|
|
| Author |
Message |
minnesotti
science forum beginner
Joined: 23 Jun 2006
Posts: 6
|
 Posted: Fri Jun 23, 2006 2:13 pm Post subject:
numerical modelling interview -- help
|
|
|
My supervisor's hinted to me that there are positions open to apply
for.
I checked who was on the selection board. My supervisor is the
selection board chair, however the immigrant who "failed" me in the
last interview is on the board, too.
He is a numerical modeller, and last time he was digging deep into my
claims I knew numerical modelling. It is his nature, it is how they do
the science in his home country. Last time, it looked like he did not
want to find out what I know and if I can do the job; instead, he
wanted to find out what I did not know.
The interview with him happened like this. He tells me:
-- Tell me about your experience with numerical modelling.
-- I did this project and that project, using the CFD packages X and Y.
-- So, what did you find out ?
-- I conducted the numerical experiment, and found the property Z and
W.
-- No, that's not the property Z and W. OK, tell me more. What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions... ah... and
solid wall boundary conditions... well, why are you asking me these
questions ? I am not a specialist in numerical modelling, however I
think what I know is enough for doing this low-level position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about your
experience in numerical modelling ?
....
....
....
Well, s.r.c. auditorium, I have a plan to write as little as possible
in the application when addressing the selection criteria. The idea is
to give that guy as little opportunity to sink his teeth into me as
possible. However, I am not an expert in numerical modelling, so that I
do not know where to draw the line between sharing the information
about the specifics of the task I set in the CFD modelling package, and
fobbing him off because he set to test me logical capability and minute
knowledge of modelling packages. Can you advise me on what should I do
at the interview ? Thanks.
\/ |
|
| Back to top |
|
 |
Hot_Potato
science forum beginner
Joined: 21 May 2005
Posts: 5
|
| Posted: Fri Jun 23, 2006 2:32 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
In the long run, if you hate numerical modelling, do no pursue it. It
will only make you miserable, in spite of the paycheck. Pursue the
activities that you love and you are good at. Personally I do not want
to see you whining for the next 10 years on the usenet how unhappy and
unsucessful you are after you decided to move to numerical modelling
and suffered trying to learn it , fake it or get along. Otherwise you
can make a huge effort to love it and be good at it. It is your
decision. |
|
| Back to top |
|
|
Straydog
science forum beginner
Joined: 26 Aug 2005
Posts: 32
|
| Posted: Fri Jun 23, 2006 3:04 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
On Fri, 23 Jun 2006, minnesotti wrote:
| Quote: | My supervisor's hinted to me that there are positions open to apply
for.
I checked who was on the selection board. My supervisor is the
selection board chair, however the immigrant who "failed" me in the
last interview is on the board, too.
He is a numerical modeller, and last time he was digging deep into my
claims I knew numerical modelling. It is his nature, it is how they do
the science in his home country. Last time, it looked like he did not
want to find out what I know and if I can do the job; instead, he
wanted to find out what I did not know.
The interview with him happened like this. He tells me:
-- Tell me about your experience with numerical modelling.
-- I did this project and that project, using the CFD packages X and Y.
-- So, what did you find out ?
-- I conducted the numerical experiment, and found the property Z and
W.
-- No, that's not the property Z and W. OK, tell me more. What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions... ah... and
solid wall boundary conditions... well, why are you asking me these
questions ? I am not a specialist in numerical modelling, however I
think what I know is enough for doing this low-level position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about your
experience in numerical modelling ?
...
...
...
Well, s.r.c. auditorium, I have a plan to write as little as possible
in the application when addressing the selection criteria. The idea is
to give that guy as little opportunity to sink his teeth into me as
possible. However, I am not an expert in numerical modelling, so that I
do not know where to draw the line between sharing the information
about the specifics of the task I set in the CFD modelling package, and
fobbing him off because he set to test me logical capability and minute
knowledge of modelling packages. Can you advise me on what should I do
at the interview ? Thanks.
|
If your current boss is going to be chair of this sellection committee,
how about you ask him for _hints_ ? ! Also, if you are going to fill out
some "background" on the application, then see if he will give you hints
on what to put onto the application, too. And, ask him if he can tell you
anything else that might help you "do better" on the interview that might
help you get that job.
