“My computer won’t boot.”
Mrs. TheLeanThinker’s computer was hanging on the logo screen, keyboard unresponsive.
I know already that if the CPU were bad it wouldn’t get this far.
I also knew that the system hasn’t even tried to boot the OS from the hard drive yet, so that likely isn’t the problem.
Working hypothesis: It’s something on the motherboard.
Start with the simple stuff that challenges the working hypothesis:
- Hang test a different, known good, power supply. No change.
- Pull memory cards and reinstall them one by one. No change.
- Pull the motherboard battery, unplug, wait a few minutes to possibly reset the BIOS. No change.
- Try holding down the DEL key on power-up to get into BIOS settings. Nope, system still hangs, though it does read that one keystroke, the keyboard is dead after that.
- Try Ctrl-Home to reach the BIOS flash process. Nope.
There is no evidence that the motherboard is not dead. Final test:
Get the numbers off the motherboard, find the same model on Amazon, order it for $37.50 to the door. (Intel hasn’t made this processor type since 2011).
New motherboard arrived today. Switch it out, takes about 30 minutes.
Boot up the machine, works OK, set the time in the BIOS, and pretty much good to go.
Convince Windows 10 that I haven’t made a bootleg copy.
The Threshold of Knowledge
I learned to code in 1973 on PDP-8 driving teletypes. Although my programming skills are largely obsolete these days, I am comfortable poking around inside the box of a PC, and I generally know how they work. Thus, the troubleshooting and component replacement I described above was not a learning experience. Yes, I learned what was wrong with this computer. (The “bad motherboard” was a hypothesis I tested by installing a new one.) But I didn’t learn anything about computers in general.
Rather than working through experiments into new territory, I was troubleshooting. Something that had worked was not working now. My experiments were an effort to confirm the point of failure.
Therefore, as interesting as the diversion was, aside from a little research on some of the more arcane troubleshooting, it was not a learning exercise for me. It was all within my Threshold of Knowledge.
In the Improvement Kata, “threshold of knowledge” refers to the boundary between “We know for sure” and “We don’t know.” Strictly speaking, we only say “We know” when there is specific and relevant evidence to back it up.
In this case, my challenge (fix the wife’s computer) was well inside the red circle.
But this wouldn’t be the case for everyone.
The Threshold of Knowledge is Subjective
Someone else with the same challenge may not see this as a routine troubleshoot-and-repair task. Rather, he has to learn.
I had to learn it at some point as well. The difference is that I had already learned it. I had already made mistakes, taken a week to build a PC and get it working many years ago. I learned by experimenting and being surprised when something didn’t work, then digging in and understanding why. On occasion, especially in the early days, I consulted experts who coached me, or at least taught me what to do and why.
Coaching To Extend the Threshold of Knowledge
Learning is the whole point of the Improvement Kata. That is why we call the “improver” the “learner.” If someone encounters a problem like my example and I am responsible for developing their skills, I am not serving them if I do something like:
- Sit down at their machine and troubleshoot it.
- Tell them what step to take, and asking what happened so I can interpret the outcome.
That second case is deceptive. The question is “Who is doing the thinking?” If the coach is doing the thinking, then the coach isn’t coaching, and the learner isn’t learning.
In this case I would also have to recognize this is going to take longer than it would if I did it myself. That is a trap many leaders fall into. They got where they are because they can arrive at a solution quickly. But the only reason they can do that is because, at some point in the past, they had time to learn.
“My computer doesn’t boot.” If my objective is for this person to learn, then I need to go back to the steps of learning. Given that the challenge is likely “My computer operates normally,” what would be my next question to help this person learn how to troubleshoot a problem like this?
I need to know what they know. “Do you know where in the boot sequence it is hanging up?” If the answer is “No,” or just a repeat of the symptom, then my next target condition is for them to understand the high-level sequence of steps that happens between “ON” and the login screen. That would be easy to depict in a block diagram. It’s just another process. But my learner might have to do a little research, and I can certainly point him in the right direction.
I’m not going to get into the details here, because this post isn’t about troubleshooting cranky computers.
“If somebody comes to me with a problem, I have two problems.”
- The original issue.
- The fact that this person didn’t know how to handle it.
You can easily translate my computer example into a production quality example. A defect is produced by a process that normally does not produce them. What is different between “Defect” and “Defect-Free?” Something is. We just don’t know what.
Is it something we need to learn? Something we need to teach? Or something we need to communicate?
If my working challenge for my organization is something like “Everyone knows everything they need to do their jobs perfectly.” then I am confronted every day with evidence that this is not as true as I would like.
If I look at those interventions as “the boss just doing his job” then I lose the opportunity to teach and to grow the organization. I am showing how much I know, and by doing so just extending the dependency. That might feel good in the short term, but it doesn’t do much for the future.
Think about this… in your organization, if the boss were promoted or hired out of the job tomorrow, would you look outside the immediate organization for a replacement? If so, you are not developing your people. When I see senior leaders being hired from outside, all I can do is wonder why they have so little faith in the people they already have.
*I remember when Gateway built their own machines, which I guess shows how long I’ve been playing with PCs. Then again, I remember when the premium brand was Northgate. Of course, I also remember programming on punch cards.