Originally Posted by J_P_A
The fact that MS chose to require the Kinect to be connected without any compelling gamer related uses makes me question their intentions (although I do have a theory or two). Based on their botched messaging and failed attempt at DRM, I don't think MS deserves the benefit of the doubt either. For my part, I'd like to see another internet s&*t storm over Kinnect, and either force MS to drop the connected requirement or force them to show us why the thing is useful for core gaming. Should be simple, but it really doesn't look like MS has a clue about why we, the gamers, need it.
And to keep saying things like, "this is the XBone MS wants us to have, if you don't like it, go buy something else," is silly. We've already seen that MS is willing to make changes to the console. If enough of their potential sales sit on the sidelines, they'll change this too.
The question is: Why would you want them to?
Knowing what we know, why is it so bad to have consoles that offer different experiences instead of just different specs? We already know that on paper, the PS4 is perhaps technically superior to what Microsoft has designed... so say you got your wish and Microsoft threw out the Kinect baby with the bathwater and offered a then-incomplete Xbox One for the same price as PS4. Why would you still buy the Xbox One over a similarly priced more powerful console that's going to have most of the same games?
So let's play retroactive armchair software engineers tasked with removing Kinect from the equation and look at some of the choices we would have to make going forward, keeping in mind that we are already altering our previous work on the back-end regarding DRM:
1. We know Kinect is what's being used for the Upload Studio functions. We also know that the controller design is locked, so we can't add a "Share" button a la PS4 and since current games tend to use all of the available buttons on the controller, we don't want to retroactively dedicate a button to that function. How do we address being able to save a clip of a multiplayer game or start/stop recording without pausing the game? We can't. Therefore, we might as well remove that feature.
2. You've designed the system so that controllers are no longer assigned as they were on 360, but assigned to a gamertag based on the location of each controller's unique IR emitter in relation to their biometrics, including removal of the player 1-4 lights on the controller itself. How do we redesign the system's assignment of each controller? If we go back to the method we used on the 360, we no longer have a visual representation of which controller is assigned to each player.
3. The device integration for television depends on the Kinect's presence as an IR blaster. Can we alter the functionality of the controller so that its IR emitter currently used for identification can also be used as an IR blaster so that we can maintain that functionality for non-Kinect users?
4. Certain features we have are tailored for the Kinect, such as the gesture to shrink the main window down or the ability to snap in other apps. Assuming the guide button still activates the guide in a similar fashion to the 360's dashboard, we can go back to the "press Y to go home" method to shrink to the home screen. How do we then allow for snapping in of apps without ubiquitous voice commands as the method?
5. The question then becomes that if it's easier to provide those services via voice commands over a microphone instead of Kinect, how do we make that work reliably? As our voice recognition is set to the known frequency specs for Kinect, we would likely have to tailor voice commands for a specific spec of capsule mic, such as the wired headset. How much work will it take architecturally for us to reroute voice commands from the hard-wired Kinect with a dedicated CPU core for processing to the WiFi direct input and the API for voice chat via controller?
6. Since packaging is already under way, how much will it cost us to undo all of that and include a wired mic instead? Alternately, since we could offer both packages at different price points, how much lead time is required to prepare marketing/packaging for the system without Kinect and with an included microphone? Secondary to that, how much time do we require to create packaging for the Kinect itself for those who wish to add it separately?
7. Developers up to this point have been treating Kinect as a known input method and part of the available SDK with the assumption of 100% install base for the platform. For our launch partners who have already implemented Kinect functions, how can we modify the SDK so that those functions can still be maintained via alternate methods without breaking development already under way?
8. We launch, in theory, say mid-November. Launch titles have to go gold and RTM at the latest perhaps 3 weeks before launch so they can be pressed to disc and packaged in time for launch. Let's say we have to provide finalized code to devs 4 weeks in advance of that so that they can finalize testing and optimization of their games, and we can update the core OS at launch with the other functions. This puts our timeline for provision of the launch SDK at the last week of September. Can we accomplish all of the above in the next 9 weeks?
Now, keep in mind that I'm NOT saying any of this is impossible to do, given time. But with them already under a crunch just to provide what they already have to before launch, why would anyone want them to?
People keep looking at this in a very simplistic way, as if it's easy to undo all the design choices they made before announcing the system, but Microsoft clearly have reasons to try to offer this as a platform instead of the console-only notion that anti-Kinect people have based on their experience with the previous and admittedly flaky tech (none of which is actually being re-used here except parts of the mic array system). So... why is it easier for people to demand that they undo everything rather than have a little patience and see how all of this will actually come together in practice? Because they still have time to show us more before launch, if people relax and let them. I know that asking people to relax on the internet is like asking them not to rubberneck when there's a wreck in traffic... but rather than play armchair system architect from the sidelines, why don't we let them present what they're offering first before we judge it? And if what they present doesn't suit our needs, consumers have several other options to choose from, which is an awesome thing. Options are good. Trying to make everything the same... not so much.
But hey, that's just me... I know I'm not changing anyone's mind here. I just think it needs to be said, even if it's ultimately masturbatory.