When I was a kid, I remember a friend of my mother's telling me about an interview question that the RAF had used in her day (which probably would have been the late 60s). The question was something like "How would you transport a giraffe across London?"
I think the point was not that they wanted a complete transport solution, but that they wanted to hear what questions the candidate would ask to solicit more details about the job. In particular, a question they were keen to hear was "Is it alive?"
I think your elephant answer is great. Especially coming up with a justification for the question. You should totally apply to Google and test it out :)
I like the giraffe question very much. Having done some giraffe transport myself, I would want to know "Is it an actual giraffe, as opposed to a stuffed-toy giraffe?".
"Is it alive?" is a great question, and makes me also want to ask "Does it need to be still alive on the other side of London?".
My actual answer would be a boring one of ringing round ZSL, Whipsnade, Woburn Safari Park, etc to find out what they do. But trying to work it out in my own head is fun and is an obvious thing to write if I have time later today...
You should totally apply to Google and test it out :)
:) My vague impression is they've gone off this sort of question.
I do like seeing how other people's brains work :)
It's funny - I first thought of the water displacement method but then reasoned that the most likley context to need an acurate weight rather than an aprocimation would be for medication and ruled out dunking a poorly elephant. After that I got as far as estmating volume and hoping there's a useful table of pachyderm densities to consult and was otherwise stuck. I wish i'd known enough about gravitational mapping to come up with your supremely geeky solution (I obviously have himework to do!)
I'm struggling to find accurate data about elephant density. I've found lots of people attempting the same question and so wanting to know what it is. I have found some scientific data about elephant density - but they mean how many elephants there are in a particular area!
The best thing I've found is this: "We had a question like this about a huge block of limestone when I was on a field trip with my geology class in college. Many of us whipped out notebooks to feverishly start trying to calculate it. The guy who won just read the plaque."
Oh, wow, awesome, thank you. Tiny tortoise in a spoon! Patient baby elephant being measured! Lemurs lounging on scales! A marmoset (?) nibbling food cautiously! Sealion splashing on to a platform on scales!
And they have a database of weight and other measurements for lots and lots of animals! Cute *and* data.
Aha! Now *that's* my dream job. Except I'm a little ambivalent about zoos. And when I start thinking about how I could get a job that's mostly dealing with data about wild animals ... I'm just trying to switch to a less well-funded and more competitive department rather than doing a totally different job.
On density I think most animals are around the density of water, certainly humans are. But there must be some variation; birds are less dense, because their bones are full of air (also air gets trapped in their feathers but I'm not sure how you count that on the aArchimedes principle
( ... )
Of course this means weighing any living animal by displacement of water will be inexact to the tune of their lung volume (including fish who vary their buoyancy with their swim bladder).I don't think this is the case for things that are floating, which includes fish. The force pushing them down precisely balances the force pushing them up - otherwise they would sink or float on up out of the water. And the amount of water pushed out of the way by the thing must weigh the same as the thing - otherwise there would be a net force pushing it up or pushing it down
( ... )
Oh, and on the many joyful tangents this sent me off on, I found out that captive elephants are becoming obese, which is fascinating to me in two ways. One, because it's yet more evidence blowing away the declining-moral-standards explanations for the similar weight gain in humans over time. Elephants are not playing more video games or getting greedier. And second, because measuring elephant's fat percentage is really tricky, but the person here has thought of a really neat way of doing it.
They don't spell it out - they just say deuterium, mass spectroscopy, and two simple blood samples. But that's enough to work it out!
My guess is they take one blood sample. Then inject the elephant with a measured quantity of heavy water (D2O) - or even semi-heavy water (DHO), and wait a bit. The water will quickly equilibrate through the elephant's tissues but not the fat, because exchange in to fat is slow (hours to days or longer). So a few minutes or half an hour later, they take another blood sample. They can then calculate how much not-
( ... )
*laughs* Alas, I think I might enjoy working for somewhere that asked that question.
If, for instance, the other interview questions were things like "From first principles, where can you look for well-paid jobs online these days?", "What's going on with death rates in the UK?", "What is a perfectly fair way of allocating annual leave in leap years vs non-leap years?", and "What does the evidence say about tooth brushing?" then it might well be an ideal place for me to work, given my recent LJ posting history.
A job that paid me handsomely to produce answers like these to questions like those on a slack schedule is close to a dream job for me.
Sadly, I can't think of a way to make that happen, apart from being Randall Munroe, and SFAICT he mostly makes money from consultancy/paid talks and book sales. Even Tom Scott doesn't make a living doing the Tom Scott thing - he does IT consulting stuff to pay the bills. Why yes, I have looked in to this in some detail ...
It's the fantasy aspect I object to. Ask me a real question and I'll come up with an answer, ask me ludicrous hypotheticals and it just annoys me. Same way I lose ny temper with IQ tests within approximately 30 seconds.
