You are viewing a single comment's thread from:

RE: The Curation Conundrum

in #curation7 years ago (edited)

I think what we need is to develop some type of system where we can use the wisdom of the crowd to identify quality posts / gems, and provide the information to the larger stakeholders so that they can review/upvote it.

The challenge is how to do this in a way that doesn't spam the stakeholders with non-quality posts.

I think that if there was an efficient way to provide the stakeholders with a list of actual quality posts, that many of them would happily vote for them. Changes to the actual platform may not be needed.

Sort:  

I think that if there was an efficient way to provide the stakeholders with a list of actual quality posts, that many of them would happily vote for them. Changes to the actual platform may not be needed.

Yes although in some sense this is what is supposed to happen already, with the quality posts rising to the top. I think some of the curation guilds were also trying to achieve this but they also have some inherent problems. One alternative would be to have paid curators who only curate and don't actually post - not sure how you could guarantee that or indeed fund it.

I agree that it is how it is supposed to work, but the inherent problems outlined in your post are preventing it from working as designed. "The crowd" does not have enough SP to make any difference with post ranking, even if they find quality posts. Most of the stakeholders who can make a difference do not have the time to filter though all the content.

Those suggestions may work. I think we need to continue brainstorming on ways to channel the wisdom/efforts of the crowd into effective curation.

@timcliff & @thecryptofiend, I think we're starting to get somewhere, now... there might actually be a possible workaround that possibly could be modified to work here. Please indulge this quick "brain dump:"

Back "before God invented dirt" I was part of a content for rewards site named "epinions;" this was circa 1999-2000. User generated content; peer reviewed/curated. Pretty much the same issues we're facing here... after a while curation ended up being largely in the hands of a few hundred people who really "cared" about the site... NOT a viable system for sorting millions of posts. So an improved system was devised.

Now, my memory is a bit hazy, but here's the general gist: Using Steemit terminology, we would have a separate "curation reputation," earned literally as a result of interaction with content. It was simple-- you were either a "member" or a "bronze," "silver" or "gold" curator. Your curation status was independent of other activity... A gold curator could have 17 posts or 17,000 original posts. Anyway, the incentivizing part here was that your rank was basically a multiplier on your rewards... and (as I recall) there was also something built into it to where if you slacked off, you could lose rank as a curator.

I should add that the system was NOT "game proof;" there were always some people trying to beat the system, rather than use it. But it certainly DID get a lot more people involved in the "read and rate" process.

Now in the context of the Steemit ecosystem, a minnow could work his or her way to Gold curating level while still being a minnow from a posting perspective. Let's arbitrarily say that a "Gold" curator would have to read, vote and comment on at least 50 posts a day over a 30-day period to get there... and then maintain that to stay there. Which sort of addresses the issue of having both content creators and content consumers on Steemit. We could potentially attract people to Steemit who say "but I don't write," who could still earn viable rewards as curators. Mind you, for this to be financially viable, we'd probably need to go back to the 50/50 split.

As an additional wrinkle-- and look forward-- with "Communities" on Steemit becoming a reality down the road, the top curators in certain topics often ended up as "Community Managers" (which we may discover we need here), tasked with such things as recategorizing (in our case, re-tagging) content that was grossly miscategorized in attempts to falsely gain visibility... not sure how we'd do that here... maybe an authentic use for flags? These were completely voluntary (and unpaid) positions, but they rewarded indirectly because your name would be everywhere, leading to increased exposure to your content, hence increased rewards.

Again, I am NOT a developer, and NOT a blockchainiac... so I'm totally OK with you telling me to sit down and shut up! Just tossing it out there....

Edited to add: The reason I bring up this particular case is that of the 100s of attempts of "rewards for contents" epinions was the longest lived and most spam-resistant of all... I got my last royalty payment in 2014(!) before then owner eBay pulled the plug on paying contributors. 15 years is an eternity in "web years."

Really great ideas! I will have to think more about how this could be adopted to the blockchain technology, but it is a great place to start and gives lots of food for thought.

($171.00 PROMOTED POST GOT MY ATTENTION!)

"Put the carrots where you need the work to be done."

