- Are we lacking manpower in any way? Or is everything moving along as it should?
On the OpenMW side everything is going fine. We are munching through the last few missing features and bug fixing progresses rapidly.
The OpenCS side is more problematic. We still have only two people working on it. That is not enough. Furthermore there is no progress on the icons for OpenCS and the manual hasn't been updated in ages. This is increasingly problematic.
- Are we stuck on any challenges? Are any of the remaining features proving problematic to implement, or have they just not been tackled yet?
Nothing major as far as I am aware of, at least for OpenMW.
Regarding optimization:
- Do we have any manpower for this? From what I gather profiling and optimization takes a slightly different set of tools/skills. Do we have anyone on the team interested in working on this, or do we need to get the word out?
There are two sides to this. The non-rendering part and the rendering-part. We already did some profiling and optimisation on the non-rendering part. There is probably more to do, but we don't have anyone working on it right now.
The rendering-part is a bit more complex, because it depends significantly on OGRE. At some point we will have to make a hard decision: Wait for OGRE to improve our rendering speed or get into more nasty optimisation techniques (which may become redundant when OGRE 2.0 releases) and do it ourself. We aren't at this point yet and I would like to leave it at that as long as possible. After all it is possible that OGRE will have made enough progress by then that we do not need to go down the less attractive route.
Regarding refactoring:
- I know this is a terrifying word to throw out at this time, but are we at a place where any refactoring would assist in feature implemention, optimization, or bug fixes? If so, should we have any discussions about delegating any refactoring tasks?
Absolutely not. A small bit refactoring here and there is fine. But derailing the development by getting into a large scale refactoring effort now would be a bad thing.
For all I know everything is just fine and this post is completely useless, but I wanted to get some visibility for all of us lowly forum-lurkers.
Yes, everything is fine (except for the lack of man-power on OpenCS). In fact I will post the planning for the OpenMW 0.31 release today.
Edit: I just found a really fundamental bug in OpenCS that has been sitting there unnoticed for a long time (deleting of referenceables). And apparently nobody noticed. We definitely need more testing. But that means we need to get the icons and the manual sorted out first, because else most people won't know how to use OpenCS.
