The current devs believe that we should optimize existing code as much as possible before enabling SMTs. SMTs could potentially increase the number of blockchain transactions quite a lot, and we want to be sure that we are ready for that possibility.

Also, running a full Hive API node is currently quite expensive, so we still feel the number one priority is to lower that cost, so that anyone with an interest can afford to run a node. This makes the entire Hive ecosystem more decentralized (which is useful to any user who doesn't want a platform where they can be effectively be censored by a single organization).