If you can get curation from a 20hour post that has $500 on it already, then curation no longer even exists.

Curation, in the actual sense of the word, doesn't exist now.

People are already ignoring the content and gaming for maximum investment, chasing the Dragon of maximum payout – no matter what the content is. What you describe is exactly what's happening, and all you really have to do is look at the way that systems work and what is actually being rewarded to see that's the case.

Consider this from the opposite direction: why shouldn't you get curation rewards for rewarding someone who is being successful? Even at 24 hours out? Are users of the platform not supposed to sleep? Are they supposed to get up in the middle of the night when/if someone who is known for producing content that they want to reward is doing so on the opposite side of the planet?

Your description of the failure mode of extending the voting window for curation is exactly the description of the failure mode that we see right now – but because of the underlying architecture of the way that rewards are structured. It's just that right now, people race to hit that mystical 20 minute mark, gambling on whether content is going to be a big hit in the next 10 minutes rather than whether people will have time to find the content and digest the content over the next day.

It's the race and the fact that what you see is not actually curation even as you describe it which is the real problem. Extending that "curation" window only has the side effect of letting more people get involved without changing the essential drive. Given that rewards get split, no matter how long things run for, and running things for longer both let's more value accumulate and splits it more ways – assuming more people get involved – there should be no net change in overall distribution.

Except that more people will be rewarded for attempting to use the system as you say it's intended to be used and not as a 30 minute race.

Your idea for 24h curation window is fundamentally flawed in that it completely invalidates what curation is made to do - give incentive to vote on good posts that have little to no rewards in anticipation that they will end up gaining more rewards from their quality and from you boosting it up the chart a little bit.

Note that your description has nothing to do with the quality of the post. It has nothing to do with the content. It has nothing to do with actual visibility or making sure that that post gets in front of people who are likely to also want to vote it up. The only thing that's important in your description, right there, is that you engage in a betting pool on whether or not a post is going to go up in value in a 30 minute slice.

And that's all.

It is, in fact, completely trivial to have a bot go around voting on all the highest paying posts to take a share of their awards – and it's being done right now. By lots of bots. And it's being done by lots of bots because there is no way that a human being could possibly be aware of all of the posts hitting the blockchain and respond at just the right time. Because the time window is so tight and so constant, the minigame of "curation" is completely broken in favor of bots, which drags the focus even further from content because bots just don't care.

As such, the economy of reward for engaging in curation does the opposite of providing for actual good curation. Instead, think of the blockchain as a warm a mass of fluid with a supercritical density of material dissolved in it. All of this material looking to find something growing just fast enough to attract all the rest. All it takes is one very light whale vote and suddenly all of the mass dissolved near it begins to accrete, not because the thing at the middle is worth anything, but because there is been a piece of particulate which has been attached and suddenly there's a swarm of more material looking to latch on.

Keeping the "curation" period at 30 minutes keeps the energy density in the fluid high, meaning any tiny little change will cause a mass material agglomeration. Lengthening the time over which curation can be rewarded would serve to reduce the energy level of the pool in general, allowing more considered, more human, engagement with the content.

If anything, your argument points directly against your declared position. Everything that you say would result has already resulted; that's what we see happening right now.

It's not going to get worse, because it's already about as bad as it can get – except for the fact that they are intending to shorten the "curation" time even further, driving more energy into the system, bringing it to an even higher boil, which will drive even more bot engagement which does nothing but try and chase the Dragon on following around significant whale votes, and nothing for actually getting good content to get good reward.

The only thing that limits your nightmare scenario from being completely overwhelming at this moment is the recharge time on VP, and while I think that could probably use a real change at an elemental level (probably by making the default voting percentage before someone has enough SP to get the slider 20%, bringing someone's daily votes to 50 before they lose appreciable voting power, without actually changing the amount of SP that they bring into play), and that's it. Because bots have a decaying, linear degradation in the amount of SP that they can bring to bear in 24 hours, that's the only thing that keeps your nightmare scenario even remotely at bay – and the ease at which bot accounts can be created, invested, and set loose has become a true infestation since hardfork 19 which helped make the situation as bad as it is.

One good thing you mention though, is that reward payouts could happen indefinitely [...]

Another user had an idea in that particular field that I thought was particularly insightful. They suggested that instead of an upvote on articles older than seven days, there would be a "tip" button, pressing upon which would open a slider interface that would effectively allow you to send them the amount of a vote directly. I don't have a problem with that. That would lead to evergreen content that people found and liked still being able to be rewarded by whoever wanted to.

Of course, in a very direct and effective way, that's basically just allowing upvotes on anything forever, but mechanically it is identical to finding someone who's post you like that's older than seven days, going to your wallet, opening it up, putting their name in, and sending them SBD or steem up to the value of one of your votes – or beyond. Which you can do right now. It would just make it easy and reasonable to do so.

I think you are very smart in the way you describe things, otherwise I would have just passed along, so don't take offense to my stance on this it is just some food for thought (and thinking is something I get the impression you like to do)

If I took offense easily, I certainly would never get into discussions on Steemit about financial issues. There is a disturbingly strong authoritarian, top-down architecture when it comes to how to manage other people's money, as it were – and my inherent antiauthoritarian Libertarian nature raises its hackles when someone tells me that they know better than I do about what I should be doing with my cash.

I operate under the belief that the more people that you have engaged with the system, the more likely you are to get something useful out of it. Things that keep people from engaging with the system will destroy the usefulness of that system.

Cutting down the window in which one can get curation rewards from upvoting content decreases the amount of time that people can engage with the content before they can be rewarded for finding something good. I find that to be inherently bad. The idea that the content doesn't matter, that the only thing that matters is whether or not you are willing to bet the value will go up instead of looking at it as an opportunity to reward the creator for giving you something that you like, I think that's inherently bad. This is the sort of thing that has led to Steemit being a little more than problematic when it comes to getting bloggers on board, because the community – by and large – doesn't give two farts in a whirlwind about content. The system doesn't really care about content. We are extremely lucky that some individuals are really passionate about good content, and some of them are members with significant blockchain-based power – but even as power-leaders, they can profit off the fact that as significant actors, when they vote on a piece and give it a big payday, it attracts other people who want to get on board the rising balloon for the payday, not because the content is good.

What Steemit refers to as "curation" is anything but. Again, it's a betting pool. You're betting on whether something is going to go viral, not based on its inherent qualities necessarily, but whether someone with real money is going to come along and vote it up.

Even when I profit thereby, I am uncomfortable with the fact that the system doesn't do what it says it does, because I think what it says it does or wants to do is a thing that needs to be done. We need mechanics which actually help us surface content that we are likely to enjoy and want to reward the creator of. We need mechanics which make it easier and more effective to engage with creators and content that gives us pleasure, and part of that is incentivizing finding that content to more people. Not less people.

Remember, every comment is a brand-new post as far as the system is concerned, so you can still harvest rewards on it for seven days. A month old post with an active comment thread is still an active engine of financial generation.