Also, keep smiling as much as you can, keep sucking up as much as you can.
/\
|
|
| Back to top |
|
|
Shlomo Kashani
science forum beginner
Joined: 18 Jun 2006
Posts: 7
|
| Posted: Fri Jun 23, 2006 9:58 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
Dear minnesotti,
I don't think there is any place for racially prejudiced comments on
this board. Maybe instead of referring to him as "the immigrant",
consider TALKING to him to see why you failed the interview.
minnesotti wrote:
| Quote: | however the immigrant who "failed" me in the last interview is on the board, too.
It is his nature, it is how they do the science in his home country. |
|
|
| Back to top |
|
|
Hot_Potato
science forum beginner
Joined: 21 May 2005
Posts: 5
|
| Posted: Fri Jun 23, 2006 10:26 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
I dont think racism counts because they both originated from the same
country, former USSR. Both Pseudo-white trash. LOL
Shlomo Kashani wrote:
| Quote: | Dear minnesotti,
I don't think there is any place for racially prejudiced comments on
this board. Maybe instead of referring to him as "the immigrant",
consider TALKING to him to see why you failed the interview.
minnesotti wrote:
however the immigrant who "failed" me in the last interview is on the board, too.
It is his nature, it is how they do the science in his home country. |
|
|
| Back to top |
|
|
Old Pif
science forum beginner
Joined: 28 Jan 2006
Posts: 10
|
| Posted: Fri Jun 23, 2006 11:04 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
minnesotti wrote:
| Quote: |
-- I conducted the numerical experiment, and found the property Z and
W.
-- No, that's not the property Z and W. OK, tell me more. What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions... ah... and
solid wall boundary conditions... well, why are you asking me these
questions ? I am not a specialist in numerical modelling, however I
think what I know is enough for doing this low-level position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about your
experience in numerical modelling ?
|
Flow-through boundary conditions ... Holly Heavens! ... They are only
good for through-ass modeling.
With all due sympathy ... you looked totally miserable if that what
actually happened.
Let me give you my interpretation of the story: anybody who has tried
to solve at least one real problem numerically immediately noticed that
the outcome data requires physical interpretation and connection to
experiment. Knowledge and good understanding of how to set the
numerical model that is relevant to the specific experimental situation
is the core of numerical modeling. If a person does not have this the
job could not be satisfactory completed on any level - low or high -
does not matter. Your answers have demonstrated a) that you don't know
or understand all of the above and b) that you don't believe it is at
all important. So, the reaction was totally expected and justified.
If I were you I would read a good theoretical book that explains the
concepts of what you are suppose to model. With your experimental
background it would be very useful and interesting for you to see how
people implement it conceptually. Good understanding of experimental
techniques gives you additional advantage and could be advertised as a
plus because people with this combination of skills are very good with
interpretation of numerical data. But all that must be clearly
presented to the interviewer of course. |
|
| Back to top |
|
|
Old Pif
science forum beginner
Joined: 28 Jan 2006
Posts: 10
|
| Posted: Fri Jun 23, 2006 11:11 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
Aging_Recycled_Scientist wrote:
| Quote: |
Both Pseudo-white trash. LOL
|
Pseudo ...? What now quasi? |
|
| Back to top |
|
|
Shlomo Kashani
science forum beginner
Joined: 18 Jun 2006
Posts: 7
|
| Posted: Fri Jun 23, 2006 11:41 pm Post subject:
Re: numerical modelling interview -- help
|
|
|
Maybe he can use a PWTG (Pseudo White Trash Generator) instead of PRNG
to help him out with his numerical problems. I think that the PRNG is
biased or LOL ... even a racist PRNG (I).
Aging_Recycled_Scientist wrote:
| Quote: | I dont think racism counts because they both originated from the same
country, former USSR. Both Pseudo-white trash. LOL |
|
|
| Back to top |
|
|
Straydog
science forum beginner
Joined: 26 Aug 2005
Posts: 32
|
| Posted: Sat Jun 24, 2006 12:03 am Post subject:
Re: numerical modelling interview -- help
|
|
|
On Fri, 23 Jun 2006, Shlomo Kashani wrote:
| Quote: | Maybe he can use a PWTG (Pseudo White Trash Generator) instead of PRNG
to help him out with his numerical problems. I think that the PRNG is
biased or LOL ... even a racist PRNG (I).
Aging_Recycled_Scientist wrote:
I dont think racism counts because they both originated from the same
country, former USSR. Both Pseudo-white trash. LOL
|
There is another racist adjective I forgot that applies to all of us
besides "white trash" and that is "white trailor trash" (even lower than
ordinary redneck, Nazi, neo=nazi white trash). |
|
| Back to top |
|
|
minnesotti
science forum beginner
Joined: 23 Jun 2006
Posts: 6
|
| Posted: Sat Jun 24, 2006 4:23 am Post subject:
Re: numerical modelling interview -- help
|
|
|
Old Pif wrote:
| Quote: | Flow-through boundary conditions ... Holly Heavens! ... They are only
good for through-ass modeling.
With all due sympathy ... you looked totally miserable if that what
actually happened.
Let me give you my interpretation of the story: anybody who has tried
to solve at least one real problem numerically immediately noticed that
the outcome data requires physical interpretation and connection to
experiment. Knowledge and good understanding of how to set the
numerical model that is relevant to the specific experimental situation
is the core of numerical modeling. If a person does not have this the
job could not be satisfactory completed on any level - low or high -
does not matter. Your answers have demonstrated a) that you don't know
or understand all of the above and b) that you don't believe it is at
all important. So, the reaction was totally expected and justified.
If I were you I would read a good theoretical book that explains the
concepts of what you are suppose to model. With your experimental
background it would be very useful and interesting for you to see how
people implement it conceptually. Good understanding of experimental
techniques gives you additional advantage and could be advertised as a
plus because people with this combination of skills are very good with
interpretation of numerical data. But all that must be clearly
presented to the interviewer of course.
|
Thanks, Old Pif, for giving me the idea in which direction I should
develop the interview. I can indeed talk about my experimentalist's
intuition and how I verify the outcome of computational modelling where
I just plug the numbers into the software package. I even think that
this is exactly the skill they are looking for. However, my compatriot
scientists, especially of the "theoretical physics" and "numerical
modelling" ilk, like to talk not about how to overcome the challeneges
of the experiment (that's too complex for them and therefore not
interesting), but about the theoretical models which represent the real
(experimental) world. After they switch the discussion from the
experimental phenomenon to the representation of it with a theoretical
model, they start grilling the interviewee on the logical structure of
the model. They are trying to find out the limit of your capability to
logically think through the model. Of course, they are theoretical
physicists and they do this better than anyone else; it is hard to beat
them at their own game. Especially for me who does not have a clearly
expressed aptitude to logical thinking. I found this out many years ago
when I was an undergraduate student in my home country giving
presentation at the student scientific conference.
This compatriot is no exception from this "ilk". In the interview
conversation, I will be banking on the fact that the interview is
chaired by my supervisor who is no numerical modeller. I suppose this
is him who will be directing the conversation and perhaps limiting the
desire of my compatriot interviewer to steer the interview to the
discussion of the logical structure of the models representing my
experimental work. But it is not guaranteed that I can do this.
\/ |
|
| Back to top |
|
|
Phil Scott
science forum beginner
Joined: 27 May 2005
Posts: 26
|
| Posted: Sat Jun 24, 2006 4:34 am Post subject:
Re: numerical modelling interview -- help
|
|
|
"minnesotti" <minnesotti@yahoo.com> wrote in message
news:1151072037.368105.195440@m73g2000cwd.googlegroups.com...
| Quote: | My supervisor's hinted to me that there are positions open
to apply
for.
