Neonchameleon
Legend
You absolutely can if you want, but people don't have to accept your model or your definitions if they fail to reflect their experience at the table. You yourself point out we know very very little of actual play (we don't know how much char-op boards reflect widespread use at real tables for instance). I don't know what basis there is here for a working model of RPGs (and I personally haven't seen one that I have found useful for design or for play). My concern with models and definitions is they so often seem to be about getting the hobby where folks would like to see it of (don't like min-maxing? make a model of RPGs that excludes that as a valid style of play. don't like story? make a model of roleplaying where story is the antithesis of roleplaying). We can do the same with definitions. Once again this is exactly what Wick is trying to do. Clearly he favors some kind of RP-heavy campaign and is redefining RPG to exclude groups that play differently than him (even though I think most of us know a huge chunk of people play the way he defines as not roleplaying). What is worse, he also clearly doesn't have much love for D&D and so he uses definitions to claim it isn't an RPG---which is an insane claim to make). Stuff like this is exactly why folks are so wary of models and people trying to control definitions in the hobby.
What I do know is players are pretty diverse and to get a game off the ground you need to please 4-7 people at the same time. Give me a model that allows me to do that, to sell lots of books and make lots of gamers happy, and I would happily use it.
The problem is that none of these terms ever gain traction, so we're stuck going back to the old GNS theory just because that's the only thing that everyone has heard of.
And even Ron Edwards has given up on that.
My take is that it's simple. Roleplaying games are broader than any model can be (which is where the Big Model failed - it devolved into a theory that explained the presence of invisible pink hippomen and square circles). And as such models will not actually cover the spectrum of games; all they can do is highlight things and lead to a better understanding of a subset. (And from this perspective GNS was a success - the S part was a failure, but G was useful as a pushback against the "Rollplaying not Roleplaying" crowd and a focus on N (which was most of what Edwards and the Forge were interested in) lead to interesting things).
A contour map is a very useful thing as long as I don't confuse it for the whole territory.
but, I don't see why that's any better than using 'Roleplaying Game' for only pure roleplaying games with no storytelling elements, and 'Storytelling Game' for roleplaying games which include such things.
It's not as though there are any pure Storytelling Games out there, devoid of roleplaying elements. (Are there?)
I'm pretty sure collaborative writing qualifies, and there are probably more people doing that than RPGs.
And that's what's wrong with trying to use Storytelling Game for Roleplaying Games that also involve at least some author stance. That collaborative writing is numerically a much bigger field than tabletop RPGs. And Polaris and Kingdom from within the RPG community I've both heard described as storytelling without roleplaying per se. (I really must get round to reading Kingdom/Microscope).
Using "storytelling game" for a subset of tabletop RPGs is like using "Football" for a hockey rules variant in which everyone is also allowed to kick the ball. To me this is ridiculous, and the only purpose it appears to serve is to attempt to exclude people.
That's how I'd define it to. The degree to which a game is a storytelling game is the degree to which players have out of character authorial control. The extreme is collaborative writing, where all parties have authorial control, and which, contrary to an assertion above, can be done as a game. If there's another way it's used, I haven't seen it.
Possibly so

"Pawn stance" vs "actor stance" is another sometimes disputed term. For me it's just 3rd person vs 1st person play. If I think and act in terms of "I do this" rather than "my guy does this" that's actor rather than pawn.
It's a bit more than that. In Pawn Stance play, following the logic of your character's characterisation into making what you know to be bad choices is simply bad play. In Actor Stance play picking good choices against the logic of your character's characterisation is known as metagaming.