#project-smackdown week 2 report
Week 2 report for #project-smackdown, the project reporting on the top 20 self comment voters and flagging the top 20 self voted comments every day.
Flags are made with accounts @smackdown.kitty and @sadkitten
Note, all "dollar" valuations in SBD, the native unit for rewards.
Summary
This week we look at the top self voted comments with self vote reward component greater than $1. This gave a total of 517 self voted comments valuing at $2,732.82
Top 20 comment self rewarded voters
In order: @adsactly, @jejujinfarm, @sigizzang, @oldtimer, @damarth, @hardikv, @ramta, @musicholic, @ctrl-alt-nwo, @oceancoinz, @grognak, @johnsmith, @danlupi, @me-tarzan, @surfyogi, @salva82, @webdeals, @novina, @trevonjb, @libertyteeth
This ranges between $468.084 and $30.391, with an average of $97.719 and median $62.184
Top 20 self voted comments
@adsactly takes the top 6 spots (highest self reward comment at $118.542), followed by @ramta for 2, @oldtimer for a total of 5, @damarth for 3, @sigizzang for 2, and @otisbrown and @trevonjb for one each.
Particular authors rewards across week 2
Some particular authors and their votes across week 2. This graph is a stacked bar chart, ordered vertical with the largest rewarded author on top. Note that the curved filled area does not adhere to the area principle and is for illustration purposes.
@adsactly - $468.084 week total
We already know that @adsactly is the top rewarded comment self voter by both amount summed over the week and also has the top 6 individual self voted comments above.
We can see from this chart also shows us they were top comment self voter (summed per day) on the 21st and 23rd, while second on 24th and 25th, self voting on 4 days out of 7.
@jejujinfarm - $288.666 week total
@jejujinfarm made a big impact on the first two days of the week (19th and 20th) as the highest comment self voter these days over a total of 21 votes. However they did not comment self vote above $1 on any other day.
@sigizzang - $266.314 week total
@sigizzang is 3rd highest self voter this week, with 2 comments in the top 20 self voted comments and having comment self voted on 4 days. They were top comment self voter (summed per day) on 24th July, 2nd 3rd and 4th respectively on 22nd, 20th and 21st. Another high self voter, especially relatively. Remember last week they were on top of the list with $331.382
Previous reports
Please see week 1 report for more details on the project if you are unfamiliar with it.
See my blog or the tag #project-smackdown for the individual reports for the first week.
Thank you for your interest
Please continue to follow the reports and let us know what you think!
I regularly self vote as I like to get my comments to the top where they will be more likely to be seen and engaged with. It is the same reason I upvote my posts.
One of the revenue goals of steemit from the start was to allow people to self promote their material as a form of advertising. If there was no advantage to buying steem why would anyone do it?
Cream does not rise to the top, look at reddit, the top comment is the earliest OK comment. You could have the best comment ever and post late and you wouldn't get seen by nearly anyone.
Anyway as you don't like it i will refrain from up voting myself here
I read your post a couple of weeks ago about self voting comment spam, and it's close to our idea here, as I said at the time.
Self up voting comments for exposure is something a lot of people do, and feel entitled to do. We've discussed it a lot at the Coop and I still side with the idea that large self up votes, even if used for comment exposure on a long thread, are not something I can stand behind. At the end of the day, these end up being rewards for the commenter which have been self assigned. Whatever way you cut it, it is not behavior I want to reward.
Of course that's just my opinion though, and I run this bot (with the help of others) to that end. I would not like to think that you blinding follow this just because I don't like it! Rather if you do refrain from self voting that it is the result of your own self thoughts on the matter 🙂
The bottom line is that voting for others needs to be more rewarded by the system than voting for ourselves. We cannot simply appeal to altruism as some (maybe a lot) will always take advantage of that situation.
I would be open to the idea of some <1 multiplier for self votes, but if it was too much I would just withdraw all my STEEM and continue to post with a low steem account. I would guess that 80-90% of my rewards come from posting, what would be the reason to own SP?
Influence on the platform was always the goal of SP as I understand it. That extends to voting, for posts and comments and also for witnesses.
