I'd like to start a discussion on the development strategy for what I see as the two biggest core "missing pieces" of PMMP as it currently exists. 1) Enchantment/XP: The ability for players to earn XP through the normal Minecraft methods. Possibly adding to these methods (PvP). The ability to "spend" XP through enchanting objects. 2) Redstone Machines: The ability for players to build complex Redstone machines. Management of server-side Redstone "pulsing" in a way that focuses on occupied parts of the world. Performance balancing of this against more critical player "ticks". This functionality was successfully "prototyped" in various spoons, and has thus become an important part of the gameplay for many servers. Sadly the functionality broke down to a high degree with the release of MPCE 0.16.0 in October. My understanding from various posts by dktapps and others is that this code was extremely ugly and will need to be carefully re-written. It is understandable that we haven't seen fixes to the spoons, as all core developers are likely focused on PMMP and the major changes necessary to support the pending release of MCPE 1.0 (aka 0.17.x). For this reason I haven't opened any "issues" on the spoons' GitHub. I am curious whether any core developers have decided to focus on these particular aspects of PMMP? It seems like an interesting area. Is is possible to break out individuals/teams to focus on this while others are focused on cleaning up the code and core stability? Maybe this thread is a bit of a "call to arms" for any skilled PHP developers looking for an opportunity to contribute to PMMP in a meaningful way who have not yet done so. While I do not have the skill to help with coding this; I'll be happy to do methodical testing and bug reports once the new private branch (mcpe-1.0) is publicly released on GitHub. This post is not meant to be critical; only to start a discussion on something I haven't seen covered on the forum thus far.
While PMMP is trying to catch up to the Rapid-fire MCPE updates, many plugin developers have taken it upon themselves to make plugins implementing features that the PMMP team just may not have time to do right now. I think it is a really cool idea to have teams to work on different areas, and I would definitely support it if anyone ever started working on it. Spoiler: On the subject of adding new features to PMMP... I guess this isn't on the Subject of Redstone/XP (But it still has to do with adding features that should be in Vanilla PE), but me and @Sandertv are working on reviving a dead plugin to add Mob Support/Spawning/AI to PMMP ( https://github.com/RevivalPMMP/PureEntitiesX ). We also plan to do other plugins (Maybe implement Redstone and XP/Enchanting as well?) I may make a separate thread for this though.
It is a waste of time to implement features that should be in PMMP itself in plugins. If you would like to add these, put them into PMMP directly. The only trouble is that most developers fail to properly implement these things (especially mob AI and redstone). Thus will waste their time as we will not merge poorly implemented pull requests.
@Magicode Your project looks very cool; I'll definitely check it out. What @SOFe is saying makes a lot of sense. It is, however, challenging for developers to work on major features in the main PMMP code base at the moment. I would say the main challenges are: 1) The most up-to-date code is in the private branch (mcpe-1.0); and there may be many dependencies on code that is changing for 1.0 2) The level of quality demanded by the core team is very high for core features; and few developers -- especially new ones -- may feel they are able to meet those expectations. For the second challenge, I wonder if a team approach might help. A leader from the core team could mentor a few less experienced developers on an example project like "Enchantment/XP". The lead developer could build the main outline with extremely basic functionality (enough to support a handful of enchantments). The less experienced developers could focus on testing every possible enchantment case, and adding code that mimics and expands on the basic functionality. Maybe there could be a behind the scenes chat for the team that would allow mentoring without the embarrassment that comes from submitting a pull request and getting negative critiques. The first challenge is more difficult. At some point the benefits of keeping that branch private may be outweighed by the benefits of allowing wider development. I can, however, understand the benefits of keeping a super-alpha version private for the core team. It allows them to focus with less noise.