Well, I just read a post about some of the team aiming for a September beta. Now without being too blunt what I think about that notion, we should as a team have some goals in mind.
But firstly I want to make sure that we remember that we are to have fun with this project, rather than be all consuming for a date.
What is our definition of a Beta? In most cases its used as a platform for introducing it to the public, so that developers can receive feedback and bugs to fix. I think this is an important point; will the team be ready to have a version that we can definitively say is a version that we can fix bugs and not be introducing new ones?
Under this definition, I think the bare minimum requirements would be:
Quests finished, likely with bugs. We want to ensure that we do not go in cycles of bug-fixing with each release. We do not want to test from scratch because we introduce so many new factors. The below elements will have a major impact on that.
Models and Textures 100% replaced, unoptimized. Lets say instead we agreed that we have a goal of 50% replaced at Beta. Are we truly testing the game? Are bug reports legitimate or worth acting upon?
I personally would be far more comfortable if we are releasing anything "official", that we don't distribute the old assets anymore. We need about 600 further models replaced (excluding armor, weapons and creatures). Even if we were to receive 10 models per week we are still looking at 60 weeks. This is without any type of further work to insure quality and optimization of models and textures.
Collisions unoptimized or optimized? Unfortunately all the replacement models are using full detail meshes for collision. Creating low poly versions of collisions is a major project. Again are we truly testing the game unoptimized? We can probably rollout with more quality control if we have a dedicated member or team, but with thousands of models, I cannot stress the additional time involved.
World 100% navigation-meshed. Is the world to be navigation meshed? If so that also in theory means that landscaping should be finished as well. Otherwise we are falsely testing.
It is unrealistic to have all the interiors cluttered and thus perfectly navigation-meshed, so I wouldn't include this as fully testable on Beta. Quests might mean that specific locations need to be finalized. But we have disclaimers that bugs will occur in interiors with NPCs and Creatures.
I’d estimate that it would take 1 person exclusively about 6 months to navigation mesh the outside world. ArchieVision worked on Dragonborn underwater navigation-meshing so we can get a better estimate from him.
Installer that is robust and easy. I'll shoot myself in the head if we spend our days instructing individuals on their problems with getting the game going.
I'll point out some of the things that I think under a tight schedule would not be included; distant objects (LOD), gameplay/mechanics/balance wanted, special effects, special lighting or weather, interiors cluttered, creatures with their animations fully completed, armors, clothing and weapons will still have some placeholders. The list of course is much longer, but this just to illustrate.
You can see even if we have the bare minimum in for Beta (for a forced date), we leave out so much of the elements of the game, that we expose ourselves to major cycles of re-testing.
If we have 3 fundamental goals for the project; Release, Quality and Optimization, Features, then we can use a distribution triangle to give each a weight, so that we know as a team, where our priorities lie
Release, the weight to delivering the game as soon as possible.
Quality and Optimization, beautification and realism of landscape and interiors, models and textures and getting everything as streamlined as possible for performance and esthetic reasons.
Features, are things like the mechanics of the game. Gameplay, spells, crafting, etc. How many extra things are added like, special effects (ash/blight storms), variations in creatures/environment, special combat (underwater/spears), voice acting, additional scenes to name a few.
So if we had a game that had all the features we wanted and the quality and performance was that of a AAA game, and we gave ourselves as long as we wanted to do it, the perfect game as such, this is what the weight distribution would be;
.jpg)
If we are wanting to have as an early Beta release as possible, my interpretation is more like there would be few features and not much weight to quality and optimization, with the delivery of the game having the dominating factor.
.jpg)
I'm fine with however the group decides what sort of game we want to deliver as a Beta, but we must decide where we want to be within this distribution and what our actual definitions are for the term.
This will decide what we leave out, aim to put in, and the quality of the Beta and/or what we aim for users to test.
I cannot stress the importance of what do WE gain out of a Beta? How does it benefit the project progressively to a finished game?

I understand the not all the
I understand the not all the team will be around forever, that is why we need to plan ahead as a community project.
It is very feasible that members at the end of the year are completely different to those who are working on it now, including myself.
So it's important to outline goals for the community as a whole, and have realistic goals and contingencies for levels of contributions for the next few years.