Our internal recruitment system runs on competence based questions. On applications it's about real stuff you've done, and in interview the questions are either on real stuff you've done or how you would deal with reality-based scenarios. Not an esoteric thought experiment in sight.
Comments 21
I think the point was not that they wanted a complete transport solution, but that they wanted to hear what questions the candidate would ask to solicit more details about the job. In particular, a question they were keen to hear was "Is it alive?"
I think your elephant answer is great. Especially coming up with a justification for the question. You should totally apply to Google and test it out :)
Reply
"Is it alive?" is a great question, and makes me also want to ask "Does it need to be still alive on the other side of London?".
My actual answer would be a boring one of ringing round ZSL, Whipsnade, Woburn Safari Park, etc to find out what they do. But trying to work it out in my own head is fun and is an obvious thing to write if I have time later today...
You should totally apply to Google and test it out :)
:) My vague impression is they've gone off this sort of question.
Reply
My vague impression is they've gone off this sort of question.
What, creative silliness about elephants isn't a reliable predictor of employablity? Nonsense :)
Reply
Reply
It's funny - I first thought of the water displacement method but then reasoned that the most likley context to need an acurate weight rather than an aprocimation would be for medication and ruled out dunking a poorly elephant. After that I got as far as estmating volume and hoping there's a useful table of pachyderm densities to consult and was otherwise stuck. I wish i'd known enough about gravitational mapping to come up with your supremely geeky solution (I obviously have himework to do!)
Reply
I'm struggling to find accurate data about elephant density. I've found lots of people attempting the same question and so wanting to know what it is. I have found some scientific data about elephant density - but they mean how many elephants there are in a particular area!
The best thing I've found is this: "We had a question like this about a huge block of limestone when I was on a field trip with my geology class in college. Many of us whipped out notebooks to feverishly start trying to calculate it. The guy who won just read the plaque."
Reply
(Also, tiny tortoise in a spoon!)
Cleveland had to weigh an elephant on four separate scales cos they didn't have a big one - which meant training the elephant... http://www.clevelandmetroparks.com/Zoo/How-Keepers-Weigh-Our-Elephants.aspx
Reply
And they have a database of weight and other measurements for lots and lots of animals! Cute *and* data.
Aha! Now *that's* my dream job. Except I'm a little ambivalent about zoos. And when I start thinking about how I could get a job that's mostly dealing with data about wild animals ... I'm just trying to switch to a less well-funded and more competitive department rather than doing a totally different job.
Reply
Reply
Of course this means weighing any living animal by displacement of water will be inexact to the tune of their lung volume (including fish who vary their buoyancy with their swim bladder).I don't think this is the case for things that are floating, which includes fish. The force pushing them down precisely balances the force pushing them up - otherwise they would sink or float on up out of the water. And the amount of water pushed out of the way by the thing must weigh the same as the thing - otherwise there would be a net force pushing it up or pushing it down ( ... )
Reply
They don't spell it out - they just say deuterium, mass spectroscopy, and two simple blood samples. But that's enough to work it out!
My guess is they take one blood sample. Then inject the elephant with a measured quantity of heavy water (D2O) - or even semi-heavy water (DHO), and wait a bit. The water will quickly equilibrate through the elephant's tissues but not the fat, because exchange in to fat is slow (hours to days or longer). So a few minutes or half an hour later, they take another blood sample. They can then calculate how much not- ( ... )
Reply
It's a very good interview question: if it is asked, you know that you cant possibly ever work for such people, so you dont need to worry about it.
Reply
If, for instance, the other interview questions were things like "From first principles, where can you look for well-paid jobs online these days?", "What's going on with death rates in the UK?", "What is a perfectly fair way of allocating annual leave in leap years vs non-leap years?", and "What does the evidence say about tooth brushing?" then it might well be an ideal place for me to work, given my recent LJ posting history.
A job that paid me handsomely to produce answers like these to questions like those on a slack schedule is close to a dream job for me.
Sadly, I can't think of a way to make that happen, apart from being Randall Munroe, and SFAICT he mostly makes money from consultancy/paid talks and book sales. Even Tom Scott doesn't make a living doing the Tom Scott thing - he does IT consulting stuff to pay the bills. Why yes, I have looked in to this in some detail ...
Reply
Edit: Ah, this would make sense if linked to your previous heartfelt advice not to work for the Civil Service.
Reply
It's the fantasy aspect I object to. Ask me a real question and I'll come up with an answer, ask me ludicrous hypotheticals and it just annoys me. Same way I lose ny temper with IQ tests within approximately 30 seconds.
Our internal recruitment system runs on competence based questions. On applications it's about real stuff you've done, and in interview the questions are either on real stuff you've done or how you would deal with reality-based scenarios. Not an esoteric thought experiment in sight.
Reply
Leave a comment