The whitepaper is very clear that self voting needs to be less attractive in terms of rewards than voting for others, i.e. than curation. This was messed up by HF 19 at the expense of the good idea of giving more voting power to those with less SP.
The whole point of voting at all is to curate other people's posts. The rewards are motivators to this end. Self voting does not help this goal, it is simply pay for exposure. Who would really want that? It sounds like advertisers advertising to other advertisers.
If you withdrew your SP as STEEM because of comment up votes being less rewarded, or even entirely unrewarded, I guess you will have missed the point and would not have the opportunity to be rewarded for being part of something larger than getting people to view your own posts and comments.
I don't dismiss this reaction though and I am curious if there are a lot of people, especially high SP holders, who would do (or say they would do) the same.
Is there a PHP API that I'm unaware of to scrape statistics data similar to your reports from Steemit?
Thanks in advance & upboated.
This post received a 4.1% upvote from @randowhale thanks to @wolo! For more information, click here!
I don't think there is a PHP API but I'm not sure about that. I use the JavaScript library Steem-JS.
Time to brush up on my JavaScript it seems.
This post received a 3.9% upvote from @randowhale thanks to @wolo! For more information, click here!
I hadn't been doing much curating lately, too busy dreaming up DIV... but after 3 days mostly offline and not curating, I was pleased to spot this post about the project that I helped pioneer (and named), and that it is still going, and getting more sophisticated.
Good work @personz and @the-ego-is-you - and whoever else joined #steem-coop.
I had to leave because my detox-induced personality changes make me ... overzealous? Well, I'm settled a bit now, but I'm still calling it when I see it, no different to before.
I am not sure if the publicity wasn't good though...
Good to hear from you l0k1 :) The project has changed internally a couple times since you left, but continues to work it's way forward.
I am still struggling to manage my having two accounts. I need a script that resteems my OP's and votes the same as @elfspice so I don't keep posting as @l0k1 and keep complicating my fund management. I mean to bleed l0k1 dry and then throw it away... The 'immutability' of the steem blockchain I strenuously question, in any case, as you know. Technically it cannot grow more than 4x its current size without completely failing (256gb total shared memory size, currently pushing 64Gb for RPC nodes)
Thanks @l0k1!
Upvoted, Resteemed. Thanks for doing this work. It's necessary to hold a mirror up to our human behavior to find out what is going on.....
UPVOTING ALL OF ONE'S COMMENTS feels like the lowest form of human...I am appalled actually to find out this info.
Thank you @stellabelle, that means a lot, glad you agree it's important.
Just to be clear though, there is no information here which says that any of these authors up vote all of their comments.
Nice presentation interesting to see the top self-comment upvoters especially after seeing their comments on several of my recent posts! Self-comment upvoting in addition to the rewards also offers great self-promotion which makes it surprising I am not doing it!
It makes you look a lot cleaner the way you currently do it Jerry =)
Selfvotes, especially in cases where the vote seemed only to be cast to get the money, have always been a signifier to me of the likely psychological focus of the commenter. When someone selfupvotes, that makes me a lot less likely to vote for them unless others also voted and the comment really provided me direcly/indirectly with huge value. If everyone prioritized voting for himself rather than others, what kind of "social" network would we end up with after all?
Self-promotion has been a recent argument to keep selfvoting, as to a certain extent the promotion opportunity may be seen as adding investment value. In my opinion there's nothing dodgy about that argument, but there could be developed better ways of using this function that doesn't also give direct rewards in Steem or SBD.
However I should say that this issue is far more complex than most people realize (although I think you actually do realize much of it as I read two posts from you about selfvoting) and involves the risk of a "sybil" attack. That is, in other words, the risks associated with creating multiple accounts and, in the least bad scenario, of making a fortune through bots while slowly depreciation the value of the network. In worst case, attacking the network to strenghten a competitor. While I can't calculate how big this risk is as of today, it certainly exists.
What's worse is that trying to fix it, if not done with extreme care, could easily make it worse. So that's why I personally think the bot is a good first step to signal the problem to users and to put some psychological pressure on big upvoters, but next will have to be a much harder discussion of how to significantly block sybil attacks. This issue is so hard in fact, that @ dan used it as an argument to allow selfvotes in order to somewhat counter the use of multiple accounts.
