Wouldn’t the smallest of mistakes corrupt a file, or is there room for error?
In: Technology
Computer scientists have spent decades studying this exact thing: how to send data in a way that lets you detect errors and fix them.
Basically, there are special ways to encode data that allows you to detect if part of the data got messed up.
The example given to beginner computer science students is: imagine for every chunk of data, say, eight bits (a bit is a one or a zero, you probably already know that computer data is made of a bunch of ones and zeros), you use seven bits for the data you want to send, and the last bit is a check bit: if the sum of all the other bits is odd, you put a 1, and if the sum is even, you put a 0.
So let’s say our data is 0011010
There are three 1s, so it’s odd. So our chunk of eight bits would be: 00110101
Now as we’re sending the data, imagine it gets messed up a bit, and one of the zeros turns to a 1: 01110101
On the receiving end, the computer reads this and says, wait a minute! There’s an even number of 1s in the data section, but the check bit is 1! That can’t be right. Can you send me that chunk of data again?
Now, you may notice this method fails if *two* bits get flipped. That’s why computers actually use a more advanced system that can even tell you *where* the error is, allowing the receiver to correct it automatically.
And this is only one of many systems computers have to ensure data integrity. When transferring large amounts of files you might use checksums to ensure they transferred correctly, for example.
The overall takeaway is: there are special ways to process and send data that has built-in checks to make sure the data doesn’t get messed up. If the receiver detects that the data is messed up, they can correct it if possible, or otherwise request that the data get sent again.
Latest Answers