Learnlets
Clark Quinn's Learnings about Learning
(The Official Quinnovation blog)

19 April 2009

Debugging

Clark @ 3:44 pm

This starts out slightly technical, but eventually gets to the learning!

My first job out of college was designing and programming educational computer games (FaceMaker & Spellicopter may be the two best known titles).  When I went back to grad school, I went for the design side, though I’ve kept the ability to understand what technology does, and do a bit more than the average bear.

So when my internet connection started getting flaky, I realized I had a problem.  Not the basic problem, but because there were several comonents in the chain, and it’s hard to isolate one without substitute parts.  And, of course, when you have two potential culprits, it’s so typical to have the manufacturer of one blame the other, and vice versa!  Now, my DSL modem and my wireless router were both quite old, at least 4 years and maybe as old as 6 or more.  And I didn’t have a spare of either router or modem. What was I to do?

(Tech details)

For those who are curious, the DSL signal comes in on the phone line, and then the modem translates it into ethernet. That could go straight into the computer, but instead I have that go to a wireless router to serve all the devices in the house (currently 4 computers, a Wii, two DS’s, and my iPhone; when no one’s visiting!).   The phenomenom I was seeing was the connection starting to hang on various accesses.  Rebooting both router and modem solved the problem (rebooting only one never seemed to work), but only for a while (8-24 hours). BTW, this behavior was described both by the ATT guy and a guy at Fry’s as classic hardware going bad.

I called ATT, and they agreed to send me a new modem (I reupped for a year).  That came and I managed to get it installed.  Took several tries, since they forgot to tell me that the modem now stores it’s own account login details, so those detalis don’t need to be stored in the router!  Ahem.  That one bit of info, and I was up and running again.

For a while.  Then the flakiness happened again.  So off I went to Fry’s for a router. For a ridiculously low price they had a refurbished one available, so I nabbed it. Same brand, only a newer version of my old one, which I was happy with.

Taking that home, I finally accessed it’s settings, but couldn’t make it talk to the modem!  The lights on both modem and router said they were connected, but no traffic would go through.  And I couldn’t access the settings via wireless, and it took a long time for the settings page (you control the router through a web page it hosts internally).

I took the router back and exchanged it.  I was willing to bet that the first one was just flaky.  With the new one, the settings page came up almost instantly, and I could access it wirelessly as well.  OK, that seemed better.And the lights indicated everything was fine. But, no traffic was still going through!?!?

I was pretty sure that, it being the weekend, I couldn’t get help ’til Monday, but I searched the Netgear site anyways, and they said they had phone help 24 hours, so I called and got through.  The guy there first said he couldn’t help me by phone for a refurbished modem, but then proceeded to tell me just what the problem was (turns out he couldn’t walk me through online, but could give me the details, which was all I needed).  Of all crazy things, the modem and the new router both want the same URL!  He had me reset the router’s IP address to something different, and viola’, I’m online!

(End tech details)

The learning here is severalfold.  First, systematicity helps.  Now, I know that, but it’s nice to have a chance to practice it.  One of things I miss most about not programming anymore, besides the ability to create new experiences, is debugging.  I loved using logic to try and figure out what’s wrong, and testing, repairing, and so on. I used to work on cars with my Dad, and the same process would be followed.  I think systematic research and testing is a meta-learning skill, and one we really don’t teach in school, yet it’s critical!

Another meta-learning skill, or really attitude, is persistence. I didn’t have an option, because no internet connection would be a critical business issue.  Fortunately I had a connection, it was just flaky (and with all this online seminar action coming up!.  And I admit there were times when I was tempted to use bad language (or did), and/or had to take time out to cool off.  But I kept thinking, testing, talking, reading, and more.

Of course, the two critical pieces of information would’ve been devastating if I didn’t have them.  And I didn’t find either in a discussion forum, I talked to people, live. I’ve learned to be very clear about the steps I’ve already taken, and that helps to short-circuit what can often be very basic stuff (e.g. “did you plug it in, and are the lights lit”). I mentioned that I’d tried the manual, and my steps, which helps build credibility with the tech person (the router person commented that he thought I could do with just the instructions).

So, I think we could and should spend more time developing reasoning skills as well as rote knowledge (duh!), and help people learn to share their thinking to help identify the problems faced.

And now, here’s hoping it was the hardware and not a different problem!

Note that this ate up a lot of my time this past week, what with store, and time on tech support, etc, so this is also an apology for my lack of blogging this past week!

