Botched Experiment in Citizen Engineering

Poor Old Bob Cringely tried to do an experiment with Bit Torrent and prioritized VoIP and forget to control a variable: the bandwidth consumed by the VoIP call whether it’s prioritized or not. Naturally, he jumped to some very stupid conclusions:

With this new knowledge I did a simple test that you can do, too…I did a couple BitTorrent downloads of specific files, measuring how much time and total bandwidth was required. Then I deleted those files, changed my Internet Gateway settings to give priority to my Vonage VoIP packets, called my Mom on the phone and started downloading the same two BitTorrent files…

My test results were clear. I had no problem downloading the same BitTorrent files, but it took longer. That was no surprise. After all, I WAS talking to my Mom, which would have taken some bandwidth away from BitTorrent. But the more interesting result was that the total bandwidth required to download the same files using traffic shaping versus not using traffic shaping was almost 20 percent more, which undoubtedly came down to increased BitTorrent overhead due to contention and retransmissions involving the priority VoIP service.

In other words, he downloaded some files while he wasn’t on the phone and then downloaded them again while he was and it took longer the second time. Duh, isn’t that what you would normally want?

Cringely also makes the startling discovery that Bit Torrent is very aggressive in its search for bandwidth, and warns that throttling it makes it take even more bandwidth. That sound like yet another flaw in Bit Torrent, and I don’t see why all VoIP users should stop making phone calls so Bit Torrent can run faster. His argument is along of the lines of “we better not make those terrorists mad by killing them.”

Is this guy on drugs?

3 thoughts on “Botched Experiment in Citizen Engineering”

  1. It’s not a very good experiment.

    The fact that similar bandwidth contention would have occured without QoS (Except for the fact that the call may have been dropped), but Bit Torrent behavior isn’t predictable from session to session and I doubt any attempt was made to verify that he connected to the exact same peers each time. He may have used the same tracker, but I doubt the peers were the same.

    Shouldn’t we require sniffer traces from these folks in order to validate their results?

  2. Well, it is Cringely you’re talking about, so the answer may well be “why, yes, he is high on rock cocaine, why do you ask?”.

    Though in fairness he’s not smoking them rocks nearly as often as Dvorak, who does it intentionally.

Comments are closed.