I checked who was on the selection board. My supervisor is
the
selection board chair, however the immigrant who "failed" me
in the
last interview is on the board, too.
He is a numerical modeller, and last time he was digging
deep into my
claims I knew numerical modelling. It is his nature, it is
how they do
the science in his home country. Last time, it looked like
he did not
want to find out what I know and if I can do the job;
instead, he
wanted to find out what I did not know.
The interview with him happened like this. He tells me:
-- Tell me about your experience with numerical modelling.
-- I did this project and that project, using the CFD
packages X and Y.
-- So, what did you find out ?
-- I conducted the numerical experiment, and found the
property Z and
W.
-- No, that's not the property Z and W. OK, tell me more.
What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions...
ah... and
solid wall boundary conditions... well, why are you asking
me these
questions ? I am not a specialist in numerical modelling,
however I
think what I know is enough for doing this low-level
position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about
your
experience in numerical modelling ?
...
...
...
Well, s.r.c. auditorium, I have a plan to write as little as
possible
in the application when addressing the selection criteria.
The idea is
to give that guy as little opportunity to sink his teeth
into me as
possible. However, I am not an expert in numerical
modelling, so that I
do not know where to draw the line between sharing the
information
about the specifics of the task I set in the CFD modelling
package, and
fobbing him off because he set to test me logical capability
and minute
knowledge of modelling packages. Can you advise me on what
should I do
at the interview ? Thanks.
\/
|
This interviewer has a friend, an India Tcch grad probably,
competent to a fair degree at least...coached by this
interviewer he can be more than 100% competent to do the job
and shine at the interview...he will work for less than you
pay in taxes. End of ballgame.
thats simply how it is these days.... so noq you and those
who read this are not advising their children to go into the S
and E fields because of this mess... it will just get
terminally worse at light speed as US worker skills atrophy
from disuse or become obsolete in 2 to 5 years.. or less.
(the LaRouch people have legislation they are pushing to stop
this by the way)
across all of US industry and science... we will loose our
brain trust.. and will be hiring our enemies to program our
weapons systems...that will be exciting.
accordingly... myself for instance am back in the trades...
that will last another 5 years, then those rates too will be
at starvation levels...
fortunately (or not so fortunately) I will be fully 'retired'
by then...others will get completely screwed.. .unless we
legislate worker rights and maintenance of our industrial
capacity,,,,and/or the US middle class unionizes with an
absolute vengence, and gets trade barriers errected ..that
will save a few of them.. .but the rest of the world,
willing to live in sqallor will still beat us to death in the
open market.
its greedy corporations and a corrupt US governance driving
this mess..both of those are well beyond critical mass
imo..and will not be stopped before they decimate the working
class and the nation those support.
In your case..... you will be working real cheap or you will
be gone in my view... As an engineer these days any potential
offers are half or a third of what I was getting in 1990
adjusted for inflation... not enough to cover the wear and
tear, I was surprised to see on this NG that even the PhD's
were getting screwed.
Working cheap guts a person from the insidem as these put out
more value than they recieve.
Even those in the higher pay brackets are ruined..but in
different ways... I spent some time at lunch in Silicon
Valley today...what a phony mess...nerds fitting themselves to
the corporate mold...fakery... no courage in these people...
its easy for these to become vicious and cowardly..that type
is short a heart as well.
You perhaps want to win in that environment. I did for
decades in the time frame it was on its terminal slide out the
bottom...
I learned too late.. the price I paid was almost fatal..it
took 10 years to recover, Some in that boat dont even
realize they have been decimated... they think its normal.
it isnt. Those lead diseased lives...they think thats normal
too,
OK so how to 'win' that interview... what I had done in the
past was burn the midnight oil for weeks, dry run all the
technology, and discovered all the achillese heels, or even
invented a few.....and developed solutions .... then ***T
controlled the agenda...I did not let the viciious idiots run
with the ball..and they will as you have mentioned.
