Carroll Kong wrote:
> >
> > Look, first of all, I'm obviously not endorsing that anybody
> with x years of
> > experience are automatically handed a ccie number. They
> would still have to
> > pass the test just like anybody else.
>
> I trimmed down some of my extra fluff in the quote, sorry, just
> read
> the older archives with the same thread name.
>
> Oh I never suggested that either, I just said this initial
> "filtering" process is not clear cut, and we might be filtering
> innocent, bright individuals.
>
> > Therefore the idea is simple. You use a minimum number of
> years of
> > experience to eliminate the labrats. So instead, you get
> router-caressers
> > (hmmm, sounds like some people enjoy networking a little too
> much). You
> > then eliminate those guys with the test itself - if that
> highly experienced
> > person didn't actually learn how to do all those things you
> mentioned, then
> > it's unlikely that he would pass the test.
>
> Right. I am saying, it is NOT the number of years that matter,
> is it
> the quality of the number of years. One year of hardened fire
> fighting, troubleshooting, advanced deployment, cut over
> experience
> is sure worth a lot more than...three years of "maintaining
> the
> network" aka Router Carresser. But who gets to judge the ratio?
Well, obviously Cisco gets to judge the ratio. Hey, right now, Cisco gets
to determine that people are internetworking 'experts' just from a 1-day
test that deals with only network configuration but no troubleshooting, and
we've all learned to accept that, so what exactly is so outrageous about
Cisco also judging whether you've had 'enough' experience?
The point is that in any profession, somewhere along the line, somebody is
making an arbitrary decision. Medicine, law, you name it - somewhere along
the line an arbitrary decision is being made. To say that the CCIE process
should be any different is really to hold the program to perfection.
>
> So, do we 'weight' the one year of hardened experienced more?
> Or
> less? I am not talking about the exam yet, just, what about
> the
> legitimate people you are filtering out? What if they make it
> "three
> years of experience" because that is how long it takes for the
> "average" IT guy to figure out that Netbios can run over TCP/IP?
>
> What about the guy who figured it out in 5 minutes? Surely we
> do not
> want to disqualify him just because he figured it out in 5
> minutes?
> Of course not, so how do those guys still benefit?
All this presumes that the only way a prodigiously precocious engineer will
find work is if he gets his CCIE. If a guy is really so preternaturally
brilliant that he can figure things out in 5 minutes what takes normal
people 3 years, then surely some company will pick him up and he will then
accumulate the experience necessary to meet the experience threshold. Is it
really such a tragedy to force that guy to wait for a bit to get his ccie?
After all, a guy with such networking perspicacity probably won't even care
about the ccie after spending 3 years in the workforce - he's probably
looking at getting his PhD and/or looking to join Howard and write BGP drafts.
Consider the case of airplane pilots. Just to get an pilot's license, you
must have a certain minimum number of documented flying hours. To be hired
as a pilot for an airline, you must have documented proof that you had at
least several hundred hours of flight time, and sometimes several thousand.
But you might say what if I'm the next Chuck Yeager and I can learn in 1
hour what it takes normal pilots 10 to learn? Too bad, you still have to
have that minimum number of documented flying hours to qualify. Simple as
that. Or consider doctors. Every single Medical Board requirements dictate
that you must spend a mandated amount of time in an approved
internship/residency program that deals with the medical specialization in
question. Even Doogie Howser himself can't flout those requirements - if
you want to be Board-certified, you have to fulfill the time requirements.
So if time requirements are OK for pilots and for doctors, why are they so
inappropriate for network engineers?
>
> > Now obviously, this is imperfect. You will still have some
> guys who carress
> > routers (man, that just sounds disgusting) and then bootcamp
> their way to
> > getting their ccie. I agree. But there is no perfect
> solution. It's better
> > than what we have today, where labrats bootcamp their way to
> their ccie.
> > Bottom line - a caresser CCIE is on average more skilled than
> a labrat CCIE.
>
> Perhaps that is true. (I am not going to argue either way, but
> I
> think it's debatable. :) )
I really don't see how it is debatable. The lab-rat CCIE has just the CCIE
to his credit. The caressers has both the ccie and some experience. They
have everything the lab-rat has and more.
Or, if you prefer a more quantitative explanation, when x(i)>= y(i) for all
instances of i, then MEAN{x(i)}>MEAN{y(i)} except for the special corner
case of x(i)=y(i).
>
> However, this is akin to the scorched earth tactic. I suppose
> until
> we find out how many people passed the CCIE, are considered
> WORTHWHILE, and find out how many years of experience they had,
> we
> will not know how many "innocent" victims we will fry with this
> tactic.
