So @wang voted and then it is as if more than three @wang equivalents came and flagged the post.
Not equal at all and very much screwing up any legit data. I made another post stating how we could do a controlled experiment. That is not what this is.


In reality it is WORSE than my math showed it. That CURVE n^2 people have been referring to makes it so there is not a direct correlation in steem power like my math showed. The difference between @wang 1% and @smooth 1% would likely be even greater than the math I showed you. I told you that @wang voted and then @smooth countered with a vote more than 3 times as powerful as @wang's. It may have been actually worse than that due to the curve.

This is exactly what happened to me. I managed to pick up 6 downvotes. According to steemd.com, the downvotes were as follows:

abit -18,019,851,490,916
smooth -1,010,494,623,757
engagement -422,414,143,516
berniesanders -7,379,281,241
thecyclist -245,377,738,122
nextgencrypto -102,619,770,165

but my top six upvotes were
hendrikdegrote 8,219,277,610,560
jaewoocho 5,372,742,249,385
lafonaminer 2,650,557,916,332
wang 2,310,680,937,121
delegate.lafona 1,316,562,433,153
fyrstikken 105,068,108,918

They weren't just cancelling the whale votes - they were cancelling all the dolphin and minnow votes as well.

Which makes a complete mockery of "lets see what happens when just dolphins and minnows vote"

Also - this thing affects curation rewards for the little person. Even if you were a minnow, it was worth voting manually to see if you could spot a great post at 30 minutes, that would later be picked up by a whale. There is no point bothering now, because each post only seems to earn a dollar, so if you are a really small account, you won't get a curation reward for it anyway. They've broken the entire system.

I expect this experiment to STOP soon. I'm actually trying to have some productive talks with people to see if I can help someway. I am not a powerful steem power person, but I care about this project and the community. Not sure if I can help, but if I can figure out a way that I can, I will.

Abit insists on continuing - and stated that it would be "good" if people start leaving - and smooth apparently agrees with continuing as well. I think other whales should simply stop complying with the requests. This is both a mockery and a failure. When something like this is haphazardly sprung on the community with no defined goals or parameters, and then a large portion of the community says, "this isn't good," then it's time to stop, evaluate where you went wrong, and work to improve the "testing" for the next attempt, if there is one. You don't double down on the stupidity and extend the time frame.

Coin Marketplace

STEEM 0.20
TRX 0.13
JST 0.029
BTC 66649.04
ETH 3352.77
USDT 1.00
SBD 2.70