YOU start the meeting before THEY have a chance to start the
meeting...YOU say... look...Id like to point a few problems in
this business that sink the average numerical modeling
project...the let em have it.. dont let em take a breath
literally... I refused to even sit at many of those
meetings...and I always arrived ahead of time and staked out
the head of the table.
.. YOU ask the interviewer if he knows why that is NOT so far
addressed? And what he does to address it..
I also had about 20 lbs of books with tags in them and
binders, which I made a point to drop on the table from about
2 inches up... as everyone was just seated...I kept standing.
These would always invite me to sit...I always declined their
generous offer.... I had anticipated their dumb ass questions
in advance and had slam dunk replies ready..not canned replies
but in depth research on the mess, the nature of the bogus
notion and the cure..I could go for hours on any of those
points.
I also didnt let anyone at all ever interrupt me...not for a
billionth of a second...if you allow that you will be toast.
That takes a personality type most dont have...so they get
screwed or they become a schooozing back stabber... I got that
personality from my earlier days on the motorcycle flat track
racing circuit...that was nasty. One mistake and all your
bones got broken... I did meeting with the same sort of
presence,
That worked then... I dont think it works now... that would be
seen as 'not a team player'... the back stabbing slime has
taken over these days... still...if they want to talk to me...
I stay standing. Catering to these jerks is entirely fatal.
you then become the leading vicious rat in the race.. unless
you develop a few other characteristics, a killer sense of
humor in the face of direct assault no less....but always.with
a dozen of them out to eat you alive...
I did that...'successfully'...thats how I got paid a
premium... the cost to me was nearly fatal...the money after
taxes...a pittance.. the entire mess is nasty...It is
apparently worse now.
These days its electrical work...or plumbing? I walk in,
tell em what its going to cost to fix or do the job (I get 75
dollars an hour or more)...if they dont like it...their toilet
or whatever can flood their living room up to their eye balls
for all I care... as I head off to do another job... which I
quit at 2 pm after a 4 to 6 hour day...I dont like working
myself too hard.
My few industrial clients expect me to lie for them though...I
dont...they hate that...so they send in their beat down idiots
who screw things up
....when it all hits the fan they call me to fix it...I get 75
dollars an hour...I show their client what the errors were..
fix it up... and let the chips fall where they may... lies
never work well. I tell em, no one is perfect, when we mess
something up a little...we simply fix it.
People that dont mind being low paid drones can stay in
corporate america..not much more damage can be done to them..
those with real brains, or who want to earn more than the cost
of maintenance, and anyone with a conscience need to get chear
the hell out of corporate america.
Those working should develop a plan B and get it started on
weekends...so that they can leave and still survive.
The best plan B's involve accumulationg rentable assets if you
want to end up as a heartless rentier that is... or developing
a killer skill set, and staying unbelievably healthy in this
very unhealthy world...thats been my approach. I think it
will play better at 85 than most other options... even being
rich..the rich dont do particularly well... no challenge,.. a
challenging life is not seen as 'success' by most people,...
that standard is a key part of the problem.... a culture that
seeks and targets for eaxe, destroys itself....we grow and
stay strong under productive stress. key word is productive.
The stress inflicted on us by corporations and bogus
government is not productive...it fatal.
Phil Scott
> |
|
| Back to top |
|
|
Phil Scott
science forum beginner
Joined: 27 May 2005
Posts: 26
|
| Posted: Sat Jun 24, 2006 4:40 am Post subject:
Re: numerical modelling interview -- help
|
|
|
"Straydog" <asd@panix.com> wrote in message
news:Pine.NEB.4.63.0606231101480.14299@panix2.panix.com...
| Quote: |
On Fri, 23 Jun 2006, minnesotti wrote:
My supervisor's hinted to me that there are positions open
to apply
for.
I checked who was on the selection board. My supervisor is
the
selection board chair, however the immigrant who "failed"
me in the
last interview is on the board, too.
He is a numerical modeller, and last time he was digging
deep into my
claims I knew numerical modelling. It is his nature, it is
how they do
the science in his home country. Last time, it looked like
he did not
want to find out what I know and if I can do the job;
instead, he
wanted to find out what I did not know.
