Hacker News Reader

Comment by galangalalgol | Hacker News Reader

Not that crazy. Steel was the same at first.

Wild guess is that the dopant that creates wells doesn't always end up where it should. The paper that claimed superconductivity in layers of graphene at very particular angles also seems to be very sensitive. A similar one claimed graphene with alkanes was observed to superconduct. Perhapes whatever impure hydrocarbon they were using held the sheets at the perfect angle. All the quantum wells these things are claiming to rely on seem terribly difficult to arrange perfectly enough to work consistently. Assuming any of them ever did.

View on HN

No indeed. Lots of this in our history. You can't do most of electronics without semiconductors. But, if you have no idea what's going on you can make some rudimentary electronics experiments work - unreliably - without knowing that - e.g. the "Cat's whisker" crystal radio technology. The reason this actually works is because it's a semiconductor, but since you don't know what those are yet, you just know if you fiddle about with a fine wire and certain types of crystal, sometimes it does what you wanted, and if it doesn't keep fiddling with it until it does.

I'd imagine early history of sugar products is the same. Today you can precisely control the temperatures and so you can engineer getting exactly the desired products from sugar, but if you're not so good at either measuring or keeping careful control of temperature, you get... something. It's sugar so in most cases it's delicious anyway, but if you wanted fudge but you've made toffee you may be disappointed. With practice you can "eyeball" it without better equipment, like the cat's whisker, but with better equipment an idiot with no experience can make it do what they wanted because the numbers were correct.

> you just know if you fiddle about with a fine wire and certain types of crystal,

Even an old rusty razor blade will work as a diode if you try hard and long enough.

Samurai sword making took what, a thousand years of trial and error? The forgers had no idea how making steel worked, they just found a way to make it work.

More interesting that they found a way to work with poor quality iron

I have repeatedly read that Japan has poor quality iron, but what exactly does that mean? Iron is an element, so all of the reserves were in an unfortunate oxide requiring sophisticated refining? Just low total abundance?

Lots of impurities and limited quantity. Most of it was in ironsand, and even then most of it was in fairly low quality ironsand.

They had to get pretty creative in their use of ironsand to produce high quality steel.

Getting a pure ingot of iron is no simple task if you don't have access to a good furnace. So you'll end up with 'bad' (impure) iron if it doesn't come out of the ground in reasonably pure chunks.

All the cool swords seem to come from societies without access to good iron. Ever seen a 'Viking' sword?

Impurities.

Notably phosphorus is very hard to get rid of when it's in your metal. Before better metallurgy, the best steel worldwide was made out of low-phosphorus ores.

Or bread… have you ever thought what must’ve gone through somebody’s head to think of grinding all those cereals, adding random stuff, and baking it? How the heck did anybody come up with that??

Yes, based on the materials and type of process, it seems like the default assumption should be, if it's real at all, it's going to be a tweaky, unreliable process at first. A more surprising result would be if the published process reliably works 100% of the time.

Reply on news.ycombinator.com