Industrial, physical uses of gold account for a very small fraction of its price. Gold coins originated as IOUs of some King. Bitcoins are nothing more than decentralized IOUs for the "real money" traded at bitcoin exchanges.
I agree that the physical use of gold does not explain its price by itself. Two things though: First of all, the physical use of gold is undeniable, and it's plausible that a sort of "bubble" -- perhaps not quite the right term given how stable that bubble is -- bootstrapped from this physical value.
Second, from what I understand, the first gold coins that were intended as currency were made of gold precisely because gold was already valued for other reasons. I do know that historically speaking, the value of gold coins was usually unrelated to the price of their gold contents, so it is true that they played the role of an IOU. However, (a) there is a reason why governments chose gold rather than a different material (psychology based on the pre-existing value of gold), and (b) throughout the ages, gold was also used as a payment-commodity in trade.
That latter use is probably overstated by the goldbugs, but it would be wrong to dismiss it entirely.
> I agree that the physical use of gold does not explain its price by itself.
Gold was around production costs ($250) before the "general commodity bull-run 2001-2011" and is now back at around production costs (nowadays roughly around $11xx - $12xx, many opinions about this and different ways to quantify but production costs are in this range these days, no longer in the low hundreds as 12 years ago).
Gold is very similar to Bitcoins because production is hard. The government can't create it any cheaper than anyone else. This rarity is what helps stabilize its ability to hold value.
Gold coins are not the same, and are similar to IOUs as you mention. But gold does not base its value off the fact that at one point it was used as a currency.
