O/T : Looking for an entry level LAMP developer for contract work.

Steve Litt slitt at troubleshooters.com
Sat Aug 1 16:09:32 MST 2015


On Sat, 01 Aug 2015 06:04:56 -0700
Keith Smith <techlists at phpcoderusa.com> wrote:

> 
> With problem solving being so important you would think it would be 
> taught.  

But nooooooooooo!

This could get interesting. Thanks Keith!

Problem solving is considered a "soft skill" and few educational
resources teach "soft skills" for technological professionals. For
instance, the main educational resources for systematic technical
troubleshooting are the military, Troubleshooters.Com, and the school
of hard knocks.


> Maybe it is now, however it was not way back in the punch
> card days. In the last century things were less complicated and
> easier to fix, for the most part.  

For obvious reasons, I've given these things a lot of thought.
Somewhere around 1970-1980, goods sold to consumers became too
complicated to troubleshoot via guess, swap and hope. But lots and lots
of technologists still guess, swap and hope. This is why you have so
many problems getting your car, air conditioner, or computer
repaired, and why most folks consider buying a new $1200 TV before
getting their current $1200 TV repaired. It's why many IT centers
remain down for hours or days instead of minutes or hours.

> The problem David described is a
> very elusive one.  I understand why it took a while to find it.

I'm highlighting the following sentence...

> However you would think those familiar with the configuration of the
> system would have been able to find the problem rather easily.

The preceding sentence is common sense. I'd say 90% of the general
populace and 60% of technologists still believe it. Specifically, they
believe that troubleshooting productivity is proportional to knowledge
of the system and technology under repair. This turns out to be very
untrue for one specific reason...

The Troubleshooter is human!

The following typical (though not universal) human attributes require
the human to follow specific processes and adopt a specific attitude and
mindset:

* Humans' pattern recognition is superior to almost any AI system.
* Humans easily exploit knowledge from "similar situations".
* Humans can retain only about 7 facts in memory at once.
* Humans take minutes to recover from an interruption.
* Humans forget without the aid of writing or an organizational system.
* Humans make a lot of errors.
* Humans performance quickly degrades with pressure, panic and anger.
* Humans' results hugely depend on their beliefs.

Humans trying to troubleshoot like robots inevitably end up at guess,
swap and hope. Guess, swap and hope sometimes works for technical
problems similar to what they've experienced, but fails completely when
confronted with novel technical problems on complex systems.

> Sounds like there was some politics at play which made the problem
> worse.

This is often the case. Personally, I've often violated policies when
necessary, in order to get things fixed. But of course I can't teach
that at a troubleshooting course the company is paying for.

 
> For the most part problems are easy to identify.  

If you mean it's easy to articulate the symptom, that's absolutely
true. If you mean it's easy to find the root cause, well, that's easy
on a reprocible symptom if you have the right mindset and follow the
right process. On intermittent symptoms, nothing's guaranteed easy.

> The bigger
> challenge is having the foresight to not build problems into the
> system.

If you're discussing building for repairability in the preceding
sentence, or if you're referring to building a system as simple and
robust as possible, current fashion declares repairability, simplicity
and robustness as a necessary sacrifice for cool new features. Guys
like you and I just troubleshoot around that. Matter of fact, doing so
is the likely subject of my next book.

My replies are only about Technical Troubleshooting, which is my area
of expertise. Technical Troubleshooting is one of very many facets of
problem solving.

Keith, Amy, David, Sesso, Jason, thanks for the very interesting
observations and discussion.

SteveT

Steve Litt 
July 2015 featured book: Rapid Learning for the 21st Century
http://www.troubleshooters.com/rl21


More information about the PLUG-discuss mailing list