In my first post, i demonstrated that the downvote is necessary to efficiently make certain adjustments to the reward pool allocation. In my second post, I attempted to explain how Steemit voting is unique, in that we use it to determine, not to enforce, the rules of the platform.
Today, I would like to explain to you specifically the major effect of the cultural taboo on steemit against downvoting, and how i propose to combat it.
I suppose i should start with the cliff notes, so i don't end up putting my pitch at the end of a novel. I have started a curation trail here. The purpose of this curation trail is solely to downvote low content, automated, or statistical posts that are receiving high payouts but bring little subjective value to the platform.
and now for a science lesson:
Karenia Brevis
Karenia Brevis is is a type of microscopic marine red alge most common in costal waters near the US, canada, and South America. It is invisible and probably harmless to people and other marine life in low concentrations. It produces very small amounts of a poison called brevotoxin to defend itself against larger microscopic predators. K brevis feeds on fixed nitrogen converted from atmospheric N2 by the cyanobacteria trichodesmium (blue-green alge), and organic nitrogen from marine waste (fish poop and dead animals).
Sometimes, there will be an abundance of fixed nitrogen in some parts of the water. This can come from a Trichodesmium bloom, an abundance of dead animals or marine waste due to weather or man-made causes, or nitrate fertilizer runoff. The increase in available nutrients causes rapid growth of the k brevis population in the area. In addition to this, there can be weather or other meterological conditions that concentrate the normally relatively slow moving alge in to the area of abundant nutrients. This accelerates the natural growth of the local population even further.
Once the alge population reaches a certain critical mass, the brevotoxins that they produce begin killing off their natural competitors for food. This frees more food for them to consume, which allows them reproduce even faster. Additionally, their dead competitiors now become food when their decaying bodies release more fixed nitrogen into the water. The toxins also destroy the blue-green alge, which the k. brevis also consumes.
Eventually, the density of the K brevis population becomes so high that the brevotoxins begin killing off fish, and marine predators. As these animals decompose, they release more fixed nitrogen into the water, providing even more food for the K. Brevis bloom.
The dead fish and marine waste are eventually either consumed or moved out of the area of the bloom by tides and meteorological conditions. Without a local trichodesmium population to fix nitrogen from the atmosphere, the K. brevis, after killing off nearly everything else in the area of the bloom, dies.
What i have just described is called a red tide. It teaches us a principle that we can observe many places in nature: unconstrained growth is fundamentally destructive.
The creation of Steemit enterprises
There is a proccess that occurs on steemit which turns regular posters into enterprises. It goes something like this:
- Poster X is posting about Y. Maybe the post is objectively good, maybe bad.
- Poster X gets attention from a large whale
- The whale, brings the attention and support of other whales.
- The whale support brings in front running bots.
- The poster ends up at the top of the payout list no matter what he posts.
- Because the poster ends up at the top of the payout list no matter what he posts, he has incentive to post as much as possible.
- If his posting is actively horrible, he'll lose support from the whales. But inertia being what it is, posting that is merely lackluster usually won't lose him support.
- SO poster X has financial incentive to post as often as possible, provided he can keep the quality above terrible.
Because quantity is more important, and only a certain of quality must be maintained, poster X will strive create posts that are more general and formulaic.
Now poster X can be a person posting individually. I would argue that people like @piedpiper @stellabelle and @rok-sivante were all enterprises at one time. Poster X can also be something like @steemsports @steemvoter featured author, hidden gems, curie statistical posts, Project Cervantes. That is to say, concept posts that are not intended to be the writings of a specific author.
Steemit Posters reaching a certain critical mass tend to grow if everything stays the same
Take random concept-poster X. Set @smooth and @dantheman to auto vote everything he posts 30 minutes in. Have him post once a day for a few weeks.
His posts, regardless of quality, will grow in payment. This is due to a variety of factors, many of which were probably not considered in planning the voting system.
- front-running bots like wang and the recursives
- Cross-linked curation trails
- increased exposure on the "trending" page
- bandwagoning mini-dolphins and minnows who (incorrectly) believe voting for already-high-paying posts will bring them increased curation rewards, or who just want to be part of the percieved majority who like the post.
Enterprise Posters are not Bad. Unconstrained Growth is Bad
The posters and organizations mentioned above are not evil people, anymore than K.brevis is evil. But, given the opportunity to grow without constraint, they will grow without constraint until they are stopped, or they become large enough (and consume enough) to harm the system.
Contraints on Growth
Individual enterprise posters are constrained by the amount of time and effort that goes into composition. Because they cannot publish flatout laundry lists and expect to maintain their whale support. Since concept posters often make posts that require little or no compisiton, they are less constrained.
But with these types of posters, there is still at least a potential constraint to growth -- they have to maintain the support of the whales that backed them initially (or ones that came on board later). So they can grow, but only until the whales start to say "enough". However, these constraints are wholly theoretical. It just all depends on how much the whales backing the enterprise are willing to tolerate.
Until now, these enterprise posts have grown to the point where they created a user backlash, and that user backlash ended up costing them their whale support. However, this proccess was messy, and often engendered unnecessary bad feelings.
