The problem with gold is that it's easy to find 20 reasons to be bullish or 20 reasons to be bearish (whether sober and rational or paranoid and conspiratorial), but rarely do any of these reasons point to an appropriate valuation.
There's Asian demand, the draining of Western bullion bank reserves, decreased ore grades on average worldwide, Chinese moves to weaken the dollar as a reserve currency.... I don't know, any number of reasonable things.
So do these factors imply that gold should be $1700/oz? $17,000/oz? Maybe they're already priced in at a too-high probability, so gold should be $700/oz? Without any rational criteria for valuation, gold "investing" is just speculation. Because there's no DCF analysis to be done, you can make pretty much any claim about price that you want and it's almost impossible to rationally argue with it. I think this is why the asset class attracts so many cranks, conspiracy theorists, and doomers who can't be bothered with things like probability weighted cash flow analysis and would rather just argue from the gut about which way the wind is blowing (for some reason, usually up from the fiery pits of hell).
Nothing wrong with gold as an alternate currency (albeit a volatile one), and it may well have a place in a balanced portfolio as a diversifier, but arguing around in circles for hours about whether it should go up or down is more or less an exercise in futility IMHO.
Oh, and the whole "fiat currency" argument is a straw man as far as I'm concerned. Of course paper money is a bad long term investment. Of course gold will outperform it over the long haul. Maybe paper money will devalue like clockwork at 2%/yr, maybe it will experience periods of faster devaluation. I don't know anyone arguing differently. That doesn't mean gold still isn't a loser.