The Most Thankless Job in Tech Support
June 5, 2006 | Life | Technology | Travel
Tried to use the web without paying the $10 extortion tonight. No go; way slow. So I went to the upgrade screen and authorized the billing. But nothing changed—still super-super-slow, as billg would say.
So I called the tech support line, cringing all the way. Can you imagine a worse job in tech support than fielding calls from semi- to fully-clueless people paying $250+ per night at random hotels, trying to get their hopeless Windows laptops onto the web? People in airports and hotel lobbies regularly ask me, "Can you help me get the wireless working?" And I say, "Windows? Sorry, I use Mac, no idea. Would help if I could." The support guy was not totally clueless, but he was basically working from screenshots, and I'm running down ping speeds (1,500 ms!) and packet loss (35%) and MAC addresses, and sub-netted IP addresses, and he's not sure what to do with it all. The symptom presented as if the hotel network had cached my MAC address and was routing it through the old connection and slow equipment. The ideal "hit it with a hammer" fix would be to clear the router cache located somewhere in the bowels of the hotel. Good luck with that, Notio!
What worked was to plug in an ethernet cable, fooling the laptop into thinking the network port had changed, then switching back to wireless, which for whatever reason got things working again. At least it's not as bad as Minneapolis in 2004, when Internet service was provided by housekeeping. OMFG, that was scary, but worked out okay in the end.
Comments
That report sent me to this thought. The tech support person is going to have it in his bag of tricks now, if he didn't before, to plug into ethernet, then go to wireless if something inexplicable happens. And sometimes that's going to work. If it doesn't he'll go to the next trick. If he stays in the job long enough he'll build up a correlation between symptoms and responses, with no technical model that you would recognize as such. Essentially a neural net. Constructed by "jiggling wires", which is my method of dealing with electronics.
So what if user interfaces were designed with lots of wires to jiggle with the emphasis being on wide-ranging effects rather than logically structured tool sets? For maximum rate of patterned learning rather than understanding in the designer's sense of it? One obvious answer is that "jiggling the wires" would occasionally "short something out" and when the expert arrived on scene and asked "What the hell did you do?" no one would be able to say. So there's something missing in this idea, but there's something in there, too.
I know that more than once I've tossed together a program, expecting its lifetime to be a week, and come back years later to find people using it. Meantime a lore has sprung up about what the program "expects" or what it "means", and particular ways of structuring things so that the program will work. Things that I never anticipated and cringe to think people have been living with all that time, but that they seem content with now that results have become predictable. In a sense, the program was structured after it was written.
Posted by: Doug at June 6, 2006 09:16 AM
Post a comment
Thanks for signing in, . Now you can comment. (sign out)
(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)
