MOST SEASONED internet users will by now have heard the legend that Al Gore invented the internet – while others smugly nod and remark that it was actually Sir Tim Berners-Lee (though he invented hyperlinking, and therefore the world-wide web, and not the internet itself).
But it turns out that the internet was invented neither by Gore, nor Berners-Lee, nor the US’s Defense Advanced Research Projects Agency which created the first working model of ARPANET back in 1969.
Nope. It turns out that the idea is probably thousands (if not millions) of years old – and that the internet is basically built on the same logic that ants use simply to communicate with each other when they’re looking for food.
Stanford University says the similarity was observed when a university biologist noticed the systematic way in which a colony of harvester ants goes hunting for food – and asked a computer scientist if it looked familiar.
The computer scientist’s response was to note that…
The algorithm the ants were using to discover how much food there is available is essentially the same as that used in the Transmission Control Protocol.
You might not recognise the words ‘Transmission Control Protocol‘, but its acronym could ring more bells: it’s TCP, which when added to Internet Protocol (IP), gives you the TCP/IP method used by computers, phones and servers to communicate with each other online.
A deep similarity
TCP, in short, governs the way that individual bits of information are sent between two computers. If you’re sending a file from X to Y, X will split the file into a series of ‘packets’ and send them to Y one by one.
Y, as it receives each packet, will send an acknowledgement to X to confirm that the information has arrived safely. This gives X the ability to figure out how quickly the packets are travelling (i.e. the bandwidth of the connection) and it can therefore slow down the delivery so that the transfer doesn’t create a bottleneck in the system.
Harvester ants use a similar system to decide how many ants from a colony should go out searching for food. If an ant goes out for food and returns back quickly, the system observes that there must be a decent food supply nearby – and greater numbers of ants are dispatched to find it.
The comparison goes deeper, though: both the internet and the ants kick off the process by sending out a relatively large batch – in the ants’ case to maximise the area being explored for food, while the flood of packets is to get an immediate impression of the speed of the connection.
Stanford also points out that both paradigms have a ‘time out’ mechanism too: if an ant fails to return within a specified period, or no acknowledgement is sent for a piece of data, then both processes shut down altogether.
Predictably enough, the comparison has led to the ants’ system being dubbed the ‘anternet’ – about which more can be read in the open access Computational Biology journal.