Plus, retention is elusive. We can't expect people to be glued on Steemit 24/7, some will take off and do their own stuff and maybe come back later. This is the true test of a good social platform and it's hard to measure.
All in all, the point is that guilds are distributing far more votes than a few whales could possibly achieve themselves. I personally don't see what all the accusations about "centralisation" of power are about.
without getting into a debate about the success of the retention program, I just have to point out that this is super convienient (and also super bad) "logic"
Obv, the ones that youre still voting on are still here. otherwise, you wouldn't still be voting on them. The question is how many have been lost because even scraps you throw them (while retaining the lions share of the reward pool for the under-the-table funding of various projects all run by the same group of people) aren't sufficient to keep them interested.
you can't both claim that youre retention efforts have been successful, and that measurement of such success is elusive.
Yup definitely not arguing about whether or not guild efforts have been successful retaining people. We can't control anyone. The point is helping whales distribute far more votes. There was a period back in middle 2016 when the zipfian distribution was heavily skewed only for the few early adopters. Now the tail is fatter.
I don't necessarily agree that zeta is the most relevant stastic to measure rewards, but i think it would be interesting to see a real statistical analysis of late july zeta. vs zeta right now.
A more valid stastical metric for the success of reward distribution between mid july and now would be weekly lists of users, grouped according to standard deviation from the mean for rewards, and how much of the reward pool went to which standard deviations.
I suspect (though i havent yet been able to figure out how to pull out the information in an automated way) that neither metric would support your position.
I'd like to see that each one of the guild member leading one whale's vote, but not all of the members leading all of the whales. This will smooth out the reward distribution.
This is arguable. One whale can hire one curator to vote 40 times a day, 5 whales then there will be 200 times in total, it's similar to the number of votes cast by the guild everyday.
Yup, that would work too, although it's slightly different for a guild like curie. There needs to be 2-3 vouches for a single post to get voted on (I believe this is the case for SG too).
It's not just up to the whims of one person to vote on anything anytime. For example, I have friends who joined Steemit but I'll personally avoid suggesting / voting on any of them.
Even if I were to help out by always suggesting their stuff, another one or two person still needs to vouch for it - I can't just submit a friend's crappy posts and get them voted on that easily. It's a way to reduce chances of insider help and avoid abusing voting power given by the whales.
Imagine whale-powered solo curators coming up with good looking sock-puppet accounts to vote on them easily. Without other parties to vouch for such posts / accounts, such an arrangement could very well be abused. @abit
Also to add - it's not going to be possible for one curator to cover such big grounds, especially if the # of posts are gonna be growing. It's easy to vote 40 times a day, but it certainly takes way more time to check through pretty much all posts of the day and decide where the 40 will go to. (in terms of Curie's generic curation).
Certainly arguable, but I hope you see that it's not that straight-forward.
Edited to add this part:-
If you're unaware, we don't frontrun whales by voting with our personal accounts.. we'll always use a non-personal account. Our personal account votes will never trigger anything from the whales.
(added: by the way , what you've mentioned is better suited for comment curation).
Thanks for the reply.
IMHO that's just how capital/market works. Everyone give out their opinions, of course different people will have different opinions, at the end, the ones get positive feedback from all voters get the most rewards, the ones got less positive feedback get less rewards. If I understood correctly, your approach is more like communism, only reward the ones that you all agreed to reward, as someone pointed out, it's less likely to scale well.
By the way, when I used the word "leading", I meant that one member vote with a whale's account, but not necessarily mean front running, from an author's point of view it's no difference.
[tree limit..] @abit
Ah yeap, leading / frontrunning (not accusing of any intent to cheat backers btw hehe - you're right there's no difference).
IMHO everything that happens is what happens in the free-market, no matter if things look like capitalism or communism or socialism. I've got nothing against solo or group curation - everything has pros and cons. That said, different parts of the free-market will quite simply, develop different structures until it dies off or evolves according to market sentiment at any moment in time.
So how Curie works at this moment is anyone will be able to suggest posts on Curiesteem.com, and there are about 6 of us on the approval side of things (looking to "decentralize" our approval powers too). So there is no account list to support. The only "list" is any account below a certain reputation (level 62 at time of writing).
There's also a curator score / approval rating system in place too to promote great curators and reducing spam / low quality submissions. In that, I believe Curie will be able to scale very well since the "workforce" is essentially the entire community itself.
It's a community project now that incentives accurate, quality submissions - https://steemit.com/curation/@curie/curie-and-community-an-open-invitation-for-all-to-be-part-of-our-curation-works-powered-by-streemian
(Many may not have read it since I guess there's some bad name for guilds in general, so just linking it here for your perusal and consideration - imo, Daily Curie list is getting much better these days)
SG on the other hand is rather different with its growing fixed set list of authors to support (basically selected authors from level 63 and above, at time of writing).