Roadmap updates / priorities


#1

So I was browsing Trello briefly to get a sense of where development was headed. My feeling is most items fall into a few broad categories:

  1. feature-oriented (e.g. RLR, hair, terrain, …). These are easy enough to specify and likely to have a fairly local impact.
  2. platform-oriented (e.g. MAC support, VR, …). These are also relatively easy to specify but depending on how leaky the abstractions are might have somewhat broader impact. (I would not expect too much however as Xenko abstractions appear well thought-out).
  3. architecture-oriented. These seem vaguely defined and the precise intent is only clear to the original devs. These are also the ones that seem most likely to have a large impact and are also the cases I am most interested in as Xenko’s appeal [to me] is the exceedingly clean design/implementation. Some items that stand out as -possibly- in this category:
  • Graphics Compositor Improvements
  • Entity Component System Improvements
  • Scripts and Tasks Improvements
  • Plugin system (clearer in overall intent but I can imagine many possibilities here)

@xen2:

  • Do you foresee work in these “architectural” areas or, barring that, can a bit more detail be added to Trello (or elsewhere) about exactly what these would entail should someone else want to address them?
  • For me one of the highest priority items is the plugin system. Especially now that Xenko is MIT the alternative is that people fork and modify for cases that might otherwise be handled in a more modular fashion. Unfortunately, this seems to me also the primary area that requires someone with a deep understanding to produce an optimal design. Any thoughts here? e.g. Is this on the radar / how big a undertaking is this / do you already have a design in your head?

#2

Pretty good analysis.

First of all, please note that Trello board is totally outdated (I changed the board title a few days ago to reflect that).
I started an effort – where everybody is welcome to help – to setup Epic using ZenHub on Github (https://app.zenhub.com/workspace/o/xenko3d/xenko/boards?repos=136570231). Still plenty to add.
Those Epics already contain or will be updated to contain details about why/what needs to be done.

Once we have a good enough board on Github/ZenHub, we can present it nicely in a high-level & easy to understand version for everybody to feedback/vote on (maybe by attaching those epics to version milestones?)

I also agree with you that plugin system https://github.com/xenko3d/xenko/issues/13 is probably something that needs to be dealt on early.


#3

Heh heh, I somehow managed to completely overlook the red flashing “obsolete” sign. Serious forest for the trees issue here…

Sounds great! Thanks!


#4

What would be the best place to inquire or discuss future features for the engine? The ZenHub? The features I am most interested in are a networking layer, however I’ve already mostly written one of my own.

Assuming a network feature would be far off and take a lot of time (likely) a feature I would most like to see is a way to run the game in a headless mode essentially. I want to use all of the physics and entity services on a server, but would not want any graphics or sound output.

I am very interested and willing to help but I don’t want to do things in my own bubble, but I’ve not had much experience working with an open source project :slight_smile: