Hi Kevin, I have been thinking about this quite a bit lately, and even wrote a post on it. Not sure if I am correct but here is what I think:

I think that even if we implement the improvement you mentioned it may not be enough to drive the price of STEEM in the long run. I think the STEEM price is currently dominantly governed by investors speculating on the price of STEEM. They wouldn't know all the payment mechanics in steemit, they wouldn't know what the benefits and issues with it. They might even be just technically analysing the hell out of the price charts without knowing the fundamentals of it. I think this is reflected from the fact that we have yet to be able to free ourselves from following the trends of BTC.

In order for STEEM to go up in the long run independent of other cryptos especially BTC, there needs to be a organic demand that is not related to price speculation. Unfortunately, improving steemit may not be one of them. And that's because most new content creators are here attracted by the fact that they can earn money from creating contents - for free. Not to invest money in it to self-vote. Similarly, even increasing the curation % to 50/50 is not going to attract new demands of people buying STEEM just so that they could get this curation reward. That's because if it is just a passive income people want, they can just put it in a bank or traditional financial products. Putting money in STEEM for the sake of curation reward is only logical if the person speculate that STEEM will go up in the long run. Again, something dependent on, not drive, the price of STEEM.

Of course, we could also try to slow down supply by retaining users through improvement of steemit. But my question is then again, how much of the powering down is due to they are frustrated with steemit, or that they just see that the crypto-wave is over? I guess this will not be easy to find out.

Anyways, I see the improvement of just steemit and its mechanics as you outlined can slow down supply by retaining users, and potentially increases the number of users on steemit. To me this could create a demand for STEEM if we have a large enough user base for true advertising value. However, with most user's attitude towards self-advertising, I am not sure if this would be enough to attract demand. (Don't get me wrong, I hate crap trending posts myself and downvoted a particularly bad one, but we have to face the fact that these people would have bought STEEM to achieve that and therefore, would have helped pushed STEEM price up, even if just a little).

The only other option we have left is to make STEEM into a self-sustainable economy. Non-STEEM users need to be attracted through goods and services that can only be bought with STEEM, and that there are enough good and services that once you own STEEM, users do not have to sell STEEM to buy what they need. Only then would STEEM's price go up.

A good example is Steemmonster. If Steemmonster becomes such a good game and becomes viral, gamers will flock in to buy STEEM just so that they can play or own these cards. That will be a healthy demand of STEEM. I think this may be why Ned and his team is concentrating on SMT too. If we for example, have a photography SMT where people can buy and sell photos, and then they can use a different SMT to say view exclusive contents (blogposts, videos, etc) that can only be bought using SMT, then eventually we might achieve a point where once people buy into STEEM, they can get what they want by changing into different SMT without needing to sell STEEM. This would then truly drive demand of STEEM without being tied to the wimp of investors.

Sorry for the long reply. Just thought that this might add to the debate. Ideally, steemit inc should work on both improving steemit and SMT in parallel. But may be they do not have enough resources, in which case it seems to me that SMT development is much more important than fixing steemit for the good of the whole blockchain.