The short version of the argument is that allowing a lot of character customization
a) Can't fully achieve the goal of having the player "become" the character, as the gameplay and narrative of the game provide their own limits;
b) Doesn't really solve the problem of the interaction of race and video games; and
c) Limits the games, because it prevents them from using meaningful character details as driving the narrative, gutting it.
This misses the point to a great deal;
For (a) All creation has limits but that doesn't make it valueless or not an act of creation; even if the limits are that born within a game system.
For (b) it's true but character customization was never really aimed at solving the interaction fully.
For (c) not all details of a character limit the story of a game (would it really matter if Gordon Freeman was black?) and if a game is anything other than a railroad it needs to branch at some point anyway, so the branching of a game in response to character creation (see Dragon Age's multiple origin stories) is not a meaningful limit of narrative.
In longer form, his argument is full of holes in general; he starts off by begging the question, complete with passive-agressive "I'm going to get modded down for this, but" bullshit:
Now, to offend half the blogosphere offhand: For the purpose of this article, we will consider avatar customization a convenient narrative cop-out. We shall also assume that no mechanisms are in place stopping developers from writing and designing heterogeneous yet fully structured, narrative-based computer games with carefully constructed and immutable, unchangeable characters.
So he assumes the practice he's complaining about is the only thing stopping him from getting the games he wants (it isn't, but I can see the assumption as useful for purposes of argument) and then assumes the practice he disagrees with is valueless (it isn't). He even admits that in terms of narrative etc he's dismissing the value with nothing more than the word "seems":
(Obviously, there are occasions wherein the “tabula rasa” scenario is a fully motivated one, either by its ludic or narrative function, but assuming this to be a default state to be aspired to seems ultimately misguided beyond the MMO.)
As he asserts this without evidence, I'll dismiss it with little more (At very least, games in the line of Fallout or (from what I know about it) Dragon Age are clear examples in opposition to this).
He goes on for a while about minorities and gaming, nothing that minorities are underrepresented in gaming, and that the common approach of reading % of characters as a measure of this is a bit of tokenism and misses the point – that the experience of growing up white and growing up, say, Latino are different and this affects a lot of things in subtle ways, and just changing a character's skin isn't going to reflect these ways. And that making this irrelevant works against both the white and Latino's experience. This is true as far as it goes, but it really doesn't have much to do with character creation:
a) I've always thought the % studies as a quick and dirty measure of how much of the creators are working to take those experiences into account. If the numbers are heavily lopsided, then it's a sign the probably aren't; if the numbers are more even there's at least a chance they are.
b) More importantly, the ability of a trait to help someone connect with a character isn't necessarily connected with the importance in the game world. To paraphrase from a shadowrun sourcebook, “Who cares about the color of someone's skin when the guy over there is a rock with hands as big as your head?” This is even true for characters set initially in our on world (c.f. Gordon Freeman). So the race of the character could end up being meaningful for the player and not meaningful for the game world.
c) Even where it is relevant, it can be branched; this can help the narrative in a way that it doesn't work with books. Flipping back to Dragon Age – the plan for the game is you select one of several “origin stories”, which set the first few hours of the game and have impacts through the rest of the game. This allows them to examine more of the game world – improving the narrative/story of the game as a whole rather than weakening it - while giving freedom in character creation. While it does it in context of class and species rather than (say) race and gender, which are rendered irrelevant, there's no reason it couldn't be used for the latter.
Then he gets tripped up on language for a bit:
Yes, the act does resemble that of “creation” in that players apply their imagination to a restricted set of tools, much in the same way one would other forms of art, but a process of “birthing”, like Alexander calls it, it is not.
After all, the word “birth” is far removed from the tangible actuality of the interfaces to which our creativity is ultimately tied to
I dunno, people get pretty creative with the interfaces used in birth. (Well, people that aren't slashdot readers.) But I'm really not sure what he's getting at here; yes, it is a restricted set of tools for creation. And? It seems like he's trying to confuse himself with terminology while knowing what people actually mean by it:
In video games, then, we do become one with our character – at least as much as acting out a role in a play allows us to vicariously experience being an another being.
Well, yes, but this doesn't actually lead to the rest of his point:
audiences often demand protagonists to whom they can relate, whom they admire, to motivate gameplay and enhance immersion – so isn’t the best way to “get it right” to allow players to build their own.
For designers, writers and ultimately companies to seek to “get it right” in this manner, from my narrative-obsessed standpoint, is what I mean by avataritis.
This is the dualistic fallacy of the avatar: Customization may seem to offer developers and players alike a chance to mask, to separate an avatar from its perfunctory position and move it closer to the player, bridging the gap between various players of different origins, but due to the avatar’s function as a literary element, a character never does become perfectly liberated from its original environs and place of creation.
The idea that an idea (character creation) allowing a character to move towards the player rather than the creator is fallacious because it's not perfect is absurd. That something isn't imperfect doesn't mean it's valueless; customization does not always weaken the narrative (as I noted above) and even when it does the tradeoff may be worth it.
He goes on to say we can connect with people that aren't like us, which no-one has ever disagreed with. Skipping past the first 2 summary points, a note on the third:
Third, I sought to explain how offering players avatar-based customization can lead to beautification, stereotyping, archetyping and the ongoing perpetration of an established discourse of the avatar that allows companies to purport and rely on the assumption that players (or viewers) only want to relate, desire, admire or be themselves.
So I'm an elf, robot, alien, human, and talking cow-person of various genders? Alternately, avatar customization can also be used to put yourself in someone else's shoes and experience them, or just mess around with different characters in general; while avatars allow us to make characters like ourselves they don't require it. And often they lead away from it; if customization and the game are sufficiently robust, they can even encourage it, by playing as multiple characters. If you're replaying a game on a different path, you may as well make a different character as well.