Boldly predicting the Internet is safe

If you want to make a name for yourself, you can't be bashful about making bold predictions. Witness Bob Metcalfe (columnist).

Sure he invented Ethernet and started 3Com, but what really put him on the map was predicting the Internet would fail by a certain date. As you know, the Internet didn't collapse by the appointed date and Metcalfe had to eat crow (I'm not sure he ever made good on his promise to also digest one of his newspaper columns).

Bolstered by that experience, I hereby make my bold prediction: the Internet will never fail! Obvious? Not really, given the vagaries of the Internet and the great unknowns. Regarding the former, there isn't a day goes by without some portion of the Internet failing somewhere. Companies that are in the business of delivering Internet-based services - meaning they have servers scattered all over the world and have a global view of the Net - say outright failures are more common than anyone would suspect. According to one executive, it's "really flaky".

Where the prediction gets bold is regarding the latter point about the great unknowns.

The late Bill McGowan of MCI fame used to say what he worried about most was what he didn't know. McGowan would read everything about anything, regardless of whether he knew the subject matter. His theory was the more you know about what you don't know, the less likely you are to be surprised by some development coming out of the blue. But where the prediction gets bold is regarding the latter point about the great unknowns. In the case of the Internet, the beast is so large and complex it is impossible to know if we know all the potential failure scenarios, making it impossible to safeguard against them.

Look at a simpler case, like the failure of AT&T's frame relay network. AT&T has a tradition of overbuilding things, and yet it never saw the bullet coming.

So then, how can I predict the Internet will never fail? Because, obviously, it isn't one network. Ultimately something could run amok and cause widespread outages, but three quarters of the world could go down and technically I still wouldn't be wrong. Why didn't Bob see that? To back up this bold prediction I am prepared to, uh, eat one of Bob's columns.

Join the newsletter!

Error: Please check your email address.

More about 3Com AustraliaAT&TMCIMetcalfe

Show Comments