The halftime effect: Skype and financial markets

September 1, 2007

In the 1980s, before the beginning of major televised football games in the UK, viewers were asked not to put on their kettles for tea immediately after the end of the first half of the game. The reason behind the request was that millions of electric kettles switched on at once might cause a sharp increase in electricity demand and potentially cause a breakdown in the system. This came to be known as the halftime effect

Fast-forward to the recent past: a couple of weeks ago two conference calls that I scheduled could not take place because Skype, our preferred medium of communication, was unavailable. Checking Skype’s web site for explanations, the message I read reminded me again of halftime effect:

On Thursday, 16th August 2007, the Skype peer-to-peer network became unstable and suffered a critical disruption. The disruption was triggered by a massive restart of our users’ computers across the globe within a very short timeframe as they re-booted after receiving a routine set of patches through Windows Update.

As Windows users know, when a new update comes from the Mothership in Redmond, your computer will keep nagging you to reboot. So, although a bit shaky, it is possible that this explanation holds water. You may think that this is another story about a company that treats computers as if they were isolated units and does not take into account the networked nature of today’s computing. This sentiment would right, but only partial, in my opinion. As what the explanation does not disclose, however, is the other link in the techno-social chain that led to the Skype crash. The reason so many computers tried to connect to Skype’s exchanges at the same time is dependent not only on Windows rebooting, but equally on millions of copies of Skype trying to get connected on starting up the system. In other words, it is the default preference of Skype (‘connect on start up’) that is responsible for the crash just as much as the Windows operating system insistence to reboot the system. But let’s go deeper still. What does a default setting in a piece of software (or any other machine, for that matter) represent? Evidently, it represents the intentions and preferences of the designers of those machines. So, the default setting of Skype, ‘connect on system start up’, represents the intent of the makers of Skype, to have as many Skype subscribers as possible connected to the service at any given time. This intent, when embedded into millions of copies of networked technological artefacts and combined with another technologically-embedded set of preferences (those of Windows) brought about the Skype crash.

What has that got to do with financial market, then? Well, what disappeared during the Skype crash was the ability to connect users together. Or, in market speak – Skype’s liquidity dried up. So, the Skype crash leaves us with the following insight with regard to financial crashes: if we want to improve our understanding about the techno-social connections that occasionally play part in market crashes, it can be useful to look for social intents of various kinds (commercial intents, in the case of Skype and Windows, as we saw). It is possible that such intents, long ago frozen into silicone, may suddenly wake up from their cryogenic sleep, given the right constellation of conditions and cause a hot crash.

One Response to “The halftime effect: Skype and financial markets”

  1. danielbeunza Says:

    Interesting. And I would add that it can be much worse. While the focal point in tea-and-kettle is created by an external event (football matches, windows update), financial investors are intentionally trying to coordinate. And yet, what to do with it? This kind of risk, like similar sociotechnical nightmares (computer viruses, nuclear power) is that the very lurking-ness that makes them nightmarish also renders them difficult to prevent.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: