Introducing @calibrae, @elfspice/@l0k1's new and final Steem account.
Today
has been a most peculiar day. I did my first voice interview on the Whaleshares chat with @patelincho, to answer questions for the people there who are terribly interested in this new project.
You all know me, of course, I originally signed up as @l0k1, and then more recently, started going by my old psychedelic forums nickname @elfspice.
Today @noganoo gave me this account, which apparently is a terrible thing, but I have no idea about any of that stuff. It just seemed natural that I should switch to this full time.
This is not going to be a 'corporate' account, per se, that is, I am not suddenly going to start talking like a faceless suit-wearing robot representing someone else. That's just not my style. This is my Post-Steem username.
I know Steem is still here, but I doubt that is going to be for that much longer, since I am utterly unconvinced that they have any strategy for limiting the toxic load of bots, in fact, by the little bit of chatter I have seen about @noganoo, apparently this is the end of the wereld.
Anyhoo, so, I just want to report a few new things in the Calibrae ruleset:
- Muting will have the effect of placing a downvote automatically on every post or vote by the muted account. If an account is muted by enough users, they will have no vote power. It will work by reducing their vote power, while the mute is operational.
- Following will have the opposite effect. There has to be an opposing force to this, or it would be abused.
- Voting will not decay, but instead be 100%, but the number of votes available will be limited. In fact, the combination of stake and coefficient, and follows and mutes will generate an overall post limit. Instead of using a bandwidth scheme, the number that comes out of this formula will limit limits transactions.
This way, the ability to place new data on the database is limited. The interface will show a number that indicates how many remaining transactions you can make, which can be any one of post, vote, follow, mute, transfer, or power up/power down.
There is a serious problem with the way the bandwidth system works for new accounts - they can't see how much transactions they can make. But votes are unlimited. This means a new account, that posts nothing, can be used to spam the database with transactions. This makes no sense. Limiting the power makes no real difference. Not only that, not only can a new account post unlimited vote transactions, it can post unlimited transfers, unlimited posts. Posts, particularly, weigh heavy on the data store. These should also be limited in size, like how SMS comes in blocks. If your post exceeds the number of transactions, measured by these blocks, it will tell you you can't write any more in the post.
These are just thoughts coming off the top of my head for this afternoon. I will refine them more as I work through it.
Nice ideas, see my post for more: https://steemit.com/steemit/@emble/will-steem-succeed-or-commit-suicide
I have no interest in the subject anymore, my opinions have been made amply clear to everyone. I am focusing on the new. "Let the dead bury the dead" :)
well of course I am following you ;-)
Hey there i love a lot of the changes you plan to do , good lucky !
interesting experiment! It would be great to see different variations on the Steemit implementation.
I had a blast the past few months posting but I am getting concerned frankly. The bot issues are to be expected, but the lack of constructive communication from the powers that be and action to deal with the bot abuses is not encouraging...
Also the level of interaction seems to have changed, there are less and less quality comments, which I believe should be considered the lifeblood of this thing, so that is another red flag to me.
It seems way to vulnerable to bots and manipulation. Especially the way posts are "valued" through upvotes does not seem to reflect a signal that humans valued the content.
Posts with no views & no comments and 100's of automated upvotes, should not be able to claim 100 's of SBD since they do not grow the value of the platform .
Do you think your implementation should be completely bot -proof? (if that is possible at all? )
In any case I do hope your platform and fork takes off . If anything competition is healthy and should keep everyone on their toes and try hard. I'd be happy to sign up for the beta and give it a try, where can I sign up ?
It will be bot-proof, but not through automation, but via enabling humans to 'curate' user accounts via the various mechanisms.
All accounts on steemit will be migrated, every last one, as current at the day prior to launch. They will all be reset to default starting values, and then everyone can have at it. Basically, you will just take your username and password from here, and it will unlock your new, clean, baseline level beginner account.
The initial users will even have the power to blot out all the bots and premine whales and trolls right from the first day, by muting them en masse, well, as many as they have the power to make transactions for (I'm thinking to make the initial account start with granting 12 transactions per day). Even if they show up, they will be stuck with one transaction a day, if enough people mute them. Then, if they want to stay, they can work their way up slowly by being good.
I like it :) New game, same player shoots again :)
Yes a limited # of transactions should favor humans over automated behaviour.
Do you think comments should be more recognised as a signal of a "quality" post worthy of a part of the reward pool ?
Reading your comment again it made me think about how this method could be used to create markets that can't be manipulated by HFT bots.
It requires a reputation system to stop the bots!
Steemit's model is based on financial markets, which deprecate the influence of social factors. But they put a forum in, which means eventually the social will have to rise to equal the power of stake... Or someone will fork it so it does.
Yes I guess in the end bots can be programmed to replicate any human behaviour.
I naively thought that you could score a post as human interaction through counting the comments, subcomments and upvoted comments and the number of unique commenters.