The interview with him happened like this. He tells me:
-- Tell me about your experience with numerical modelling.
-- I did this project and that project, using the CFD
packages X and Y.
-- So, what did you find out ?
-- I conducted the numerical experiment, and found the
property Z and
W.
-- No, that's not the property Z and W. OK, tell me more.
What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions...
ah... and
solid wall boundary conditions... well, why are you asking
me these
questions ? I am not a specialist in numerical modelling,
however I
think what I know is enough for doing this low-level
position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about
your
experience in numerical modelling ?
...
...
...
Well, s.r.c. auditorium, I have a plan to write as little
as possible
in the application when addressing the selection criteria.
The idea is
to give that guy as little opportunity to sink his teeth
into me as
possible. However, I am not an expert in numerical
modelling, so that I
do not know where to draw the line between sharing the
information
about the specifics of the task I set in the CFD modelling
package, and
fobbing him off because he set to test me logical
capability and minute
knowledge of modelling packages. Can you advise me on what
should I do
at the interview ? Thanks.
If your current boss is going to be chair of this sellection
committee, how about you ask him for _hints_ ? ! Also, if
you are going to fill out some "background" on the
application, then see if he will give you hints on what to
put onto the application, too. And, ask him if he can tell
you anything else that might help you "do better" on the
interview that might help you get that job.
Also, keep smiling as much as you can, keep sucking up as
much as you can.
|
You gave practical advice...these days its suck up or
loose.
However on that issue, as you probably know its fatal for the
person,,, one justifies the behavior, being subservient...and
soon has no balls.
in this world, eventually its only ones own courage and
integrity that will save him...in the final analysis
we have seen what catering has done for those in the S and E
business. These days though, not catering is fatal to ones
career.. so I let it go on that basis. it seems you have
too.
Phil Scott
|
|
| Back to top |
|
|
Phil Scott
science forum beginner
Joined: 27 May 2005
Posts: 26
|
| Posted: Sat Jun 24, 2006 5:13 am Post subject:
Re: numerical modelling interview -- help
|
|
|
"Old Pif" <OldPif@gmail.com> wrote in message
news:1151103869.460253.28130@y41g2000cwy.googlegroups.com...
| Quote: |
minnesotti wrote:
-- I conducted the numerical experiment, and found the
property Z and
W.
-- No, that's not the property Z and W. OK, tell me more.
What
equations did you use ?
-- Mmm... I think the solver was Eulerian.
-- And ?
-- Ah... uhm... the problem was set in two dimensions.
-- What else did you do ?
-- Uhm... I set the boundary conditions...
-- What boundary conditions did you use with this solver ?
-- Uhm... those were flow-through boundary conditions...
ah... and
solid wall boundary conditions... well, why are you asking
me these
questions ? I am not a specialist in numerical modelling,
however I
think what I know is enough for doing this low-level
position job, and
I have already told you that. Do you agree with me ?
-- I do not agree with you. What else can you tell me about
your
experience in numerical modelling ?
Flow-through boundary conditions ... Holly Heavens! ... They
are only
good for through-ass modeling.
With all due sympathy ... you looked totally miserable if
that what
actually happened.
Let me give you my interpretation of the story: anybody who
has tried
to solve at least one real problem numerically immediately
noticed that
the outcome data requires physical interpretation and
connection to
experiment. Knowledge and good understanding of how to set
the
numerical model that is relevant to the specific
experimental situation
is the core of numerical modeling. If a person does not have
this the
job could not be satisfactory completed on any level - low
or high -
does not matter. Your answers have demonstrated a) that you
don't know
or understand all of the above and b) that you don't believe
it is at
all important. So, the reaction was totally expected and
justified.
