[lpi-discuss] RE: Linux Professional Institute changes Recertification Policy -- the "2 issues, " plus LPI's direction ...

[Anselm Lingnau]

Bryan J. Smith wrote:

> Then it must have come up at the normal Advisory Council meeting, if
> there was one.

Nope, there wasn't -- that I was aware of, anyway, and as I said I was around 
the LPI booth during much of the convention. I would assume that if there had 
been one I would have noticed one way or the other.

> When I was involved with the 2006 April TAC discussion on LPIC-3, the
> plan as I understood it was that you would _only_ need to pass *2*
> exams to achieve LPIC-3 certification.  There were other discussions
> as well.  E.g., to get a "speciality," you'd only need to have LPIC-2
> and pass *1* exam.  Or you could combine 2 specializations into the
> LPIC-3, etc...

In Cologne, Matthew Rice talked at length about the two »core« exams for 
LPIC-3, and how they were going to add four different »speciality« exams in 
due course. A candidate would be expected to pass the LDAP-and-PAM and the 
Samba exam to be LPIC-3 certified, and could then add specialities such as 
security, Apache, mail, etc. as desired. There was a half-joke around about 
how someone with all six certifications would be known as a »LPIC-3 Master 
Specialist« or some such.

Had the new recertification policy been announced in time for that meeting, I 
would certainly have made a point of asking whether LPI expected such an 
individual to re-take all six exams every five years, or whether it would be 
enough to start out with two and take one additional »speciality« exam every 
five years to keep all the rest ACTIVE.

> The idea here was to _maximize_ both marketability as well _avoid_
> people taking exams over and over again on the same content.  At
> Level 3, you are definitely getting into specializations, and you
> can't be too generic or broad.  Whatever Matt & co. has come up with
> is _very_ well _thought_out_, with _extensive_input_.

I'm absolutely impressed with the amount of work that has gone into LPIC-3. 
There have been so many rumours about LPIC-3 during the years that I've been 
involved with LPI that it is good to see us moving forward on this 
eventually. I have a nagging suspicion that, considering that the ratio of 
LPIC-1 candidates to LPIC-2 candidates is approximately 10:1 (at least here 
in Germany), and stipulating that the ratio of LPIC-2 candidates to LPIC-3 
candidates is about the same at most, the work required to support a couple 
of hundred candidates a year might be more profitably invested elsewhere 
within the LPI world but hey! this is still something of a community project, 
so if there are people willing to do the work then by all means let them.

> Apples-to-oranges comparison.  LPI is an organized entity with fiscal
> aspects, _unlike_ the kernel team.  Even the Fedora Project isn't a
> perfect example, because the board has external (Red Hat) members,
> but it was the best I could come up with.

This (apples-to-oranges) is what I said later in my previous message.

Incidentally, Debian, which as a software project is run more along 
meritocracy lines than Fedora and which, like LPI, *is* an organized entity 
with fiscal aspects, has, as you're undoubtedly aware, dealt with this 
problem by off-loading most of the fiscal aspects to a non-profit corporation 
called Software in the Public Interest, Inc. Sensibly, SPI does *not* have 
veto power about what happens within Debian -- the technical decisions rest 
with the »meritocracy« within the project. SPI just hangs on to the money on 
Debian's behalf. As far as Debian is concerned this arrangement appears to 
work reasonably well.

There is no compelling reason why LPI could not be organized more along these 
lines as well, if there was more emphasis on volunteer participation rather 
than running the certification side of things as a quasi-business. We even 
have the non-profit set up already! This would basically make Jim Lacey's job 
that of a »project leader« elected by those »LPI volunteers« who, like Debian 
developers, have shown a certain level of commitment and participation over a 
certain period of time. The comparison breaks down here, since leading LPI is 
apparently a salaried full-time position while leading Debian apparently 
isn't, and presumably we (the LPI community) wouldn't be able to attract 
executives of the caliber that we have now on an honorary basis! On the other 
hand, running a »community-driven« LPI more like Debian and less like a 
business might indeed carry some advantages -- so far, in spite of all the 
in-fighting, and in spite of Debian being a vastly larger organisation than 
LPI, Debian OS releases have occurred more often than LPI exam releases and 
in shorter intervals than LPI objective reviews, too :^) And so far Debian 
does not seem to have had grave problems regarding either sense of mission, 
street credibility, or funding, all of which are issues LPI is struggling 
with every so often, so the Debian project must be doing some things right 
after all. (I'll be taking my tongue out of my cheek now, thank you.)

Anselm

P.S. In case you're wondering, my other mail address is »lingnau at debian.org«.

(This is my personal opinion and not that of Linup Front GmbH.)
-- 
Anselm Lingnau ... Linup Front GmbH ... Linux-, Open-Source- & Netz-Schulungen
Linup Front GmbH, Postfach 100121, 64201 Darmstadt, Germany
anselm.lingnau at linupfront.de, +49(0)6151-9067-103, Fax -299, www.linupfront.de