4 Comments

  1. I hope you have a solid fix too, time will tell. And bonus, you won’t have to think about your options to switch providers, at least for a year…

    Comment by Lori TZ — 19 April 2009 @ 9:01 pm

  2. Hi Clark,

    I think this is valid. We should be able to apply or learn the skills required to debug any problem rather than just call for help. It is pertinent to identify the steps and be systematic in implementing them. I was recently planning to buy a wifi router and this would be a good review for me to know one of the situations that could come up.

    Thanks for sharing it and nicely highlighting the learning aspect related to the activity of debugging.

    Sreya

    Comment by Sreya Dutta — 19 April 2009 @ 9:47 pm

  3. I’ve learned to be very clear about the steps I’ve already taken, and that helps to short-circuit what can often be very basic stuff

    I’ve had nearly the opposite experience. I still debug code for a living on a regular basis, and whats more I spent quite a few years working in a 2nd level system support group, so I got used to helping other people debug their problems, both hard and soft. As a consequence, I used to try all of the obvious systematic tests, read all the FAQs, etc., and basically diagnose my problem before ever contacting anyone in support. What I found is that they are almost invariably (though most of my examples in recent years are with HP, so maybe they’re just especially bad) highly trained at following the procedure in the flowchart, and nothing short of global thermonuclear assault by three-headed aliens is going to shift them from it. After a couple of calls where I tell them that the monitor is faulty and they insist on walking me through the procedure to prove that its the monitor first anyways, I started just saying “Yes, I’m trying the incredibly complicated half-hour procedure now” *5-second pause* “No, that didn’t work. Because its the monitor that’s dead.” Eventually though, through the heavy use of excruciating feedback, they trained me to not investigate the problem at all before calling, since I knew they’d only drag me through the testing again anyways. I found this kind of sad.

    Short of venting, I’m not really sure where I’m going with this. But in a vague attempt to tie it back to the learning focus of your blog Clark, I’ll say this; they have a system which is _not_ designed to train me to do something, yet it _has_ trained me to do something, and the thing it has trained me to do is – presumably – costly and detrimental to the company. (Ok, so the minimal cost of a few hundred extra operators in Bangladesh is probably mostly offset by the expense saved not having to fix the problems of the people who give up in disgust. But as its definitely one of the reasons that I will never buy another computer from HP for as long as I live, it can’t be good for the business in the long run…) Whats sort of interesting about the whole process is that I’m not entirely sure how you would go about fixing it. From my own experience in support I know that for every person who has already correctly diagnosed their own problem there are 99 guys swearing that the server is “down” because they’ve kicked the power cable on their desktop out for the third time this month. Unless the people answering the phone have a deep model of how the system works, they haven’t got any idea if the steps you’re claiming to have taken are sufficient to properly diagnose the problem… and if they somehow acquire that deep model, then you have to promote them away from the phones before they leave for a better-paying job with someone else.

    Comment by Rob Moser — 20 April 2009 @ 1:24 am

  4. Lori, LOL.

    Sreya and Rob, interesting alternatives. Yes, Rob, I’ve had the same experience (and I won’t buy HP either, they fixed one problem on a new Compaq (they’d just bought them) and introduced another. Maybe the fact that if you’re calling about a router, a more technical piece of equipment, rather than a computer, they have a more educated tech support.

    I, too, hate the folks that have been hired not because of their knowledge but their ability to speak English and follow a decision tree. And any reasonably astute person’s game is ‘get them to escalate to 2nd level support as quickly as possible’. And they need the first level for all the unreasonably unastute people out there!

    I do find that when they say “I have to ask you to do this”, I’ve generally been able to say “I did that and this was the result”, and get them to skip to the next stage of the script.

    But I have seen it degenerate to the level you’re talking about, Rob, and I’ve seen it improving more of late. Yes, you have to negotiate auto-phone trees, but I don’t mind that if it gets me to the right person faster and it’s reasonably good (I’ll shout out United as being good about that, at least for a relatively frequent flyer). I think Netgear did a good job. ATT was *pretty* good. At least they seemed relatively knoweledgeable. Maybe the whole ‘customer experience’ thing is finally holding sway as a differentiator for the clued-in, and HP’s not there yet ;). Here’s hoping. (And venting’s ok, I reckon, as long as you also tout the wins. ;)

    Comment by Clark — 20 April 2009 @ 9:09 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress