The whitepaper is dead; let it die.
The thing that the Swarm likes to call curation is not really curation, or at least it is artificial in nature. This is something I've touched on before in previous posts but it's possible I haven't really talked about it much since 2018 in the long-long-ago.
Curation is centralized in nature.
So we built this thing called Condenser and called it decentralized, but that's not 100% accurate. We basically juked the system by creating a frontend that enforced backend curation. Because this was the only frontend available a lot of people assumed that having a high payout on a post and curation were equivalent. They are not.
The frontend decides what gets shown and what does not get shown. The frontend is always owned and operated by a single agent, and thus centralized in nature. Doesn't really matter that it connects to Hive on the backend; it can technically show the user whatever it wants to. This creates a lot of confusion for those who don't understand how it works.
the action or process of selecting, organizing, and looking after the items in a collection or exhibition.
What does curation even mean?
It means your content gets promoted through an intermediary. The definition of curation is inherently centralized, as the entire point is that the content in question is being filtered through a middleman. In this case that intermediary is a centralized website connected to a decentralized backend. Does that make it centralized or decentralized?
The answer is 'yes'.
It ends up being both. The process of casting upvotes and downvotes is largely decentralized. When the frontend upholds those numbers and funnels them into an algorithm that creates a trending tab, that trending tab is kinda sorta decentralized in a way even though it was channeled through a centralized middleman.
However, as I have stated before, just because something is decentralized doesn't mean it's actually a good solution. We tend to forget this in crypto land because we often wear rose colored glasses when it comes to this tech. We are, after all, enthusiasts and purveyors of this tech stack. It would be foolish to assume that there isn't some sort of bias there because there is in fact quite a bit of bias and tribalism across the board.
Reblogging == Curation
The problem is that there is no financial incentive to reblog someone else's work. We click that button if we think the content is worth sharing with our follower list, but we don't actually get anything out of it, even though that is the literal definition of curation.
Personally I'm very conservative with which posts I choose to repost. I don't want to clutter up my feed or anyone else's. Don't want to be spamming others as that's rude. It's also the reason why I follow very few accounts. The type of content I want to be seeing on my feed is pretty limited. I have opened this up a little bit lately and followed accounts that I should have been following years ago, but that's another story.
Point being that sometimes when I've hit that reblog button: that post gets way more upvotes than the account in question would normally get. So I'm like, "Wow that actually worked these bigger accounts saw that post because I curated it directly." But does lacking a financial incentive cheapen the function? Or should we be curating like this altruistically?
What if we monetized the reblog function?
How would that even work? Well someone would have to spin up a new frontend or implement the feature on an already existing frontend. Then when a post gets clicked due to a reblog that user has to be tracked. If they upvote that post it counts as being successfully curated and the expectation is that the content creator being paid would share with the curator.
This could either be accomplished with Hive or a tribe token. Using Hive means the system is opt in and curated users would have to willingly give up some of their rewards by participating in the program. A tribe token or smart contract solution could automate the entire thing (or posting from a frontend that sends money to a centralized agent to act as escrow).
Is this actually a good idea?
Hm, yeah I don't know: that's a stretch. Getting noticed on Hive can be pretty difficult especially during the bull market. There's a lot of noise and good content can easily get missed. A financial incentive for real curation through follower lists could be beneficial, but I honestly can't say if it would or would not be worth the effort.
Here comes the spam!
The main thing to consider with something like this is: how much spam is it going to create? Is everyone just going to start running around reblogging everything they see in the hopes that they could exploit a system like this for rewards? This question leads us down an obvious path where we can't force content creators to accept the deal by default as it would be endlessly exploited for personal gain.
However, on the flip side of this argument it could be really good to give content creators the ability to divert some or even all of their rewards just to get more exposure to the platform. Thinking back to when I was a new user in 2017 I definitely would have utilized a feature like this. Do I want to earn pennies for a post that I spent 5 hours writing? Or do I want to earn $0 on the post and make sure the entire platform sees it? Back then it would have been an easy decision to get some quick exposure because I knew some of my posts were really good but hardly anyone was actually seeing them. If I had the option to give 100% of my rewards to curators (rebloggers) I would have done it in a heartbeat. Good investment is good.
Decentralization can be a huge disadvantage.
So imagine one of Hive's frontends implemented something like this and it turned out to be a pretty good feature, but it didn't work very well because the other frontends weren't in consensus with the idea. So one frontend is trying to track all the clicks from the reblogs and whatnot, but it can't track clicks on other frontends because those are controlled by different entities on separate servers.
See when someone like Twitter or Facebook makes a decision like this they can enforce it effortlessly across the entire platform, while something like Hive may be a little tricky or even impossible to implement across the board. Just something to consider generally. Again: decentralized doesn't mean "superior". It's actually quite a niche and narrow use-case in many circumstances.
Conclusion
- The algorithm is curation
- Reblogging is curation.
- Curation is curation.
- Upvotes/likes are only a part of the equation when fed via the algo.
The frontend determines what gets seen and what doesn't. It's whatever that entity decides, and nobody else's. Today the ultimate curators come in the form of megatech corporations like Google, Facebook, and TikTok. Perhaps we should be taking a lesson from them and incorporating what works for Web3 while throwing away what doesn't. Easier said than done, but probably worth it in the end.