dear Daan,
i would like to comment on the recurring themes here in Level 2:
> GW:>>Yes, but for one detail: I think that research is a skill that can be
> GW:>>certified. It is also within the scope of technical skills that a
sysadmin
> GW:>that we agree with what is said here......being able to research or pick
> GW:>on new services without having to be trained will be THE biggest
diffrence
> And does it then make sense to give a candidate (fragments) of a
> (ficticious) man page, and then ask questions about it. What i am missing
> in t1a is the more-then-just-by-head questions. These are quite possible
> and not beyond the scope of what lpic can test/certify.
>
the core issue seems to be aptitude in solving open-ended problems.
the bar exam may be a good parallel universe. the applicant knows the various
laws, principles, and procedures, but not how to solve several random and
complex "real life" problems to demonstrate mastery of the various components.
another example is the attainment of the black belt in martial arts. there are
many procedures and moves that one must learn, but the final trial consists in
demonstrating each of these moves in fluid motion.
the ability to solve non-specific linux problems would seem to involve an
intelligent ability to tail and diagnose log files, use of strace (involving an
understanding of system calls), understanding of compile-time errors and
makefiles, piped commands, and most important some proficiency in reading c
code and kernel source which is the underlying glue of the whole thing anyway.
the issue of knowing what the knowledge resources are, both in terms of the
many facets of the 'man' and 'info' commands along with /usr/doc, web
resources, knowledge bases, and results-oriented use of resources such as
metacrawler would also be at a premium.
if i was to test on a generic ability to think through a problem i would
probably pick a random area for a given test... such as remote printing using
lpr/lpd in a certain situation. i would tell the applicant they need to solve
a situation such as a printer works okay locally but for whatever reason remote
host station.xyz.com cannot print to it. the situation is such that they
cannot go on-site but must issue diagnostic procedures to a 3rd party. i would
request a military-like decision-tree procedure to give to this 3rd party as
instructions to fix any possible problem in that scenario. in this way, they
must demonstrate their scope of understanding of the required concepts on
demand.
this is my two cents; i agree that it is a difficult area that involves
wholistic thinking rather than command-oriented proficiency.
best regards,
stuart
javabaw, inc. dba linux labs by stuart trusty, president------- linux labs
..... .... 230 peachtree st nw ste 705b
linux labs atlanta.ga.us 30303
"mission critical linux" http://www.linuxlabs.com
24x7 dispatch: 800.788.9319 office 404.577.7747
---------------------------------------------------------------------------
On Tue, 1 Feb 2000, daan wrote:
> On Tue, 1 Feb 2000, Greg Wright wrote:
>
> GW:>On 31/01/00 at 22:56 daan wrote:
> GW:>
> GW:>>On Mon, 31 Jan 2000 [EMAIL PROTECTED] wrote:
> GW:>>
> GW:>>...
> GW:>>
> GW:>>...
> GW:>>
> GW:>>:>Q1: Do you feel that this summary fairly explains the expected
> GW:>>:>responsibility level of a level 2 candidate, while understanding that
> GW:>>:>LPI is unable to certify skills beyond a technical level?
> GW:>>:>
> GW:>>:>Q1 Answer and Comment:
> GW:>>
> GW:>>Yes, but for one detail: I think that research is a skill that can be
> GW:>>certified. It is also within the scope of technical skills that a sysadmin
> GW:>>should have. One might argue that it is beyond level 2, i'd disagree
> GW:>>though.
> GW:>>
> GW:>>...
> GW:>>
> GW:>>:>Q5: Without getting into technical specifics, are there other general
> GW:>job
> GW:>>:>aspects that you feel need to be addressed here?
> GW:>>:>
> GW:>>:>Q5 Answer and Comment:
> GW:>>
> GW:>>Once again (i'm boring myself here): dealing with unknown topics is the
> GW:>>main quality of a good sysadmin. You can't expect that at level 1, but
> GW:>>at level 2, it becomes more urgent. No guru knows it all so this remains
> GW:>>important thru all subsequent levels.
> GW:>>
> GW:>>Please, comment on this topic. I feel like i'm crying in the desert.
> GW:>>
> GW:>
> GW:>
> GW:>I think the above is what Paul and I have been debating about, I believe
> GW:>that we agree with what is said here......being able to research or pick up
> GW:>on new services without having to be trained will be THE biggest diffrence
> GW:>between level 1 and 2
> GW:>
> GW:>He/She will be a competant person not requiring a team to get new
> GW:>jobs/services done / up / working
> GW:>
>
> thanks,
>
> GW:>
> GW:>Does everyone agree here ?
> GW:>
>
>
> And does it then make sense to give a candidate (fragments) of a
> (ficticious) man page, and then ask questions about it. What i am missing
> in t1a is the more-then-just-by-head questions. These are quite possible
> and not beyond the scope of what lpic can test/certify.
>
> :wq!
>
> Daan Hoogland Unix consultants v v
> [EMAIL PROTECTED] OO developers \ /
> ### ## ## # >---X---<
> http://snow.nl # # # # # # # # / \
> Snow B.V. ## # # # # # # # ^ ^
> Tel. (+31)418-653333 # # # # # # # #
> Fax. (+31)418-653666 ### # # ## ## ##
>
>
> --
> This message was sent from the lpi-examdev mailing list.
> Send `unsubscribe lpi-examdev' in the subject to [EMAIL PROTECTED]
> to leave the list.
>
--
This message was sent from the lpi-examdev mailing list.
Send `unsubscribe lpi-examdev' in the subject to [EMAIL PROTECTED]
to leave the list.
