The entropic explanation of time's arrow is gibberish.
The number of possible combinations of disorder is greater than order so that's why you never see a broken glass reverse time and reassemble itself. It's not that it couldn't, it's just highly improbably.
It just seems extremely weak to me. I bet it's never happened and never ever will happened and it has nothing to do with probability or "tending" towards disorder, whatever that means. The glass certainly knows how to break in an orderly matter.
son of parnas
March 12th, 2007 7:54pm
I never got the whole 'disorder/order' distinction. on the whole every state is a state of order, just that we tend to prefer some orders over others.
March 12th, 2007 8:31pm
>just that we tend to prefer some orders over others.
Since we pick some out of an infinite set of possibilities, the chance of the ones we pick, i.e. the possibilities we consider ordered, coming up are close to zero. Any _specific_ state of disorder is just as unlikely as order. Sort of like the lottery, all combinations have the same chance of coming up, yet people feel that 1,2,3,4,5,6 would be highly unlikely.
Would time be going backwards in the fringe case of the pane of glass magically reassembling itself? Not really.
Well, the math for entropy is very precise. It's a good way to measure the difference between states. Like Carbon that is in long tightly bound chains in gasoline vs. the Carbon in CO2 after it's burned.
But I agree that there is a certain "I have a hammer, so every problem can be solved by hitting it with a hammer" quality to using statistical mechanics to talk about filing cabinets, toy chests, or broken tea cups.
But it's not trendy any more for physicists to use inappropriate mathematical tools on every day problems. That's now left to the economists, who seem very eager these days to misapply their knowledge on areas far outside of their domain.
March 12th, 2007 9:03pm
Well of course the glass reassembles itself. It has to. It can't not. The foundations of modern biology require it.
March 12th, 2007 9:28pm
If it increased overall entropy for the glass to unshatter, sure.
It's only "highly unlikely" for a glass to un-shatter itself? Sheesh, with all the broken glass on this planet, I'd think if EVEN ONCE it 'leapt' back together again, somebody would have said something.
That's not "unlikely", that's something wrong with a theory that describes it as "unlikely".
March 13th, 2007 1:33pm
It's surprising how big numbers smaller than infinity can get.
An 'unlikely' chance just means "we've been watching these pieces of glass here, unmoved by wind, waves or weather for the last 10 million years and I think I saw a couple shards hook up for a brief moment back in the early Pleistocene period but not much since then."
Like that dumb "all intellectual property can be found in /dev/random" thread.
March 13th, 2007 1:46pm
Oh. I thought you meant "unlikely", as in "it's unlikely for me to win the Lottery". When of course, SOMEBODY wins the lottery every so often.
But what you meant was "unlikely", as in "it's unlikely I'm going to be hit by the planet ^h^h^h^h ex-planet Pluto".
I still think I'm with SoP on this one.
March 13th, 2007 2:42pm