Disclaimer: I'm an operator in Curie. This is my own response to the various criticisms seen in other posts.
A minimum viable product
Posts are Steemit's major unit of communication. It is in anyone's interest to improve distributed judging and resource allocation all around the platform so that the network resources are going to good, honest actors. Not plagiarists, or those with the intent to defraud the community. Personally, I think that's the minimum filter required for every post on Steemit.
The way to do it is NOT by using a bot script to distribute votes blindly and indiscriminately. If anything, this method will actually scale-up error of resource distribution. It doesn't take a genius to figure out a way to abuse such a distribution model by registering loads of small accounts pumping out random shitposts. It will be a disaster if everybody supported such a distribution bot. It's conceivably the worst way to invest.
So to improve such a situation where there's just a sea of posts, the market is free to develop whatever it wants. This improvement filter can either be in the form of bots, people, and will most likely end up as a combination of both worlds. Curie is one of many initiatives, or methods to improve the network.
Here's how Curie works
Organizational structure
We are divided into two major groups. First is the internal curation group consisting post finders, post vouchers, and ultimately, the proxy voters. Second is the external curation group. It's open to the rest of the community to suggest posts as long as these post meet the guidelines that we have developed organically over the course of our collective experience operating Curie.
All Curie members are quite simply operators. We do not have any defined roles, and we can take-up and pass on roles anytime as long as time and skills permit. Other than the more technical stuff like maintaining a witness and developing the software to assist Curie works, anybody in Curie can be part of any operational process stated below.
At the time of writing, we have nine (9) core Curie operators, and more than five-hundred (500) participants in #curie on https://steemit.chat.
Minimum operational process
Here's what happens to each and every post before it gets the @curie vote:-
- Plagiarism check.
- Account background and consistency check.
- Recommended posts must be supported by other Curie members as well. Anybody can also speak their mind about the suggested post or account as well to sway the final proxy voter's decision.
- Determine voting weight according voting power availability and quality of post.
*Please note that quality of post has more to do with the consistency of account. We do not entirely disregard posts with stuff that we may not personally agree with like bad grammar, etc. That said, it's important to have an efficient group of people speaking their minds to debate about voting issues. We are certainly still learning what it means to reward a diverse crowd.*
Funding the operations
There are several common misconceptions about the flow of funds generated by @curie's curation rewards and its daily posts. First off, curation rewards of the top influencers supporting Curie are mostly returned to @curie on top of the rewards generated by the daily posting. As per the recent Daily Curie, this is true for every word:-
100% of said funds are paid out to the hundreds of curators submitting posts, the staff processing them, and various other expenses such as software development, server expenses, proxy account acquisition, etc. All transactions are completely transparent and visible on @curie's wallet.
Nobody here is cashing out to buy a Lamborghini. Funds are mostly distributed widely to quality curators for their time and effort. There's real work done on all fronts.
Improving the operation
Over the past few months since Curie started, we have been intensely modifying the payout amounts to suit the reward pool size. In fact, we have lowered it down so much that some earlier Curie members could not continue doing quality curation full-time. They weren't easy decisions but it had to be done. Most of the active Curie operators these days are actually located in countries with lower cost-of-living.
In order to scale-up and improve the cost of operations at the same time, we are looking for ways to automate both book-keeping and post / account value-detection. This is by no means to replace manual curation entirely, but just to lessen the time, effort, and cost spent for each post.
Personally, I think the large dataset available through Curie's many thousands of votes throughout the months could prove to be beneficial to producing some type of autofilter using machine learning to process posts (and also account activities) at one point.
Also, in my own wish to develop a gig platform on top of Steemit, I may begin to propose including account meta-tagging as part of Curie operations in the future to add more value into our curation works.
Curie operations shouldn't be funded by the daily reward pool?
As mentioned earlier, most top voters supporting Curie do not earn from their curation rewards. In fact, most curation rewards are given back to @curie to be distributed to those that have performed work - spending their time and effort curating. (Edit, I removed the part that says that 100% of curation rewards from accounts like @val-a are returned to fund Curie's operation. It is more complicated than that as the funding comes in form of SP and some other considerations like liquidation period, taxes, etc)
Personally, I think a Minimum Viable Product (or Service) like Curie and Steemcleaners could use wider community realisation and support. If there's a better way to do this while instilling confidence for network investors, please spark a discussion on it. So far, @curie has been limiting itself to basic list posting, maintaining a neutral position without providing any other information other than its intended function.
I really don't see how else is @curie going to get funded if not by part of the daily reward pool. In a way, it simply means that the community is deciding that the small daily portion is for keeping up with Steemit's MVP / MVS.
The annoyance of repetitive trending posts
This is certainly a quirk of Steemit that I'm also susceptible to it. The Daily Curie is method to fund the operation - there's nothing to hide here. One way to spice it up would be to make it a unique post by featuring some up and coming writer, and then append the usual Daily Curie list and stats at the end of the post.
However that's just a play on perception, and I sure hope that annoyance isn't the only argument for assessing value. It's really not an ideal situation - everybody in Curie knows that. But for each day a Daily Curie is up there on trending, displacing what could have been something else, there is a list of other 60-120 posts that have gone through various processes executed mainly by 9 Curie operators and a good number of other quality curators on #curie.
While The Daily Curie looks repetitive and boring as hell, the list contained in it is surely not.
Future plans
The best organization is to have each person on Steemit empowered with the tools to be effective curators, especially for those with high stakes looking out for the network's best interests. In a way of speaking, the most efficient unit of Curie is simply one person without all the fat of an organization. I believe the abundance of data and community participation that comes along with the Curie initiative will benefit the system over the long-run. That's certainly needed to come up with solutions to remove the vetting organization. It is something that we're learning over time.
If Curie has a goal, it is really to figure out ways to dissolve and make itself irrelevant at one point. But for now, whales with voting power but not enough time may consider us @curie as a trusted curation guild. I hope the community continues its lively debates around curation guild and ways to improve the situation.