Actually on this topic i find the term "New Game Experience" more matching than "New Player Experience".
Any developer is free to ask John Smedley, President of SOE (Sony Online Entertainment) how introducing such a "New Game Experience" works out, especially if the majority of the current playerbase opposes the change.
@Jade Viper:
Nice example with your Lobsters, but not really applicable here. I can easily counter your example with my own working experience, i've spent almost a decade in QA IT on semiconductors. Among our products also were sensors for the automobile industry, including sensors for tire pressure, airbags and electronic gas pedals. Quality control on those was essential. (Nobody cares so much if one in a million sim-cards is broken, but for some reasons, people get really picky if they have the one in a million car which doesn't brake and their airbag is not working...

) {Edit: Actually it's their relatives who usually get angry. The owner of the car tends not to complain any more... }
The scientific approach you describe would be to put plenty of those elements into cars, see what happens and analyse the result. For some reasons, we didn't go this way, but analysed anything beforehand. Testing them at very high and very low temperatures, in moisture, while shaking and while applying physical force to their connectors was standard, along with some "more esoterical" test criterias, and we did this due to past experience of what went wrong with older product lines.
This in my book justifies our approach here: many of us here know the issues with third person view from past games. Where we had the choice of either using a disadvantageous camera position (first person), reducing our fun by playing it as yet another third person game or quitting the game completely. As both the first and second option were detremential to fun, only the most diehard fans did not go for the third option after a while of suffering.
Now, i refuse to be a "3rd person is always evil" fanatic, but i also refuse to be a religiously faithful to the developers and believe them that they can implement 3rd person view without creating issues. Unless the developers manage to somehow alter the laws of geometry at large, it is impossible to implement 3rd person view without giving a larger field of view than 1st person view. Thus a forces split of the playerbase seems to be the only option, but going this way will always stay problematic.
There were some "suggestions" on how to "fix" the issue, including black bars on the screen or fading out mechs if they would not be visible yet from 1st person view. Unfortunately i am absolutely certain that exactly the assumed
dimwitted less knowledgeable new player, who supposedly would need the 3rd person view, would be completely unable to comprehend why these limitations would be in place. All he would see would be that the game was "hiding" enemies from him and that's the reason why he lost. (And he would not be wrong, the game actually would hide enemies despite they'd be in his field of view. ) Thus the game would be bad and at fault, both in the players view and also in truth. A guaranteed way to get rid of a potential new player, and thus absolutely not an option to use.
Unfortunately all suggestions i read which might help the issue of 3rd person being a significant advantage by providing more information have this or other obvious problems, so splitting the playerbase might be the only way to go. But even this split is very likely to produce a plethora of additional problems.
One potential example would be: streaks or lasers might need to be "nerfed" in 3rd person view, as they are bound to have a bigger deployment envelope there than in first person. After this change, some other adjustments might become necessary. How long will it take, till the game is either optimized for one playstyle or equally suboptimal for both? Or will the game split up into two different branches for two different playstyles? Does anybody really believe PGI can afford such a step?
Thus on the assumption that the whole thing really is for "educating the new player", the most practical solution was already mentioned many times in this thread: a really good tutorial. During the tutorial, an outside view of the mech could be used, but i doubt that it would be necesary.
Anyway, i also think the whole "people don't get the torso twist" stuff is a strawman. We had tanks with a very similar steering concept to MWO in Battlefield 1942. The mouse controled the turret, wasd controled forward and backward movement and turning of the tanks hull. That was 2002, and the Battlefield series is still alive with the same controls. (And even before that, some HL mods built vehicles on a similar pattern! Even in a CounterStrike map, there was a driveable Jeep with the same controls. )
So if players can handle the controls in the Battlefield series since over a decade, i doubt that the potential playerbase if MWO is unable to grasp it, either. I very much doubt that MWOs playerbase is moronic at large, in the unlikely case that there actually is a problem with people not getting the controls, it's clearly a communication issue and can be remedied with a proper tutorial, which for some reason is very low on PGIs priority list.
Edited by Sylow, 30 November 2012 - 07:45 AM.