Siri, Apple and a brief history of IT communication breakdowns

By on
Siri, Apple and a brief history of IT communication breakdowns

I saw a documentary recently by Werner Herzog called Lo and Behold: Reveries of the Connected World. The basic premise intrigued me: Herzog bringing his storyteller’s gifts, curiosity and wit to bear on a history of the internet and speculation about where it might take us.

Sadly, the film is a bit of a disappointment, rambling from one talking head to the next, one technology-related topic to the next, without a well-constructed narrative thread to hold it together. If you get a chance to see it on TV some time, I recommend it – but don’t go out of your way.

It did, however, include one segment that resonated with me. Namely, the anecdote that gave the film its title, based on an event during the birth of what we now call the internet. As you probably know, back in 1969 the very first multi-site computer network was called ARPANET. What you may not know (I didn’t until seeing the movie) was that the very first node of the ARPANET was in Leonard Kleinrock’s Network Measurement Center at UCLA. 

You actually get to see that original node in the film – cool if you’re into that sort of stuff. The second node of ARPANET was installed in Douglas Engelbart’s lab at the Stanford Research Institute, which at that time was still part of Stanford University (it became a separate entity in 1970).

In October 1969 the first attempt at host-to-host communication between the two nodes was made. The plan was that the node at UCLA would attempt to log in to the node at SRI using the command LOG, which the latter would auto-complete to LOGIN. Given the sheer volume and variety of content swapped over the internet these days it’s staggeringly primitive — but you gotta start somewhere.

Anyway, the communication got as far as LO, then the network crashed attempting to send the G. Thus, the first-ever message sent on the internet failed — and the result was 'LO'*. Herzog, unfortunately, misses a nice little coda to this story. You see, frustratingly incomplete communication with SRI is an everyday part of life for millions of people all over the world these days.

Allow me to elaborate. When  Stanford Research Institute separated from Stanford University in 1970 it became a private commercial research centre that eventually was named SRI International. It has developed untold thousands of technological innovations in a multitude of fields, the majority of which end up being licensed or sold to other companies to develop further.

One of these was an artificial intelligence project called CALO, which stands for “Cognitive Assistant that Learns and Organises”. Yeah, catchy. CALO incorporated work from other parts of SRI on speech recognition and natural language processing. The idea was to create an intelligent assistant that you could speak to and which could then intelligently follow your instructions.

By now you’ve probably guessed where I’m going with this. 

In 2010 Apple acquired the technology SRI had developed for its speech-recognising intelligent assistant. It’s not called CALO any more, it’s called Siri — based on the fact it came from SRI, you see.

I’m sure many of us have stories of miscommunication with Siri. I once confused my wife by sending her a message that I was driving the kids home from the zoo – except that Siri somehow heard “moon” not “zoo”.
A quick check of the GPS confirmed I wasn’t actually that far from home. I can’t tell you how many times I’ve asked Siri to play a favourite song in the car, only to have my musical horizons involuntarily widened by some random selection. I could go on.

Next time it happens, just bear in mind that this moment of frustration, as you attempt to communicate with Siri and it just doesn’t quite get it, is not a thing that happens without a context. It is, rather, an echo of the very birth of the internet.

Lo, and behold.  

*Yes, I know that communication between nodes on ARPANET doesn’t really count as the internet and that communication between disparate networks didn’t happen until 1977. Don’t tell me, tell Werner Herzog.

Matthew JC Powell is a technology commentator, philosopher and father of two, in no particular order.

Got a news tip for our journalists? Share it with us anonymously here.
Copyright © nextmedia Pty Ltd. All rights reserved.
Tags:

Log in

Email:
Password:
  |  Forgot your password?