Carnildo:

Whiskey Tango Foxtrot? Over.:

BsAtHome:

I'm just waiting until Godwin's Law is invoked. This thread is developing in a way that would suggest it is getting there.

Hitler tried to ensure that all words had a specific, un-malleable meaning once...

He did? I thought that was Orwell.

That is easy enough to rectify.

Hitler and Orwell were actually the same person. After all, you would have to be a Nazi to write something like 1984.

See? Easy. ;-)
 

The thing about Godwin's law, is you can really apply it to a variety of other topics that will inevitably come up in a long and stupid enough thread, such as some moron posting an image macro... OH HAI

phaedrus:

th0mas:

BsAtHome:

I'm just waiting until Godwin's Law is invoked.

Why, so you can redefine it? 

So, let's get with this one.  All the people who think that they can just repossess Big-O notation are communists.

Now, all you communists, get with the redefining and invoke Godwin's Law on me.

EDIT: Yes, I'm trolling.  Yes, I'm just messing with you guys.
 

You know, the whole thread  developed into a discussion about form without actually looking at the content. Apparently, nobody bothered to think about a divide function and how such is supposed to operate (you can find it posted here). I wrote that mine would do a divide in O(1) here and that should have raised some suspicion and eyebrows. The whole focus on form without looking at content is the real WTF here.

Many problems arise from similar situations were the form is more important than the content. It is like a modern shoot the messenger and ignore the message. Therefore, invocation of Godwin's law would have been advantageous to the thread.

And, yes, we are all communists. In our world you do not posses the notation. It is our collective that is the center of the universe ;)

BsAtHome:

You know, the whole thread  developed into a discussion about form without actually looking at the content. Apparently, nobody bothered to think about a divide function and how such is supposed to operate (you can find it posted here). I wrote that mine would do a divide in O(1) here and that should have raised some suspicion and eyebrows. The whole focus on form without looking at content is the real WTF here.

Carnildo:

BsAtHome:

You know, the whole thread developed into a discussion about form without actually looking at the content. Apparently, nobody bothered to think about a divide function and how such is supposed to operate (you can find it posted here). I wrote that mine would do a divide in O(1) here and that should have raised some suspicion and eyebrows. The whole focus on form without looking at content is the real WTF here.

There's nothing particularly unusual about constant-time division in this contest. All three of my entries do constant-time arithmetic.

It is actually quite hard to do constant-time division if you have a string of digits. At least you have to go through the digits once... 

BsAtHome:

It is actually quite hard to do constant-time division if you have a string of digits. At least you have to go through the digits once...

All arithmetic operations operate in time somehow proportional to the size of the value. If you have a constant upper limit on that size, the operation is O(1) because everything reduces to a constant. This is not a particularly useful observation, so we usually assume infinite precision numbers (in which case, no operations more sophisticated than a bit swizzler are O(1); I'm not sure what the lower limit on division is off the top of my head, but it's probably something like O(n log n)).

asuffield:

BsAtHome:

It is actually quite hard to do constant-time division if you have a string of digits. At least you have to go through the digits once...

All arithmetic operations operate in time somehow proportional to the size of the value. If you have a constant upper limit on that size, the operation is O(1) because everything reduces to a constant. This is not a particularly useful observation, so we usually assume infinite precision numbers (in which case, no operations more sophisticated than a bit swizzler are O(1); I'm not sure what the lower limit on division is off the top of my head, but it's probably something like O(n log n)).

The theoretical low limit on division is O(1) in case of table lookup (at the cost of fixed operand/processing size and n^2 huge tables). Long division (shift and subtract) limit is O(n).

BsAtHome:

asuffield:

BsAtHome:

It is actually quite hard to do constant-time division if you have a string of digits. At least you have to go through the digits once...

All arithmetic operations operate in time somehow proportional to the size of the value. If you have a constant upper limit on that size, the operation is O(1) because everything reduces to a constant. This is not a particularly useful observation, so we usually assume infinite precision numbers (in which case, no operations more sophisticated than a bit swizzler are O(1); I'm not sure what the lower limit on division is off the top of my head, but it's probably something like O(n log n)).

The theoretical low limit on division is O(1) in case of table lookup (at the cost of fixed operand/processing size and n^2 huge tables).

Note the part where I pointed out that only the infinite precision case is worth describing in O() terms, because absolutely anything with "fixed operand/processing size" runs in O(1). In the infinite precision case, that requires a table of infinite length.