[lpi-discuss] Re: Linux Professional Institute changes Recertification Policy

Bryan J. Smith b.j.smith at ieee.org
Fri Dec 1 13:55:02 EST 2006

I was going to stay out of this (let alone I don't have the time),
but I want to correct some misnomers.

Anselm Lingnau <anselm.lingnau at linupfront.de> writes:
> On the other hand, frequent recertification does generate
> money for the LPI, so must be a good thing. We're beating
> Microsoft at their own game :^)

No, that's quite _incorrect_.

The "money maker" in the certification industry is either "pushing
product" or, worse yet, "pushing training."  The cost of development
typically outweighs the revenues from certification fees alone.  Many
vendor certifications are about pushing product and many
vendor-neutral certifications is about pushing training.

LPI does _neither_.
That makes LPI very unbiased.
And given LPI's cash flow, they are _very_efficient_.
That's a win-win for professionals and industry.
Which is why I drum up the "value" of LPI.

"G. Matthew Rice" <matt at starnix.com> wrote:
> I'll stay out of this, too.  Except to say that I have heard
> of people asking if they could recertify _every_ year. \
> No comment on those people, either ;)

Matt brings up an excellent point.  There are both trainers and
professionals who justify both revenue and expenditures on yearly
training and re-certification.  I loathe the concept of certification
for certification's sake, and it insults me as an engineer.  

I think (i.e., my opinion) LPI is trying to balance the "realities"
of technology against those who thrive on the certification industry
or marketing certified professionals.  It says you must re-certify
every 5 years, and recommends you re-certify every 2.  In other
words, "we know you're probably current within 5 years" but "if you
maintain a set of certified Linux professionals, we don't think you
need to re-certify them less than every 2 years."

If there is one constant in this industry, it's that you will _never_
please everyone.  But LPI comes close to trying to do so, and I
wouldn't expect an organization run by and for Linux professionals to
do anything different.

"G. Matthew Rice" <matt at starnix.com> wrote:
> Pretty much.  There are a fwe cycles here.
> 1. We rotate the exam content a couple of times a year.

And that's the standard "exam bank defense."  It's a good practice in
general.  Professors who use the same questions over and over don't
offer quality examples.  Professors that do rotate questions and
generate new questions every few years are much better.

It's the best you can do without moving to another format that takes
far more investment and infrastructure to implement.  Just look at
the expense Red Hat goes through, and even it's format is not perfect
and tends to be more focused and less broad.  There is always a

Bryan J. Smith   Professional, Technical Annoyance
b.j.smith at ieee.org    http://thebs413.blogspot.com
     Fission Power:  An Inconvenient Solution

More information about the lpi-discuss mailing list