I really love your points raised. A second option that has been discussed is making the 20% we get from posts serve as a monthly reward pool which will be shared amongst writers based on the percentage of the overall total reads generated by their articles on the platform during the course said month.

After the first couple of months, this is going to be heavily weighted toward those with established presence on the platform and almost none of it is going to go to newcomers, and that's the way it's going to be until you shut down the project.

Now, that might be exactly what you want, but from the perspective of someone who is coming along later, they have almost no chance of competing in the midterm with those who already have an established presence. You are basically holding them in with the hope that they will be someone who is recognized as a creator within this niche in coming cycles.

You have effectively, philosophically, re-created "publish or perish" as well as a system which simultaneously rewards simply having been on the platform the longest. I am not sure that was entirely your intention.

Also, you immediately open yourself to be the target of some sort of distributed bot "reading" swarm, given that you are going to distribute that 20% by "reads." If I can buy up votes on the blockchain, I can most certainly buy "reads" for the same distribution mechanism.

It's hard to get away from the multilevel marketing whiff coming off of this. After all, let's consider the degenerate case in which no books are voted up for a month. There will be no 20% to take. It doesn't matter how established the writer is, they get nothing. Now let's consider the case where the number of people publishing books on the system triples, but the number of consumer and consumer-hours does not. In that period, each view becomes worth effectively 1/3 of what it was before. Unless the amount of the 20% taken can grow faster than the number of people who want to be involved in the system by a good amount, it becomes more advantageous for those already in to keep people out in order to protect their rewards.

You've just reimplemented a reward pool that mimics the same failure modes as the Steemit reward pool. It just keys often easier mechanism to automate.

However, these are still early days for us and we would need all the help and funds (the initial purpose of the 20%) we can get to develop the product properly

Unfortunately, you need to do more than minimum viable product to have a successful rollout. You have to have a system built in theory which actually does what you say you want it to do. It needs to be resistant to exploitation by automation, because we already know that sort of exploitation exists on the platform we've got. It needs to not set up mechanics which actively incentivize bad actor behavior. It needs to decide what it wants to do before you start implementing anything.

I realize that the "don't think, go fast, fail fast" process is popular in Silicon Valley circles, but I've always been fond of replacing that whole architecture with "don't fail."

For me, as a writer who is already published, I need some kind of understanding of the theoretical underpinning of how it's all supposed to work devoid of handwaving, and which reflects a fair amount of consideration about my experience as a user, to be the kind of person that gets on board this sort of thing.

I may not be the kind of user you're targeting. You may be looking for the people who are just looking for a lighter weight, less promising vanity press. Which is fine, as far as it goes, but if so you probably need to really lean heavily on what advantages your system will provide over what already exists on Amazon and Google Books – and if you can't answer those questions, find some answers to them.

There are a lot of ways for independent authors to self publish these days, ones in which they retain most of the rights of distribution going forward after publication. You have to compete with those.