Stupid Modem Tricks
Jerry Laiserin

There's no way to predict when, if ever, you will receive and read this, but it certainly won't be on the scheduled/promised date of 24June2002. Why? Well, I'm positive that cable modem access provider Optimum Online is not the world's worst service, and that the customer-friendliness of their corporate parent, Cablevision Systems, is not the most egregious example of monopolistic arrogance on the planet. That said,...

Suppose you were a hypothetical customer of a hypothetical cable modem access provider. Suppose further that your connection had worked—adequately, at best—ever since you installed it last year. Now suppose that some unknown third party installed a cable modem on the same network segment and configured his modem with a fixed IP address the same as the DHCP-generated IP address that the access provider had provided to you. For the non-technically inclined, think of this as boarding a plane only to find another passenger sitting in your assigned seat and possessing an equally valid boarding pass. Just as two passengers cannot occupy the same airplane seat, regardless of the identical seat numbers printed on their respective boarding passes, two computers cannot occupy the same IP address inside a network. However, unlike the airline analogy, the "bursty" or intermittent nature of IP traffic allows the network "plane" to fly while the two passengers wrestle for the same IP "seat."

On a real airplane, the duplicate seat numbering on the boarding passes clearly is not your fault, but is an airline "network" problem. Therefore, it would make no sense for the "customer support" flight attendants to insist that you could fit in your seat if only, for example, you took off all your clothes, turned them inside out, left half of them off, and then tried sitting. But, cable modem service provider folks might insist for days that your inability to access their service could not even be discussed until you had disconnected your router, bypassed every cable splitter and internal coaxial wire in your home, and reconfigured every Windows Control Panel "Network" setting and every Internet Explorer "Internet Options" setting on your computer (perhaps because the cable folks think that Microsoft's customer-friendliness is an egregious example of monopolistic arrogance).

Back on board the plane, it would not be especially helpful if the airline folks insisted that your belt buckle must have just gone defective—coincidentally as you boarded—and suggested that you would fit in your seat if only you purchased a new, non-defective, house brand belt buckle from their wholly-owned retail outlet. Few cable modem access providers own their own discount electronics retail chains, but if one did, the analogy would be like telling you to whiz on down to their nearest retail store and buy a new splitter to replace your newly "defective" one (splitters are passive components unlikely to fail). After all, they only want to spare you that ominous $50 "truck roll" (on-site service call).

I could ramble on, but the truly frustrating parts of such a hypothetical experience are: the persistent insistence that all problems are customer problems, while a clearly flawed network is presumed flawless; the final resolution that calls for the innocent and injured customer to stay offline an additional half day while the newly enlightened network cops track down and shut down the fixed-IP perpetrator who continues to enjoy that extra half-day of unencumbered usage.

If there is any "take-away" lesson here for AEC, plant/process, and infrastructure industry readers, I think it is to remember that the more bizarre and unlikely the clients' complaints or concerns, the greater the likelihood that it is your fault and not theirs. Telling clients or customers that everything will work fine if only they would rearrange and reconfigure everything on their end is rarely the most direct path to user satisfaction.
JL



< back