>
> If you are okay with frying X number of "innocent, bright"
> people (I
> would be very interested in the statistics myself), then sure,
> we
> should do it, just like the CISSP. (which I strongly disagree
> with
> myself)
Again, I dispute with the notion of 'frying' them. It's no more tragic
than forcing Doogie Howser to go through his residency. Those guys who
truly have the killer kung-fu should have little problem in getting hired
and picking up the necessary experience. And besides, like I said, I don't
know that those guys would really care about the ccie anyway, particularly
after they've spent several years in the field.
>
> My argument is, should we really be frying those innocent
> people when
> I see far more 'hardened' experience people worth far more than
> the
> router carressers? Odds are those hardened experience people
> also
> have faster learning capabilities to keep up. Those are very
> good
> people we are potentially filtering out.
But again, not filtering out forever, just filtering out temporarily. If
you're good, you will be picked up by an employer and you will get that
experience. Again, it's no more tragic than forcing Doogie Howser to endure
his residency.
>
> > And you ask about the integrity of the background check
> procedure. Well, I
> > am proposing using the same procedure that some employers
> today use for
> > their job candidates, where they hire companies to fact-check
> your resume.
> > I believe how it works is that those companies then go to who
> you claim to
> > be your former employers and obtain a signed legal document
> from their HR
> > departments using official company letterhead attesting to
> the fact that you
> > worked there from such-and-such dates and held such-and-such
> a position.
> > It's not just a matter of calling up some old managers who
> may secretly be
> > your golfing buddy and assessing your skill, it's about using
> a formal
> > procedure that is subject to legal action if marred. Cisco
> obviously
> > wouldn't be doing this, but there are dedicated
> background-check companies
> > who do this as their main business and Cisco would contract
> with one of
> > them. Or if you really want to get down to it, you can do
> what the
> > government does before assigning secret clearances, including
> interviews of
> > random coworkers and so forth.
>
> Well, the first part is STILL suspect to what I described
> earlier
> (trimmed out of this post, sorry), which is some managers do
> not know
> how bright individuals are. How can they? They are not
> technically
> ultra savvy themselves! It is what I call the "local guru
> syndrome."
> The guy who is your friend's sister's cousin's computer
> expert. The
> guy who says Mavis Beacon version 10 will corrupt Windows XP
> because
> it has "keyboard drivers" in it. The guy who says Netbios is
> absolutely unrouteable and has 3 years to back it up. The guy
> who
> was left unchecked and let his ego make him believe he is the
> master
> of all technical computer knowledge! Anyway, (sadly enough
> only the
> last example was an exaggeration), so the "local guru"
> convinces his
> manager and others that he is good. I am sure he is great,
> relatively speaking, but put in a pro to compare, and he turns
> into
> silly putty. The manager, without seeing the pro, will give
> this guy
> the thumbs up because he has won the "local guru" award.
Well of course, and these people will be filtered out by the test itself.
Again, I am not proposing that guys be handed their ccie simply by virtue of
experience only. They still have to go through the test which is supposed
to weed out the highly experienced know-nthings.
>
> As for the random government contact bit, I think that would
> work a
> bit better, since colleagues "tend" to have a better view on
> the
> technical skills of each other rather than the manager.
Exactly. I think everybody agrees that it's no cakewalk to get a government
TopSecret clearance.
>
> > Now again, could this procedure be corrupted? Of course.
> It's not perfect,
> > no solution is. But I believe it's still a substantial
> improvement over
> > what we got right now. Like I said, common sense dicates
> that a carresser
> > CCIE is on average better than a lab-rat ccie. Both guys
> have passed the
> > test, but at least the carresser has worked on real network.
> True, he
> > didn't do much on that real network, but he still did more
> than the labrat
> > who has, by definition, never worked on a real network.
>
> Oh no doubt, I understand NO system can be 100% perfect.
> However,
> this solution eliminates potentially very bright individuals
> with
> less years of experience, but potentially significantly much
> higher
> quality of experience. I suppose without statistics here, you
> could
> easily argue that sample is too small. If I do see some solid
> statistics on it, I will agree with you then.
Again, it doesn't eliminate those kung-fu masters forever, it just forces
them to wait. Is that really so bad?
>
> If the sample is too small, fine. If it is a fair amount, I
> think we
> should reconsider. You are trading one set of "bad apples"
> (lab
> rates) for a set of "bad oranges" (router carressers) and
> demolishing
> innocent candidates (the people who should be certifying) in
> the
> process.
The oranges are less rotten than the apples. And again, I am demolishing no
innocent people, just forcing them to wait.
>
>
> -Carroll Kong
>
>
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=71501&t=71143
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