If I were you I would read a good theoretical book that
explains the
concepts of what you are suppose to model. With your
experimental
background it would be very useful and interesting for you
to see how
people implement it conceptually. Good understanding of
experimental
techniques gives you additional advantage and could be
advertised as a
plus because people with this combination of skills are very
good with
interpretation of numerical data. But all that must be
clearly
presented to the interviewer of course.
|
I certainly wont disagree with that.... however in my recent
experience one would be fortunate to find an interviewer who
had it straight themselves....in my business most know only
their specialties and that to a 20% level...beyond that they
have it all sideways..and accept only wrong answers.
Some that have interviewed me and recommended a hire have been
overruled by some jerk in the background for reasons tha tbat
baffleed both of us... politics... and bogus criteria....utter
insanity, gone ballistic.... and its self destructing them
also.
the mind boggles..
Phil Scott
> |
|
| Back to top |
|
|
Phil Scott
science forum beginner
Joined: 27 May 2005
Posts: 26
|
| Posted: Sat Jun 24, 2006 5:18 am Post subject:
Re: numerical modelling interview -- help
|
|
|
"minnesotti" <minnesotti@yahoo.com> wrote in message
news:1151123023.005357.312710@g10g2000cwb.googlegroups.com...
| Quote: |
Old Pif wrote:
Flow-through boundary conditions ... Holly Heavens! ...
They are only
good for through-ass modeling.
With all due sympathy ... you looked totally miserable if
that what
actually happened.
Let me give you my interpretation of the story: anybody who
has tried
to solve at least one real problem numerically immediately
noticed that
the outcome data requires physical interpretation and
connection to
experiment. Knowledge and good understanding of how to set
the
numerical model that is relevant to the specific
experimental situation
is the core of numerical modeling. If a person does not
have this the
job could not be satisfactory completed on any level - low
or high -
does not matter. Your answers have demonstrated a) that you
don't know
or understand all of the above and b) that you don't
believe it is at
all important. So, the reaction was totally expected and
justified.
If I were you I would read a good theoretical book that
explains the
concepts of what you are suppose to model. With your
experimental
background it would be very useful and interesting for you
to see how
people implement it conceptually. Good understanding of
experimental
techniques gives you additional advantage and could be
advertised as a
plus because people with this combination of skills are
very good with
interpretation of numerical data. But all that must be
clearly
presented to the interviewer of course.
Thanks, Old Pif, for giving me the idea in which direction I
should
develop the interview. I can indeed talk about my
experimentalist's
intuition and how I verify the outcome of computational
modelling where
I just plug the numbers into the software package. I even
think that
this is exactly the skill they are looking for. However, my
compatriot
scientists, especially of the "theoretical physics" and
"numerical
modelling" ilk, like to talk not about how to overcome the
challeneges
of the experiment (that's too complex for them and therefore
not
interesting), but about the theoretical models which
represent the real
(experimental) world. After they switch the discussion from
the
experimental phenomenon to the representation of it with a
theoretical
model, they start grilling the interviewee on the logical
structure of
the model. They are trying to find out the limit of your
capability to
logically think through the model. Of course, they are
theoretical
physicists and they do this better than anyone else; it is
hard to beat
them at their own game. Especially for me who does not have
a clearly
expressed aptitude to logical thinking. I found this out
many years ago
when I was an undergraduate student in my home country
giving
presentation at the student scientific conference.
This compatriot is no exception from this "ilk". In the
interview
conversation, I will be banking on the fact that the
interview is
chaired by my supervisor who is no numerical modeller. I
suppose this
is him who will be directing the conversation and perhaps
limiting the
desire of my compatriot interviewer to steer the interview
to the
discussion of the logical structure of the models
representing my
experimental work. But it is not guaranteed that I can do
this.
|
In that case I would make a fast study of the flaws
inherent in modeling and analysis out of context with the real
world or any other aspect I could imagine...and I would drive
the interview in that direction from the start...it would be
on the application summary or whatever also... I I would find
fault with a focus limited to the logic of the experiement
etc.
I would point out that one finds what he is looking for..and
that models are often staged to preclude actual discovery..or
some such.
I would not engage them in their sand box...I would look for
holes and cat turds in their sand box...and then waive them in
the air for a while.
