Rule of Three
Would it be possible in D&D Next to use some of the tactical combat rules and ignore others, or would you need to talk them all as a package?
Yes. While we are likely to design and present rules modules in groups aimed at a particular goal (in the case you mention, to provide more concrete rules for adjudicating combat while using minis, a battlemat, etc.), we also want each individual optional rule to work on its own. For example, last week I mentioned that grid-based cover and facing are two things we're looking at for the tactical combat rules module, each of those is its own distinct optional rule. So, if you want, you can use facing, but not grid-based cover, or just use grid-based cover but not facing. Similarly, if we have a "grim and gritty" module that we put together, it might have things like long-term wounds, critical hit charts, drastically reduced healing, etc. The goal would be that each of those systems works independently of the others, but when combined they really work toward producing a grittier feel to the game. Remember, the goal of rules modules is not to enforce a style of play, but to facilitate one, and part of facilitating a wide variety of play styles is making sure that those are not so tightly intertwined that they cannot be broken apart.
What kind of actions are you looking to add as maneuvers in the D&D Next playtest?
We'll be talking more about maneuvers in the near future (as I alluded to last week in Rule of Three), but the idea is that we want maneuvers to represent more than just a basic combat action. We want things like pushing someone around, knocking them down, disarming them, etc. to be covered by our check system, and teach DMs how to adjudicate those things without having to add in heaps of very specific, exception-based mechanics that bulk up our streamlined core.
Maneuvers, on the other hand, are designed to let you do those kinds of things and make your normal attack, or do things that are more exceptional. Anyone can try to push the orc off the cliff, but the character with the maneuver gets to make an attack and push the orc for free (for 4E players, think about how tide of iron works). Similarly, there are things that just can't be covered by our basic checks system that need a place in the game, like charging, power attacking, etc. that have more exceptions-based rules. Those are the perfect things for us to put in maneuvers, because players get to opt in to the complexity, and it makes it easier for the DM to remove them, or give bonus maneuvers out, on a campaign-by-campaign basis.
In D&D Next, will a different spellcasting system likely be attached the wizard, or to a different class?
For the moment, we're looking at big structural changes (like, for instance, how a character casts spells) likely requires a different character class. While we are definitely excited about the idea of having different ways of casting spells, we also want to make sure that our classes stay balanced and functioning as intended. Plus, it makes it a bit tougher to predict how other game elements (like feats, for example) interact with a class when that class's fundamental mechanics change. In the other direction, we think "alternate spellcasting method" is a strong enough hook to give a class a large portion of its mechanical identity, and again we can really make sure that the class is behaving in a unique, balanced way if we focus all of our efforts on that class's core mechanics.