Slashdot Mirror
World's Oldest Puzzle Solved
An anonymous reader observes: "The Loculus of Archimedes, the world's oldest puzzle, has been solved. It has 536 solutions. You can find the details here."
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
You would be amazed how not-easy it is to find even one solution for most people. Try this with the loculus, or with a set of tangrams, or with a set of pentominoes (which can be fit into, and cover completely, any rectangle of area 60 with both sides at least 3 units long).
The only catch is that we humans have to have a pretty good understanding of a problem/puzzle/whatever in the first place in order to program a computer to solve it. The limit still really is us humans, that is unless we develop true AI, which I really think is impossible because of what I just said.
Isn't it amazing that a computer could compute in minutes what has taken humans thousands of years to solve?
And yet humans can solve in minutes some things which a computer couldn't solve in a thousand years.
The four color problem is a lot different than this puzzle because you have to define your shapes (topological relationships) in addition to checking properties of "joining them up"
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
It would probably help your cause if you commented on the article, and posted your take on it, rather then engaging in an offtopic rant against the people you're so pissed off about. You could actually be engaging in thoughtful discourse, rather than furthering the problem that so vexes you.
This is just as offtopic as the parent, and I was going to post anon, but fsck it. Put it in your journal pally.
"Inattention makes clowns of us all" -Bean
You build a simulation. Right now this is still in its infancy, and these systems obviously have to prove their worth by producing accurate results, but virtual organ simulation is where things are headed.
It's very likely we won't have the computing power available to simulate these accurately for another 20 years - but so far there doesn't seem to be anything that would prevent you from, in principle, modeling organs on a sub-cellular basis and obtaining a reasonable simulation of their macroscopic behaviour.
Nae bother
Isn't it amazing that a computer could compute in minutes what has taken humans thousands of years to solve? We're in a time in which the sheer calculating power of computers can predict stress and failure in complex structures (FEA), lift and drag of fluid flows (CFD), and even the way a polypeptide will fold into a protein.
I will be more amazed when a computer actually comes up with its own algorithms to solve those problems. As it stands now, a computer only crunchs numbers once it's given a very specific set of rules. Without an actual person to define the scope of the problem, design the algorith, and sometimes derive the maths needed for it, a computer is pretty much useless.
Computers are helpful, yes. Computers are panacea, no... at least not yet.
R.There may be 536 solutions, but the original creator started with a single solution in the form of the original pattern and order of cuts. We may never know the exact order and pattern of cuts that created the puzzle, but I'd bet we can guess how most people would attempt to create such a puzzle.
For example the fact that the vast majority of 536 solutions are bilaterally symmetric suggests that the first cut in the creation of the puzzle was right down the middle. I'd also wager that cuts that bisect fragments are more likely than cuts that nick a fragment. Such straight-line, bisecting cutting behaviors are more likely than cutting polygons out of the middle of the whole square.
It may be a math puzzle solved by a computer, but I wonder if we can learn something about how people think from it.
Two wrongs don't make a right, but three lefts do.