The Letter from the Chairman says the quarter three release won't be at the end of September like many may have expected, because of its proximity to CitizenCon, which is just 10 days later. [Can you talk about that?]

Well, when we were talking about quarterly releases, we didn't think this through fully. We just thought, "Oh, we're going to do one every quarter, which is cool." But when we started to look at our planning, like, okay, we're going to CitizenCon and we obviously have some stuff that we want to show that won't necessarily be in 3.3, but some of the stuff will be kind of 3.3 stuff. And CitizenCon is always quite a big endeavor to pull everything together. And getting a release out is also a pretty big endeavor to pull everything together. And we're just like, okay, it's crazy. We're going to have this massive rush to get everything out for September, and ten days later we also have CitizenCon. And so there were quite a few people on the development side that were about ready to have meltdowns, like, "How am I going to do all that stuff?" So we took a look at it and we decided that it would be better, in future, our plan would be that CitizenCon and the third release of the year should always kind of align with each other, and you make CitizenCon the celebration and the beginning of the release of that. I'm not sure we'll be able to go live at CitizenCon, but we'll probably definitely be able to go full PTU to everyone by CitizenCon, because in some ways you want to go to CitizenCon and be kind of like, "Hey, here's what 3.3 is," and do stuff like that. But that's our plan, just to align them. And if we'd thought about it when we first said, "Oh, we're doing the quarterly schedule," we would have said that. So this is us just looking at it and looking at the logistics. Because the other thing is releases and things like CitizenCon take a lot of focus. And so we're trying to also make life and distraction for developers fit inside certain boundaries, so they're not always feeling they're getting pulled in 20 directions.

And to be honest with you, I've seen people complain, "Oh, look, the AtV doesn't have as much content." Well, it's not like we're deciding, "Oh, hey, haha, we've got loads of content, we're just not going to show it to you guys." But what happens, for instance, when we release 3.2 or 3.1, at that point the team goes into full polish and bug fixing mode, so they're not doing feature development, and then it takes them a bit of time to get back into feature development. Of course, we have some teams that are off cycle working on features that aren't coming out in the current release, so we can show some of that, but it definitely is the case that there is a bit more of a content drought for us to show things as works in progress for AtVs around the release period than there is once we're much deeper into the release period. And the reason why last year was so great in terms of AtV content was because we weren't releasing anything. And so we had loads of stuff to show because we were working on all these features, whereas now everyone's getting to see the features and actually play the features, which I think ultimately is a better thing. But that doesn't mean you're going to get 20 or 30 minutes of AtV full of beautiful, juicy B-roll all the time, because that's, A, a lot of work, and B, it requires everybody to be working on new features that they can show that for. With our new release schedule, we've always got quite a decent amount of the team that's working on the bug fixing and the polishing of getting the current release out.

The analogy I've always liked is it's a lot like statting your character in an old pen and paper RPG. If you put 20 points into strength, you've got to take them out of intelligence. If you put 20 points into charisma, you've got to take them out of wisdom. You can't have 20s across the board. If you want more frequent releases, we have to start pulling points out of...

Yes, that's the truth. I think the last time we were on, we talked about how there's always a trade-off, right? Yes, we would like to do everything, but even with the amount of resources we have, which is probably one of the biggest... There's a few projects that may have more people working on a single game, but there aren't many of them, right? So we've actually got a lot of resources, but still, even with all those resources and all the things we want to do, because we are literally trying to simulate a whole universe from a first-person view at incredibly high fidelity, you always have to make some choices. And some of your choices are things like polish or bug fixing. A great example is the turret stuff, right? So the turret stuff still isn't what we want for 3.2. The turrets need to be better. It's a major part of multi-crew. But they schedule it out, do the sprints: "Okay, yeah, we should have the turrets working." And then you're like, "No, they're still not good enough." Well, that has to now go back into a schedule somewhere, which will bump that team; now they won't be able to do something else they were scheduled to do until they finish the turrets. And so those are the kind of trade-offs you make, where you sometimes say, "Well, what is the most important thing?" And what we're focusing on mostly now is making sure that the current game experience is getting as good as possible. And it's going to take some time to dial that in. But that does mean that sometimes you push back some other stuff. So we haven't finalized our going-forward roadmap, but in the roadmap, accidentally someone... no, it came out last week. Yeah, so I didn't publish the QA staging; they published it, and we're still trying to vet some of the stuff properly. But that's the case of, well, some of these teams were going to work on, say, some of that profession stuff, but they haven't, because we've had them working on polishing the turrets or some other stuff that we thought maybe we would have had done or dialed in earlier, but it's taken longer and been more difficult. And so that's the ongoing aspect of managing it, but I do think the quarterly releases help you do that, because it gives you short-term goals, that you have to get your stuff out and get it in, and so people can see tangible progress, and the team itself can see tangible progress, rather than just all working on a big mess of stuff and then doing a release once a year or whatever it would be.