On Fri, 05 Feb 2010 21:46:02 +0000, nmarques78 wrote:
> I donât want to be disruptive, but this has to be said at some point.
>
> âFree, as in freedom of speech, not as in free beerâ.
Well, the discussion here has been largely about open development, which
is more along the âlibreâ definition than the âcostâ definition.
> People do want everything for free⌠Letâs make a small metaphore here.
>
> You want something to be certified for sale within European Union, it
> has to be certified with âCEâ, same applies to the US and other parts of
> the world with âRoHSâ certification.
> This process involves a credited institution like SGS, TĂźV, APCER or
> whatever. This process involves money spending.
Thatâs not necessarily the case. Let me clarify something here; I work
for Novell in the training department. Accreditation is something that
is useful, but itâs not something thatâs mandatory. IT certifications
are used to measure a minimum knowledge and/or skill level - nothing
more, and nothing less.
Also understand that coming from an IT background, I also have been
skeptical of the value of certification (I only got my LPIC-1 and Novell
certifications after starting to work for Novell as a technical trainer).
> If you realize that you want a âfreeâ course, lets compare to something
> such as Microsoft Certified (mouse) Engineer or eventually Red Hat
> certified Engineer, which is credited by Microsoft (for MSCE) and Red
> Hat (RHCE). This involves spending money.
Traditional certifications do involve spending money, yes. However, the
face of technical certifications has been undergoing a transition for the
past several years.
> Yet again, Novell is in business to make money. From the Marketing
> perspective it comes down to this:
>
> Goal of a Organization: Generate revenue/proffit (can come in several
> ways, like social profit, finantial profit, etc). Means of an
> Organization: The Product Target of an Organization: People in general,
> consumers.
>
> I would suppose Novell is spending money on this, as such they have to
> get their revenue from someplace. I would doubt they would be giving
> away free courses (certified like the commercial ones from Red Hat for
> instance), without taking nothing from the process.
Well, consider that revenue comes from products; from a training
perspective, thereâs a cost offset, but I donât see revenue generation as
a primary goal for the part of the business I work in. My goal is to
reach as many customers as I can - and I can see that there is a benefit
to Novell in raising brand awareness through the community development of
such materials. Sure, Novell is in business to make a profit - and as
big a profit as possible. But there are many ways to accomplish that
goal, and just building things and saying âpay for it or you canât have
itâ, while being one model, isnât the only model. Itâs certainly the
simplest model, though.
Another model is to invest (as you allude to in your next paragraph) in
community resources and then leverage that investment in a way that
brings in more bottom-line product revenue.
Personally, I see this type of project as that type of investment in the
community. Now, Iâm not someone with any budgetary authority inside
Novell, Iâm just a guy who manages part of our testing program in the
training organization. But I also understand the value and have been
doing a bit of selling of this idea internally as well - itâs not much
different than an idea I pitched about 6 years ago, in fact (though this
idea is much more developed).
> The sad truth is that everyone is deviating from âfreedom of speechâ to
> âfree as in free beerâ.
This is largely a limitation of the English language, though - I think
most people understand the distinction, but âno costâ is something thatâs
more tangible. âLibreâ is something thatâs more easily understandable by
developers, which I think itâs fair to say is who Stallmanâs intended
audience for the term is/was.
> Honestly for those who have done MSCE (I did it back on NT4.0), we know
> what this courses actually are, and in most cases are a waste of money.
> Where to click, what to enable, blah blah blah⌠But all of that is
> really crappy if you donât have low level knowledge on the system and
> protocols used. This why so many boxes are owned despite of all the
> progress in the last 10 years around security.
The problem here isnât necessarily that the certification has no value,
itâs that the value that organizations assign to it is not what is
intended by those who create it. A certification, as I said above, is a
minimal measurement of skill or knowledge. Itâs a starting point, not
the end point. So you can have people who understand the theory of how
an operating system works, but have no idea how that applies in the real
world.
Thatâs the application of knowledge/skills, which is something that a
certification traditionally does not measure.
Itâs not very different from getting your board certification in
medicine. Youâve passed your exams and proven what you know, but you
havenât operated on anyone in an operating room yet. The first time you
do that, you apply your knowledge and skills.
Granted, there are significant differences as well (a board-certified
medical doctor has been through much more rigorous training than the
average systems administrator - and thatâs as it should be).
> What you should aim actually is to have your local governament
> supporting Free Software, and use it on the educational system. This
> would make more sense.
I donât see that it has to be an âeither/orâ discussion. No reason to
not do both.
> I do recon a lot of places that when people apply for a job, if they
> mention things like MSCE, theyâre completly off the job oppurtunity,
> because itâs a stupidity testemony.
There are companies that rely too heavily on certifications, thatâs
true. I applied for a job once with a company that insisted that because
I didnât have my CNE certification, I couldnât possibly know what I was
talking about, even though Iâd written a book on the subject. I consider
it their loss, not mine.
That can go the other way as well; and thatâs up to the individual
candidate to decide. That doesnât mean that creating such a program has
no value.
> Itâs not a company or distribution provider that should form people on
> the real stuff, low level and protocol level. That as to come from
> somewhere else, like a degree, Masters, etc. Their role in this subject
> is what is really should be, handling some weird tools made for dummies.
> The quality of a sysadmin doesnât come from certifications At least
> most of BOFHâs around will know what I mean with this, and will recon
> that most stuff running on top of a Linux kernel is well documented, too
> well documented. If people wanna step over this learning process, they
> fail big time.
Sure, but that doesnât mean that having a system whereby someone can
educate themselves (be it in a classroom or with self-study options) and
then can validate their knowledge and demonstrate a minimum competence
level has no value.
Documentation is not always educational material; it tends to focus on
the âhowâ rather than the âwhyâ of using elements of a system. In order
to be effective as a training tool, the student needs to understand why
they do something, I think. Just knowing the rote steps to accomplish a
task isnât enough, because that then removes the ability of the student
to apply their knowledge to other similar tasks.
For example, if a manual teaches you how to use sudo to allow non-root
users to use /sbin/shutdown and it also teaches you how to start yast to
manage network settings, those are good things. But if the student then
fails to combine the two skills to allow non-root users to use sudo to
manage network settings using yast, then the learning experience has been
a failure for that student.
The way that a training course gets around this where a manual perhaps
doesnât (though it wonât always fail, because some students will take
initiative and figure out how to combine tasks just from reading the
manual) is by having an instructor (or supplemental material) that
describes what the sudo command does and why you would use it.
> Anyway, the best tool ever to help understanding linux, is vim, master
> vim, and youâll master all the rest
Well, thatâs an old-school approach, certainly. But not everyone works
the same way, either - and with a system like Linux, thereâs definitely
more than one way to do things.
Jim
Jim Henderson
openSUSE Forums Administrator