On Tue, 14 Jan 2014 18:26:01 +0000, BSDuser wrote:
> IT is a young man’s game, rarely respected by management, and a tough
> path to salaries > 150K.
Depends on the industry. I’ve worked for companies that had a high
amount of respect for their IT professionals, as well as those who looked
to blame them for system failures (and here’s a tip: in IT, systems
fail. That’s part of the job. As is fixing them.)
> The on-call duties definitely would be a deal
> breaker for me,
These days, that is a deal breaker for me as well. But some people like
that - and even thrive on it. I used to - I remember doing a 36-hour
troubleshooting session on a critical system error. Support person
handed off when their shift ended, and came back the next day to find me
still working on the problem.
I can’t do that any more.
> plus why do these cert exams cost so much money?
Because there’s a lot of development that goes into building an exam in a
defensible and consistent manner. There’s a lot of development work
that goes behind building a certification program - it’s not just a
question of siting down and coming up with a few dozen questions to ask
So, why do they cost so much?
Because it costs money to develop and maintain a certification
Cost is a “barrier to entry” that’s easy to control. You set a price
that’s too low, and people just keep trying the exam until they pass it.
Then they’ve proven they can pass an exam - big deal. If there’s a
higher stake in it, then there’s incentive to pass the first time, which
means the student learns the material well enough to pass the first time.
Many people believe that if you don’t pay for something, it has no
value. While in the OSS world, we know that’s clearly not the case, in
the business world, the value assigned to a “thing” depends, in part, on
what they paid for it.
Certification program development isn’t just about developing the
exam. You also usually have to develop the training materials that teach
information that helps the student learn what they need to in order to
pass the exam. Done properly, this means developing a set of
interrelated objectives, structuring the objectives in a way that flows
logically, and then handing them off to course development and exam
development. In an ideal world, the courses teach content to the
objectives, the exam tests knowledge (or, ideally, skills - which means a
hands-on exam rather than a written exam), but the training materials are
developed independently from the exam so as to not “teach to the exam.”
That’s actually not an easy thing to do.
You need infrastructure in order to deliver exams. You need proctors
to proctor the exam to ensure students aren’t cheating (or trying to
scrape answers). You need a way of tracking the students’ progress in
the certification path (because usually a certification isn’t “one exam
-> one certification,” although that trend has been shifting in order to
Businesses are in business to make money. The same is true of
companies that deliver exams, develop certifications, and validate those
certifications. There usually is a profit motive involved.
Higher cost also helps protect the value of a certification - if it’s
cheap enough that anyone can get it, then there’s not much value in it,
no matter how hard the exam is (because low cost means you can keep
taking the exam until you pass it, and sooner or later you might get
lucky and stumble on the solution). Being one of a billion people who
can spell “certification” doesn’t have a lot of value. Being one of even
100,000 people who have proven they can solve a complex problem in a
reasonable amount of time? Yeah, that’s got value.
And that’s just a start. Take it from someone who spent a decade working
in that line of business. A certification exam that is cheap isn’t worth
> Also I
> doubt there is much demand for unix/linux admins when most companies can
> go off the shelf cloud services by companies such as Amazon.
I’m sorry, but this is a completely false assertion. Most companies
still have in-house IT, and probably will for a long time to come.
Especially with question surrounding surveillance, data security, and
data integrity. If you think a company like Target puts their POS system
on Amazon, you’re not really thinking it through. If that system can be
compromised to the tune of 110 million people (the latest “maybe as many
as this number of customers affected” by their data breach) is inside
their network, imagine how much higher the risk is if it’s not secured on
systems in their own data center.
Go look at any job board - there are needs for people managing UNIX and
Linux systems. Even if it’s in the cloud, so what? Someone still has
to manage it - just because it’s in the cloud doesn’t mean it doesn’t
have to be managed. Managing an AMI instance is not very different from
managing it installed on bare metal. Amazon doesn’t run the systems in
the virtualized instances.
openSUSE Forums Administrator
Forum Use Terms & Conditions at http://tinyurl.com/openSUSE-T-C