Phil Scott
|
|
| Back to top |
|
|
Old Pif
science forum beginner
Joined: 28 Jan 2006
Posts: 10
|
| Posted: Sat Jun 24, 2006 11:46 am Post subject:
Re: numerical modelling interview -- help
|
|
|
minnesotti wrote:
| Quote: |
I can indeed talk about my experimentalist's
intuition and how I verify the outcome of computational modelling where
I just plug the numbers into the software package. I even think that
this is exactly the skill they are looking for.
|
Regrettably, the current status of physics based modeling is such that
you CAN NOT just plug in numbers and get the answer. You probably need
more physical intuition than anything else to connect modeling and
experiment. Intuition is basically developing with practical work. You
must have enough of that from your experimental experience.
| Quote: |
However, my compatriot
scientists, especially of the "theoretical physics" and "numerical
modelling" ilk, like to talk not about how to overcome the challenges
of the experiment (that's too complex for them and therefore not
interesting), but about the theoretical models which represent the real
(experimental) world.
|
Well, naturally, people in such circumstances talk about what they know
best. However, the numerical modeling as far as job is concerned could
mean many different things. It could be development of the new models
that don't exist or it could be the usage of commercial packages. The
former requires the combination of theoretical physics, applied math
and programing, the later requires conceptual understanding of what is
behind the GUI. I understood that the later is your case. which makes
the whole story substantially easier as it is impossible to acquire all
three ingredients in short period of time whereas to read one book with
the description of concepts is quite possible and very beneficial.
| Quote: |
After they switch the discussion from the
experimental phenomenon to the representation of it with a theoretical
model, they start grilling the interviewee on the logical structure of
the model. They are trying to find out the limit of your capability to
logically think through the model. Of course, they are theoretical
physicists and they do this better than anyone else; it is hard to beat
them at their own game. Especially for me who does not have a clearly
expressed aptitude to logical thinking. I found this out many years ago
when I was an undergraduate student in my home country giving
presentation at the student scientific conference.
This compatriot is no exception from this "ilk". In the interview
conversation, I will be banking on the fact that the interview is
chaired by my supervisor who is no numerical modeller. I suppose this
is him who will be directing the conversation and perhaps limiting the
desire of my compatriot interviewer to steer the interview to the
discussion of the logical structure of the models representing my
experimental work. But it is not guaranteed that I can do this.
|
Well, it as I said depends on what the future jobs is. If they clearly
understand it for themselves of course. If this is development of the
new computational model from the ground up it is probably not your cup
of tea. On the other hand, if it is to be a commercial package then you
bound to what is available. You can not change anything you can only
choose between the available options. What is required in that case is
the demonstration that you can make the right choice consciously, that
you understand both the capabilities and the limitations of the
package(s) and that you know how to use the degrees of freedom that are
available.
Basically, you have two tasks for the interview preparation. The first
one is to inspire yourself with what is supposed to be your job and the
second to bring yourself to the level of discussion with the interview
panel.
For the first one consider the computational package as an experimental
setup where instead of handles, limbs etc. you have menus and mouse.
The rest is in principle the same - physical phenomenon that you simple
address from different perspectives.
For the second read 1) a good book with the description of concepts and
2) try to understand the capabilities and limitations of the
perspective package(s) and preferably find some comparative study of
them.
And don't demonstrate your total ambivalence to the perspective line of
work. Imagine reaction of conductor who learns from the first violin
that he hates the music. |
|
| Back to top |
|
|
Google
|
|
| Back to top |
|
|
|
|
The time now is Sat Jan 10, 2009 3:04 am | All times are GMT
|
|
Sprint Ringtones | Myspace Layouts | Buy Anything On eBay | Free Advertising | Debt Consolidation
|
|
Copyright © 2004-2005 DeniX Solutions SRL
|
|
Other DeniX Solutions sites:
Electronics forum |
Medicine forum |
Unix/Linux blog |
Unix/Linux documentation |
Unix/Linux forums
|
Powered by phpBB © 2001, 2005 phpBB Group
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Forum index
»
Science and Technology
»
Math
»
num-analysis
