I think its because a lot of people on HN run businesses themselves which could or are negatively affected by piracy, because software is easily pirated - so they (we?) are naturally hostile towards the idea that we're doing business wrong and that our hard work should be free.
Honestly, I don't know what I think of it. On the one hand the pro-piracy people do make some sense, but on the other hand, a world of free everything is impossible to sustain and messy to do business in. Unlike OSS people like to think, support is not enough to.. err.. support all businesses. It works in some cases, sure, but it doesn't scale very well and for some businesses simply doesn't make sense.
The author is postulating that the price of a good is determined by its cost of production, which is essentially the Marxist theory of value that has been discredited for over a century. Else, he is proposing that the demand curve is completely parameterized by cost of production, which is a radically new idea that I sadly don't see much promise in.
I agree with you that demand isn't dependent on costs always - and I don't think the author is either.
He's saying that demand is still high, as is supply. But because the marginal cost of each unit is so low (on the postulation of infinite copies), that piracy is how the price is "fixed" economically speaking.
This isn't a pro-piracy position, but if you see it as each copy of the game had to be physically manufactured piracy would drop to zero - otherwise it'd be called theft.
That the marginal cost of copying media is zero is a pretty mundane observation. The author goes beyond this to hypothesize that demand moves according to some abstract "fair price" (that happens to suit the author's semi-Marxist definition of fairness), and that piracy only occurs when the market price is set above this value. This second point is the only novel bit of the author's post, and it's just not very good.
Is it because the argument is fundamentally flawed, or are we seeing a little groupthink here?