Theoretical vs Practical Knowledge [was: Re: Opportunity for Usability Evaluation]

Tue Dec 11 16:31:07 PST 2001

On Tue, Dec 11, 2001 at 01:16:32PM -0500, Gwendolynn ferch Elydyr wrote:
> On Tue, 11 Dec 2001, J C Lawrence wrote:

>> So you too can write job req's that demand book larning' for
>> positions that require practical, pragmatic trench experience.

> I'm going to wander afield from where this thread started.

> IMHO, it's very important to have a mix of theoretical and practical
> knowledge. This isn't a suggestion that you must have a degree - but it's
> a strong suggestion that you must have some form of theoretical background.

> The major failing of a person who only has pragmatic experience is that
> they only know what they've encountered. My classic example for this is
> the wonderful SAGE-II/III that I had working for me, who managed to
> tar up and remove lib, because he was completely unaware of it's purpose!
> (not really a surprise there - it's not knowledge that the average hands
> on sysadmin would need to know).

 Whoever classified this person as approaching a SAGE III was in 
 grave error.  A SAGE I should understand the concept of .DLLs and
 shared libraries.  A SAGE II should be able to identify the major
 shared libraries (by basename or location) for each of the OS in
 which they claim competance.

 By SAGE II or III I expect an admin to have the good judgement to
 test their "theories" on non-production systems prior to performing
 "experiments."

  (Theory: These lib*.so files are redundant on this system.
   Experiment: Let's try deleting migrating them into a tarball).

> A bit of book larning' would have nipped that problem cleanly in the bud.

 Depends on which books!

 Actually it's *far* more likely that an *experienced* admin, even of a
 home system, will have encountered a damaged libc.so file (or a corrupted
 /etc/ld.so.cache or some other breakage in the shared library loading
 mechanism) than that the graduate with a B.S. in C.S. will have had
 this *specific* example explained to them.

 Your alleged counter example, to me, argues *against* your intended
 point.  It *supports* the claim that experience is more important
 because I've met far too many degreed professionals (all the way up
 to the doctorate level) who couldn't identify the shared libraries
 on their system with a stack of text books and manual and a root shell 
 prompt.  It's a practical detail that differs from one specific system
 to another.

 This is not to say that I'm rejecting your argument; just your example.
 The obvious observation is that there are varying degrees of 
 "understanding" beyond just "practical" vs. "theoretical."  

 I met a mathematician (a professor, teaching Calculus) who could 
 not understand how an automative differential was, in fact, an analog 
 computer which solves a specific differential equation *by* applying 
 power/torque to a pair of drive wheels as they round corners.  
 To him differential equations were abstractions with no physical,
 mechanical manifestions.  Obviously most automotive mechanics can work
 for their whole careers with only an intuitive understanding of what a
 differential does.  Show them a set of formulae and claim that some of
 the gizmos that they work with are "solving" those and you'll get a 
 blank stare.  

 Ironically every experienced mechanic knows that if the rear 
 end of a rear drive car is up on jackstands, and you apply torque
 to one of the wheels (say to loosen a lug nut) the other wheel will 
 spin in the opposite direction.  Most mathematicians would not guess
 this from an abstract description.  The underpaid tire monkey at the
 local Sears knows that he has to loosen the lug nuts *before* lifting
 the car and that he *must* drop the car before tightening the lug nuts
 later.  They "understand" properties of the differential in practical
 ways that a mathematician often would miss.

 The biggest difference between a SAGE III and a SAGE II is not
 the quantity of their experience or their formal training.  There is
 a vital qualitative difference.  A SAGE III or IV doesn't simply 
 do things a certain way by rote; he or she asks (or has asked and 
 discovered) how things work and why they need to be "that" way.

--
Jim Dennis 