My wish is that the community will at some point start to discuss these risks more deeply and come up with a real solution, not merely trying to push it into the future with minor tweaks - even as I of course would accept any improvement.
Love your exploration of this in depth because of the conclusion we need to keep self upvotes to limit the value of making duplicate accounts! When I started out on Steemit, I upvoted myself a lot until I realized it felt better to give for others. Self upvoting presents a great opportunity for most of us to learn that we feel better giving to others than to ourselves! Still it is so tempting to drop that upvote on myself and I appreciate you helping me remember to save my upvotes for others!
Thank you 😊 Can I ask why are you not doing it?
@personz thank you for asking because the only reason I am not doing it is that I already did do it when I first joined Steemit. With exponential rewards then, I could only get $1 a comment with my full upvote and then I realized that after blowing all my voting power on myself, I did not feel very good and better to vote for others. Kind of like sex where it is better to do it with someone else, voting on others is a lot more fun which I guess is why you do not vote yourself up as a witness?
Since then the situation has changed though and it's much more rewarding to vote on yourself than it was. I like your story because it shows this clearly. Now what we have is people going through the first step (up voting themselves) and not having the realization that it's better (and more rewarding) to vote for others. I still think we need a system level adjustment to improve this situation. That's just for new people - I think experienced members are seeing self voting as interest payments on their investment, which was never the intended purpose of voting.
I used to vote myself as a witness but I became convinced that the purpose of voting is primarily to vote for others, since a vote is necessarily evaluative and we are not qualified to judge our own work or our own suitability as witness. Until such a time as this is corrected on a system level I make the example of what I think should be the case.
As I am engaged in the controversial work of disagreeing with self voting and taking action against it, I also need to be sure that I am "practicing what I preach", which I am happy to do, it makes sense to me.
Thanks for the comment and question.
Thanks for doing this, I might make a squeaky wheel, but it takes someone with your skills to make a difference.
Is this the prime factor in the rewards dropping off?
Thank you 😊 I don't yet know if it is a prime factor. I only have the computing power (and time) to scan for the top self voted comments at this time. I'd really like to see a deeper data analysis and I hope to do it soon in collaboration with some others in the #steem-coop.
What I do know is that much of these top self voting rewards are so far above what the vast majority of users actually are rewarded that it is an important issue for the shape of the platform. Are we going to say that we stand by this? Or will we oppose it?
Do you think that increasing the power-down time again might help mitigate this? It occurs to me that 'self-voters' are in general probably less interested in the long-term success of the platform than 'community members'.
That's perhaps a reasonable assumption. It might, and I think it might be a good thing in any case.
I think we just have to shed the light and the bugs will scatter.
It is up to us to decide what behaviors we will tolerate.
I'll try to keep up with what is happening.
I feel like ideally this problem is dealt with at the code level. I am not sure policing like this can scale to millions of users. There should be a limit of self votes allowed per day... users that are not self voting feel like they are losing out out to the ones that are...
I agree with you completely that it should be system code change. We are not intending to run this forever. Perhaps the solution is some kind of bot (though I really hope not) but yes, we are campaigning for a hard fork change.
Does the bot take into account the percentage of a person's vote that goes to themselves?
For example if a whale upvotes 100% every time and votes for themselves 2 times and others 98 times, there are clearly not being selfish, but those 2 votes would activate your bot, no?
No, it only orders comments by pending payout (the amount the self vote is worth at time of vote) and the top 20 comments by this order are flagged. So the bot makes no distinction how often you do it, just if it is of significant value.
The report is on the greater statistics gathered.
We are talking about a phase 2 which would take more information into account, more information on this coming when we have discussed further. It's not as simple though, and some of the strength of the current approach is that it is very clear why and how comments which are flagged are chosen, and it is largely non subjective.
interesting report. some even comment on your post and upvotes their comment without even upvoting your post. it hurts
You should add Spam votes to the list. These are not the most upvoted comments on Steemit. This report should inlcude comments with 3000 bot upvotes. They are very common: https://steemit.com/spam/@ausbitbank/so-i-just-flagged-a-spammer-with-3400-accounts-voting-numbered-comments
We're looking at starting a #project-smackdown sub project looking at this stuff. If it materializes you'll see it here 😎