Several people have wondered why I'm not spending my time opposing the so-called "Econonic Improvement Proposal" this time around. As a term it's one-third true: it's mostly about fighting spam, and it's very much not an improvement, but I can't really argue with it being a proposal. However, while I still think it's a bad idea, I no longer think that it matters all that much, because over the six months since the last time we fought about it I've become convinced that Steem has such large-scale issues that any tinkering with the rewards curve is essentially irrelevant. Even if the proposed changes really did move the rewards curve further toward optimal - which they don't - optimizing the rewards curve isn't merely rearranging the deck chairs on the Titanic, it's more like tightening the screws on the deck chairs on the Titanic. So I haven't really seen addressing it as worth my time.
Instead, though, I'd like to put forward some potential paths for making Steem viable in the long term. These are meant to address three large-scale issues with the current design of the blockchain:
Here I'm going to present three possible futures where Steem could become more viable. They aren't the only ones, and I'm sure I could come up with some others, but I think they're essentially representative of the variety of available approaches. All three start with the same initial rehabilitation step before processing to larger-scale changes.
1. Reduce the power-down time to one week.
A huge part of what makes Steem Power a mathematical nightmare to hold is what has become a ludicrous unstaking time. Eos and Tron have established three days as the industry standard, and in the long term Steem should seek to emulate or exceed them. However, current limitations of the Steem economic system mean that the power-down time can't reasonably be shorter than the post payout time. So in terms of immediate changes with minimal technical effort, one week is the best we can do at this time.
2. Professionalize the block producers.
If you compare the top block producers on Eos or Tron with Steem's top witnesses, the differences are stark. The more successful chains are operated by BPs who are formed into legally existing entities, have websites, mission statements, official representatives, and just by and large treat the job like something they're pursuing seriously. Steem has a guy who likes to cosplay a space cowboy from a terrible TV show. There's no witness on Steem who would be considered for a minute as a BP on any top chain, and that's a problem. People considering committing their time and money to Steem are going to notice that.
Fixing this is ultimately on the existing top witnesses or on the witness voters, which generally translates into being on 
This proposal addresses Steem's issues through moving most blockchain activity to Smart Media Tokens rather than the core Steem coin.
1. Develop SMTs in such a way that they can take over the content voting system. Remove content and voting from the base layer.
This isn't quite what SCOT is doing, but it's not too far off from it. By moving all voting from the Steem coin itself to Steem-based tokens, we have the opportunity to obviate Steem's distribution issues, solve the too-long power-down issue, and allow for diverse development all at the same time. SMTs could operate on varied economic systems, allowing for optimization through rapid trial and selection, rather than guessing and hardforking.
2. Convert Steem rewards to more-standard Proof of Stake payouts, and reduce the power-down time to industry standards
Steem inflation under this system (probably drastically reduced) would either be paid out on a purely stake-weighted basis, or use the Tron system of paying all the inflation to the top block producers and allowing them to reward their voters. The latter, while considerably out of the line of current Steem culture, seems to work pretty well.
Content rewards would be paid entirely in SMTs.
This proposal maintains content voting on the base layer while allowing much more flexibility to users.
1. Convert the voting system to fully atomic voting.
We took a big step in this direction with Voting Mana in Hardfork 20, but there's still a significant amount of work to do to get to the ideal. The concept here is to make each staked Steem vote and recharge independently - actions would still take voting power, but that voting power would move smoothly between actions. Rather than delegating the vests that create voting power, you'd delegate voting power itself. The key benefit here is that a fully-charged unit of Steem Power could be powered down without any penalty at all.
2. Remove the power-down time.
Simply: If you have 40% Voting Power, you can power down 40% of your SP immediately.
This proposal is technically difficult but would remove all barriers to holding and using Steem Power. If you power up and vote, you just need to wait for the vote to recharge before you can withdraw. If you hold a steady amount of SP, you can withdraw a chunk of it at any time. Voting Power would work somewhat differently than it does now, but could be made essentially consistent with current usage.
This proposal relies on two observations that I believe are quite solid:
Thus we can maintain Steem more-or-less in its current technical form only by reducing the aspirations for how much content it should be supporting, and moving Steem from a generalist social media presence to a subject-specific one.
1. Rebrand Steem as a Libertarian/Crypto-Anarchist think tank.
This is clearly the subject matter which holds the most interest among top stakeholders, and would retain most of Steem's core communities. This can be stretched to include open-source activism, drug freedom, metal investment, and all sorts of cryptocurrency crosstalk. It would suck for the artists, and would make me considerably less interested, but it would make Steem more functional.
2. Rebrand curation from "quality content" to "idea quality discovery."
That's a terrible phrase for it, and someone should come up with a better one. But what I'm talking about here is essentially what a think tank does: push forward the discussion of ideas related to its philosophy through rewarding and organizing the producers of those ideas. This is something Steem's existing curation/trending algorithm would do quite well, especially in a Libertarian context, as opposed to content discovery which it's quite terrible at.
This addresses the fundamental issue with existing think thanks, which is that they're extremely committed to gatekeeping. Philosophical growth, when it happens, comes from an extensive and expensive process of recruiting new intellectuals to the organization. A Steem-based think tank, operating well, would allow universal access to the arena of ideas, and rewards would flow automatically through the community's perception of idea quality.
In this situation the EIP might not actually be a bad idea.