There is a third type of enterprise poster, however, that is not constrained in this way. It is an enterprise where the gimmick is buying votes. One such enterprise is steemvoter. This business felt completely entitled to upvote an empty post to over $90 with its users votes, because it had established its right to do so in its TOS. This was allegedly as a test, but the votes were never removed, and the comment would have paid out had @abit not flagged it (a flag steemvoter attributed to a conspiracy against him within steemit of korean gangs).
There is really no constraint on how much steemvoter, or a similar service can grow and take from the pool. The growth of enterprise posters backed by whales is troubling. The growth of enterprise posters backed by vote buying is even more troubling.
other constraints, designed into the steem curation system, are impaired by the downvoting taboo and changes to curation rewards
- Decreased curation rewards for bandwagoning -- curation rewards are already so low, that they cannot provide incentive one way or the other. Dantheman made 400 steem last week in curation rewards. Smooth like 151. The best performer was berniesanders, who got a return of 2000 steem on a 2M+ SP balance. And its very likely that much of that was due to his participation in curie, and being able to vote first there (my guess). Curation rewards that cannot provide incentive for whales cannot have a significant effect of reward distribution.
- Users, including whales, downvoting bad, overvalued content. This is proposed as the main countermeasure against bad content somehow supported on the n^2 curve. However, Even major stakeholders\owners like dantheman and ned seem hesitant to downvote material they feel is overvalued. The ned/master yoda thing and the dan/steemsports thing are a great example of this. If the taboo is so strong that even the people who run the place are hesitant to use the downvote feature, it seems unlikely that other curators will.
Intimidation and prior restraint
Not all people are ethical people. Therefore, it stands to reason that some percentage of enterprise posters will also be unethical. I do not know the people behind steemvoter and steemsports, so I cannot speak with certainty to their ethical fiber, but their recent actions give me pause.
An enterprise like this that is able to increase its marketshare by making changes to its tos, then bullying anyone into silence who voices an objection is dangerous.
@steemvoter recently had this to say to a user who questioned whether or not the money going to his company was a responsible use of the reward fund:
Well take it elsewhere. If not we will ask our community if we can use their votes to down vote the hell our of annoying whiners.
(note -- I am not entirely sure that their TOS does, in fact, require them to ask before using the votes they have been given access to to downvote their critics... i suppose if it did, they could always amend it though, and make not agreeing to the amendment opt-out)
Also here
the steemvoter account flags a user for downvoting his payout post, and makes sure to announce it. The purpose of this is clear -- to prevent others from also adjusting the payout downward by threatening reprisals.
After making these threats, he recently posted this:
We thus propose a security service where we defend the community from potential attackers, firstly by providing passive defence strategies and early warning systems that could be enabled in a user front-end like Steemvoter, with defensive comments posted at intervals, and increased as multiple auto-flags are detected to neutralise the attacker like we did.
There's a name for that. Its called a protection racket.
Thank you so much, we are glad you are like-minded on this, a flag police is what is needed and we are prepared to be just that, we are tired of it as well and like you, believe it will kill the community.
Now, he proposes to use his purchased votes to tell others how they are allowed to vote.
The user @walden in rocketchat, apparently a representative of @steemsports, has made similar threats. He monitors the steemitabuse channel and attacks people who are critical or unsupportive of SS's level of funding. In the following image, posted by noisy, he indiactes that he will use the SP that all this reward money has gotten him (as well as his completely unassailable reputation) to enact reprisals.
In a system where it is already quite difficult to adjust enterprise support downward, a force like that attempting exert influence on voters and perhaps even critics in the name of expanding their portion of the reward pool could be a dangerous actor.
So here's my pitch
The two main effects of the downvote taboo with regard to enterprise posts are:
- There is a hesitation to downvote over-rewarded enterprise posts that comes from an unwillingness to be the "first in" and thus being forced to bear the brunt of reprisals alone.
- There is also a hesitation that comes from the potential of being accused of downvoting because you are an asshole, jealous, have personal motivations, or for some other inappropriate reason.
So my pitch is--let me be the jelly asshole. If @walden wants to do a press release, let him do it about me. If someone wants to accuse you of being attacking them for personal reasons , tell them sigmajin attacked them, you're just on his trail.
No, i can't hide your downvote. And I can't promise that you will never suffer a reprisal. But i can say that i will use the trail to respond to reprisals with upvotes (this is the only thing i would ever consider upvoting with it, and only then to just above zero), and that i will also use my own accounts, including the 80K steem power I have in small accounts, to attempt to cancel out reprisals should they occur.
I have started a curation trail here the sole purpose of this trail is to serve as a check on the growth of enterprise-poster rewards beyond what is healthy for the system. Should it become necessary, the trail will also respond to help users being intimidated or attacked speaking critically about these enterprise concerns. This trail will use no more than 7 full power vote equivalents a day (so like 14 50% or 7 full power or whatever). If this is too much, i believe you can set your strength using the slider.
Later today, in the comments to this thread, i will release my specific criteria (no votes will be cast until 24 hours after the criteria are released), but in general, i will be concentrating on low content\no content posts that have a payout in the top 25. I do not ever intend to flag a post or comment below zero, and i do intend to give special consideration to new posters.
Can this make a difference?
I believe that it can if it gets even a moderate amount of support from medium/large dolphins, many of whom have expressed concern about these types of accounts.