Re: Revised: A Proposed Concensus for Recertification and Renewal

[Alan & Susan Mead]

Kenneth,   I'm sorry that this topic and my reply anger you so much.  Since this is a consensus point, please indulge me in trying to explain the issue as I see it:   If (hypothetically) I certified on Linux and then never touched a system for, say, 60 years and on my deathbed (as my last wish) wanted to mess with your production system, how much confidence would you have in my certification?  I hope very little since I am sure I would forget many critical little details.  So this contrived argument is intended to show that it is improper for an organization such as LPI (which depends on its reputation to "validate" its certifications) to make open-ended certification decisions.   I think there is a critical difference between organizations which offer training/education and which issue a degree/certificate/whatever to mark that a pupil has completed the experience and those organizations like LPI which offer a "stamp of approval" that a person knows something.  In the second case, the certifying body has no control over the individual's training or efforts to continue training and thus need to be cautious about making open-ended statements.   You're CPR example is apt.  What if human anatomy and physiology was changing at the pace that Linux does.  The zyphoid (sp?) process is moving around, the other bones are evolving into a more flexible plug-in systems and the heart should never have been implemented in "body cavity space" and is being migrated into "cranial space" to protect it better.  Don't you think you would want someone doing CPR on you to be recertified periodically?   Of course, the flip-side is that using CPR training is a low-base-rate event and knowledge not used for a couple years probably fades.  Whereas we expect the vast majority of Linux ceritifiers to be using Linux actively on a daily basis.  I think that's why LPI refuses to require recertification.  I imagine it working like this:  say a Linux professional certifies, they work with Linux for three years and then look for a new job, the employer says "Oh, I see you certified three years ago and you've been active in Linux since but LPI recommends your recertification" and the candidate says, "Yeah, I could recertify tomorrow if you want to pay for it but I don't feel like I need that validation anymore"  And the candidate gets hired (or the employer takes him/her up on the offer).  For candidates seeking employment from the occasional organizations run by morons (who might ignore actual experience in favor of a test), they could just take the test again.  This would certainly be no more a hardship than any other (Intel-centric) computing professional faces.   -Alan