Assuming flame wars would not be voted or flagged down.
But I guess once the bot programmers figure that out and they have access to 1000's of bots that could be defeated as well... sigh....
Yup. It's a win for the community, and a lose for antisocial moneygrubbers :)
I am modelling this after polycentric legal systems like the ancient jewish and middle ages icelanders. Judges are subject to judgement by their peers. Then they can't become tyrants.
I am sure that it won't take long for the power of the system of security by human driven judgement actions to send a clear message to the scammer/bot/asshole community, that the whole platform can turn on them in a matter of days and render all their efforts fruitless, and even lock up their deposits for days. I am thinking also to add a transfer limitation when thte account hits zero reputation, so the asshole has to choose between trickling out their stake, or making good so they can get more of it at once. Any liquid assets they have, they have the choice between trickling it out, or setting a power down on their stake... you see what i mean, it makes the dilemma on the miscreants, and clarity for the good people.
I am working on the right formulas for mute/follow effects on holding or releasing Stake power via modulating effective reputation score, to avoid flamewars, some of it is going to be hard to model because of its complexity, but it's complexity is due to concurrency rather than complex sequencing.
It should not be sequence dependent so as individual accounts change the parameters only one operation is required per change. Virtual operations I guess they are called, storing a transient state derived from deterministic data coming from the database state.
I don't see how there can be any patterning discovered in comments that can positively determine humans. The best turing test is other humans, and the best turing test of all, is many other humans. That's why this model puts the tools to submit these judgements in the hands of the users rather than trying to guess it with heuristics.
Well, I'm not sure about that, there isn't really an opposite to commenting, unlike follow/mute up/down vote. Whether a comment is supportive or not isn't what comments are about. In a debate the merit, or lack thereof, can only be judged by humans. I'll be interested to hear proposals to this effect, but bots can comment the shit out of a post, and it means nothing.
This is quite salient. Furthermore, it is a feature, designed into the platform. As the possible reasons for including this feature in the design aren't elucidated in the white paper, which instead states that such manipulations are an existential threat to the platform, we are left with examination of the effects of the feature on the platform, and accounts, to ascertain the reason for it.
In any case cryptic reasons are usually malign, just as the incessant examples provided by candidates for political office show.
I do look forward to finding you on Calibrae, along with other humans capable of reason.
Perhaps, in time, the apparently injurious flagging of @calibrae will prove to be a good thing, by weeding out those poseurs that fail to pursue it because they can't be assed to penetrate the fog of flaggotry that protects it from such, like a moat and bailey.
Keep us up to date! We are on the edge of our seats here ;p
I will make posts on a daily basis, as things happen. Today was crazy hectic. It seems like all the pieces are in place now.
Definitely with you about the bots, they seem to be procreating like crazy recently, this is most definitely NOT good for the Steemit platform as it will become a social(?) platform where bots do the upvoting and maybe soon even the writing too (then upvote their own posts)...until the whole thing crashes and burns. Keep us informed about calabrae.
You got an upvote & resteem from me.
Steem on bryond the moon, lets take it past the moon. Good luck to us all.
That's it!
After I built my botnet and people seen it I know many other people did the same. I reported the exploit that was used to create many accounts to Steemit Inc. and they completely ignored me until it became a huge problem.. There is a Russian man with 80,000 accounts now.
So if there is someone who has 80,000 accounts, plus all the other with more than 100 or 1000 accounts....there are not really 200,000 Steemit users in fact.
The charts all show huge growth but a lot of it is bots...sad news for Steemit.
Thanks for that info.
1,000 upvotes... I'm pretty confused by the post, but will be a new follower anyway, peace
That was @noganoo... he discovered a way to register hundreds of new accounts on the free signup. He submitted the exploit to @ned who gave him 100SBD. So he kept probing, found another one, and paid himself accounts instead. Grey-hat I suppose.
Anyway, if there was any sensible mechanisms of defense against spam on this platform, such an array of bots would be impossible to use and not very useful, because people could stop them posting. It's an interesting demonstration of a flaw in the system.
hmm, curious... ty
Thanks for the update mate. Looking forward to seeing how this plays out. Once live, I'll be cross posting content to help support you on the new chain.
Good luck.
If you create a script to automatically do this, and I certainly encourage this, it would be great if you could add it to the https://github.com/calibrae-project/contrib repository when/if you make such scripts. I will see to it also that the Shimmer interface will add an option to do this also for the submit post page.
I find this quite interesting, I did not know about this project until now and I am a bit confused but will read the old posts to try to get it.
I think Steemit may be entering on a decay era (not something too serious) because of the rise of bots so I am starting to follow you.
I think it would be accurate to say that only last night the complete set of rule changes have been made, I will be updating this. I am writing the wiki page for calibrae right now to put all the information into one place for easy reference. The link will be dropped in the next report.
This post has been upvoted and resteemed in the name of free speech. I may not agree with what you say, but I agree with your right to say it.