I see so many people enjoying the process of determining what others are allowed to do, and so little time improving the tools for a more attractive and fun user experience.

Most developers I know working within the Hive ecosystem are not at all focused on determining what others are allowed to do.

Maybe you're mixing up stakeholders and developers? Hive was designed with downvoting, and stakeholders do use the capability. But that has very little to do with the current software development process here. Now, it's true that people do simple queries of databases to analyze rewards of accounts, etc and post about those rewards. But, in my opinion, that's just barely software development. It's simple data mining.

The reputation system has needed an overhaul for a long time.

As it happens, building a reputation network is the main reason I decided to tie my future work to Hive. It's because Hive has a strong technical base layer and a nicely mixed community of users who believe in change rather than the status quo (note I disagree with many individuals here on various specific issues, but I think many of us are united in the sense that we think things can done be better) and I think that's a good launching point for an information rating system.

But building a functionally useful decentralized reputation/rating network is a very big challenge. I expect this will be the most difficult software I've ever attempted to develop in the past 30+ years of software development, and that's after working on very complex software in the past (computer circuit simulation software, engineering physics simulations, distributed computation systems, p2p networks, blockchain technology, etc).

If you think we (Hive developers) are not doing enough, then I think you're focusing too narrowly on the code being changed for the hardfork. From a coding perspective, the changes in HF25 were some of the least difficult things we've been working on. I intentionally kept the goals for my own team for this HF to a reasonably small scope, because we have many other tasks going on. While the curation changes will likely have one of the greatest immediate impacts for the social media applications, these didn't take us a lot of time to do, relatively speaking.

The larger focus we've had has been on building a framework for the development of Hive-based applications (I've been referring to this as "modular hivemind" as the codename, but it'll probably get a better name once it's released).

Current apps face a couple of serious problems: 1) they have to decode data from the blockchain, 2) they have to come up with a scalable way to serve up the data they compute from the blockchain, and 3) they have to properly manage when the blockchain experiences a microfork which results in reversions of recent block data (without this ability, the applications will either have to rely on older "irreversible" data or suffer from potential data consistency errors across their nodes). Solving these problems is key to easing Hive-based development and the ability to build scalable systems (systems that can support many users doing lots of stuff) with strong data integrity.

After we've completed initial development of the modular hivemind technology (I'm thinking within the next 2 months), I plan to start developing a reputation system using modular hivemind as the base technology. We're also planning to build a smart contract platform based on modular hivemind. These are both very large undertakings.