Does the [WALLET] [PERMISSIONS] [POSTING] in each Steemit account ONLY allow voting and resteeming capabilities without sacrificing Active and Owner keys that would give FULL access to that person who lets say was TEMPORARILY working on a shift to help CURRATE for 24hours or possibly weekly?

Given that person would get a pecentage of the rewards for the work and time and effort after his or her shift was finished and curation rewards paid out MANUALLY or AUTOMATICALLY.

TL;DR

Cuaration/Curator Position (FULL TIME or PART TIME)

Freelance Temporary with possibility of handling the Rented or Delegated Steemit Account after say 3-6months of Curration Performance.

Any wrong doing or misuse of this Rented/Delegated account would have the Permissions revoked or changed. (The posting key is used for posting and voting. It should be different from the active and owner keys.)

The posting key would give the ability to vote (curate), post, comment, and resteem without access to the account's funds. Giving the user delegated SP and asking for a cut would accomplish basically the same thing. smooth used to hire curators back in the day btw - pretty much exactly what you described.

Having a separate account NOT a personal account which may be seen to play favorites with having a particular steemit user name showing they UPVOTED or RESTEEMED a CURATED post. For example:

Newly Created Steemit account name
@curationteamsteem1
@curationteamsteam2
@mondaycurator
@tuesdaycurator
etc

These handles can be 25 rep with delegated or funded SP which can be recycled to the next Freelancer Temporary Curator. (a set of new POSTING keys would be recommended)

I suppose the question is, if such a system works most of the time does it matter? By trying to only find solutions that are 100% game proof we may end up with generally subpar systems in reality.

Further if such a system were found I am almost certain some person somewhere would figure out a means to exploit it

Bad human behaviour can never be completely eliminated. All you can do is try to mitigate against it where you can but it isn't always possible.

This is the difference between how many people who work in computing think vs what we see in biological systems.

Biological systems work optimally most of the time and in most situations, whether they work in exceptional circumstances or not is somewhat irrelevant.

I'm not saying we should ignore potential problems or risks only that we should not lose sight of what is important and the vast bulk of common circumstances, merely to mitigate against exceptions.

From where I am sitting, "mostly works" is a LOT better than "sometimes works."

Perfect solutions seldom exist outside of intellectual think tanks.

Exactly. Yet it seems some people are stuck in this form of thought.

Something I agree with

I see examples with this solution every day. There are curation guilds, trails, and groups of people who compile lists of quality content. It serves a purpose, is thoughtful and rewarding.

The problem arises where these people are paid to present someone elses content while the original author receives minimal rewards. No doubt due to similar reasons that you've outlined.

Another function that could fill this roll is resteeming. The issue here is that it's not rewarding at all and can be misused without even realizing it. (at least in the minds of your audience)

I think there are indeed better solutions and we should keep thinking to see what we can come up with that is fair, rewarding and effective.

Those suggestions may work. I think we need to continue brainstorming on ways to channel the wisdom/efforts of the crowd into effective curation.

Exactly. Thanks for you wise words mate. I'm just glad that there are people in the community like you who understand such issues and are thinking about it too. I am confident that it is a problem which we will find better solutions to through such discussion and probably some degree of experimentation. This is one of the reasons I find behavioural economics so fascinating - things don't always work out as expected and incentives are not always a simple thing.

I think we need to continue brainstorming on ways to channel the wisdom/efforts of the crowd into effective curation.

@timcliff I described a filtering system that can allow us to do that and help curators a great deal. If you are interested: https://steemit.com/steemit/@borislavzlatanov/proposal-make-it-easier-to-find-quality-posts-on-steemit

@thecryptofiend I think my proposal addresses a lot of the problems you raised, too.

I would be happy to get some feedback because I currently don't know if people don't see my proposal as good or the post just doesn't get noticed amongst the flood of other posts.

Thanks for sharing I will give it a look when I have a moment. It is hard to get things seen in the sea of posts.

It seems like an interesting project, but can you please explain what it has to do with our discussion on curation? The way it is randomly plugged as a reply to my comment kind of seems like spam.

The developer think tank has a dbase for centralizing these types of issues and solutions at http://developmentcenter.ml (linked in that article)

Cool, thanks!