Anselm Lingnau wrote: >> Lest anyone thinks I have a poor opinion of the RHCE exam, I don't. As >> exams go, this is an especially good one and Red Hat put a lot of work into >> making sure it is relevant and meaningful. Luckily, false positives (people >> who passed but didn't deserve to) don't appear to happen in my experience. >> > > I guess the false negative rate means that Red Hat wants to make *very* sure > that the Anointed do know their stuff. If ever so many people who might (or > ought to) have passed the exam in fact fail the exam, this is not really a > big problem, from Red Hat's point of view, as long as there are enough people > around who *do* pass such that enough certified practicioners are available > to just satisfy the market's demand for official experts in Red Hat's > offerings. This goes, of course, to the philosophy of why an IT certification exists, which is significantly different between LPI and Red Hat.
> It is as well to remember that Red Hat's main business is selling Linux > expertise to commercial customers, with sidelines in developing and packaging > software and providing training for the Red Hat »priesthood«. This is very > different from LPI, which unlike almost every other certification provider in > the IT industry does not have actual products to peddle other than > certificates. If the goal is to create an elite (for the purposes of artificially limiting the number of officially-endorsed experts), then Red Hat is on the right track. Even the higher pricing and limited-exam-location models suit such objectives. But this is a philosophy borne out of maximizing one vendor's product sales, and I would imagine that most Linux people don't want to limit their careers to Red Hat shops. Having said this, such tactics -- using certification as an indirect component of vendor lock-in -- have been successful for Oracle, Cisco, Microsoft, and of course Novell which pretty-well invented the tactic. It is reasonable that if you are going the be RH certified then you must be able to use the "Red Hat Way" to solve problems, even if (as is almost always the case with Unix and Linux) many alternative ways exist. To use the drivers' license analogy suggested elsewhere -- you may have your own driving techniques, but you have to drive (and obviously know) the "correct" way during the exam to pass. The moment you get your license you can drive any way you want, but at least once in your life you had to prove you could do it the officially-sanctioned way. (In the early days when we were trying to explain the benefits of vendor-neutrality to an IT environment dominated by vendor (or vendor-consortium) created certs, I would start by asking "would it be acceptable to you if you could only get your drivers' license from Ford?" While many saw that as an absurd notion, it didn't take much to demonstrate that situation as the status quo in the IT world. It's a line that the media seemed to like as well, as it was repeated in articles such as http://www.computerworld.com.au/index.php/id;1727500083) LPI does (or, at least I hope it still does) take a different approach. Rather than using certification to artificially create a scarcity of experts, its goal for certification is to allow anyone to prove at least a minimum skill level. If there are more qualified people then jobs, that fact alone does not (and should not) diminish peoples' identified skill level. Maximum accessibility, rather than pseudo-elitism, is an explicit goal. OTOH, one would expect that a program such as LPI would be more concerned with results (ie, was the config file edited properly?) than technique (did you use the "right" editor to change the file?). > In other words, LPIC certification is a first-class citizen; it can stand on > its own without having to be constantly tweaked to deliver what external > players such as the product development, product marketing, partner liaison, > etc. departments expect from it this month. This approach isn't something that evolved over time; it is a core ethic that existed almost two years before LPI delivered its first exam. There were many community members who were quite aware of what they hated about vendor-driven certification -- the original group was determined to avoid as many of those characteristics as possible. Unfortunately, at the current LPI website history begins in 2005, so my assertions are hard to prove. :-) - the other Evan _______________________________________________ lpi-discuss mailing list [email protected] http://list.lpi.org/cgi-bin/mailman/listinfo/lpi-discuss
