[lpi-discuss] Re: Certification v. Non-Certification Training -- WAS: LPIC-1/2 training

enigma at riddlefixer.com enigma at riddlefixer.com
Mon Jul 26 14:16:50 EDT 2004

On Mon, Jul 26, 2004 at 03:02:25AM -0400, Bryan J. Smith wrote:
> On Mon, 2004-07-26 at 02:26, enigma at riddlefixer.com wrote:
> > I have mixed feelings about non-real world training.
> > Who does it really benefit?
> Well, that's a larger issue.  But the problem is that many companies use
> IT certifications as a "measuring stick."  So most training surrounds
> certification these days.  It's sad, but if someone cares about that
> sheet of paper and you can't prove to them otherwise, then you have to
> play the stupid game.

Well, there are other choices.  I have chosen not to play the stupid
games (at least when I recognize them) ;)

> But the LPI program is very well designed.  Part of the reason is their
> certification program isn't about "pushing product" which most are. 

Well, it is about pushing Linux but I believe the community aspect of 
LPI has been a good and necessary thing.  As LPI gets more popular, 
one of the challenges will be to keep it a valuable certification.  
Some of that responsibility should also fall on those of us who train 
and develop training solutions.  

If you take someone into Operating Systems cold trukey (i.e. they have
had no other exposure to Linux or real world implementations of any
other OS), cram LPIC-1 training into their skulls such that they pass
the exam, you have devalued the certification because they will not
really posess the skills of the ideal LPIC-1 candidate.  Furthermore,
if they don't have a job doing it for a while and don't breathe it at 
home, all of what they did have learned will evaporate.

I am not sure there is a solution to this problem, other than a lab-like
test which LPI folk have continually been able to talk me out of :)  
If there were a way to certify their hands-on experience, that would be
ideal.  For example, you have to have so many hours of hands-on-linux
in certain core areas before you can get a certification (like pilots).

At the same time, we need to keep this available to those that don't
have a job or income.  Bringing the cost of the certification down is
important for obvious reasons as well.  

The LPI team has done very well addressing all these very challenging 
problems, but the landscape changes too, so we must all be on our toes.

Some ideas might include LPI sanctioned internships, jobs, online labs,
and self study labs (e.g. build a mail server that does xyz).  Yes,
validation is still a problem and creates more infrastructure
requirements and costs to LPI but can we really rely on only the exams?

LPI has the opportunity to show how a community-run certification effort
can be more valuable than the classic certifications we have now.  Linux
is changing the way people think about software and Intellectual
Property; why shouldn't our certification effort?  I believe LPI is
doing a great job.  As supporters of LPI, we must keep that spirit in
the things we do as well.

Frankly, I don't know if there is a solution to other than at the actual
interview; perhaps interviewers should have an LPI interviewer's resource 
guide or something.


More information about the lpi-discuss mailing list