For an interesting opinion piece about why standardized tests are not a good determining factor when hiring new employees, read Paul Kimmel's latest article:
http://www.codeguru.com/columns/VB/article.php/c10283/
Printable View
For an interesting opinion piece about why standardized tests are not a good determining factor when hiring new employees, read Paul Kimmel's latest article:
http://www.codeguru.com/columns/VB/article.php/c10283/
I know that I get frustrated with companies that use HR departments to give people technical tests and then use the results as a screening tool.
The second job I ever had involved a technical test using C. I didn't know the answers to half the questions. There were things about pointers to address to pointers to static locations that needed to be dynamically allocated, etc. etc... The levels of indirection were twisted so that it wasn't simple.
I ended up saying that I didn't know the answers to several of the questions, but then discussed the questions and speculated based on what I did know. By the time we had walked through the test, I'd learned quite a bit about some of the obscurities in C. I also ended up getting the job.
What helped to get me the job was the fact that I didn't claim to know answers I didn't have as well as the interest I showed in wanting to learn the answers. It showed a level of aptitude on my part.
The scarey thing was that I realized many months later that all the obscure things I'd been question about were actually in use on the project we were building. oye.
As to there being too much to know..... I gave up trying to retain everything there is to know technically a long time ago. Just look at the number of classes and such in the .NET Framework. Nobody can be an expert on all of it. If you also mess with VC++ and MFC, you add on top of that. Throw in a little bit on Java and you could sink ships with the amount of information. 99% of it, however, can be found quickly so there is no point in allocating brain cells for it. I may not be able to answer a question immediately, but give me 5 minutes and I bet I can :D
Brad!
Using HR departments to give tests over the phone is only usefull if the tests are multiple choice, as otherwise they have to interpret your answers if they don't match exactly the answers that the tech people give them.
But in general I've found that tech tests are hit and miss when it comes to getting the job. I've had tests, such as on C, where I've got fewer than half the questions right and got the job. At the other extreme I've got 23/25 and not got the job!
I've heard before about the interviews at Microsoft, and I agree with Kimmel that it may be the best test, but not all companies have and army or people that can afford spending an entire day with a single person. Imagine a company with 5, or 10, or 20, or even 50 employees, spending one day with each person. Just not feasible, IMO.
As for questions like
I've heard (from friends) about similar questions asked in interviews at a big IT company in Timisoara. Personally I was a little bit disappointed on how the interviewers reacted at some answers. When you ask how would you move mt. Fuji I think you should be a little bit opened for the solutions provided.Quote:
why are manhole covers round? How would you move Mt. Fuji? How would you go about counting all the stars in the sky?
PS: If someone asked me how would I move the Fuji mountain, first I would ask:
- How do the people of Japan feel about it?
- Are you aware of the enviromental implications of the task?
- Why does it need to be moved?
- Where does it have to be moved?
- What is the timeframe to move it?
- Are you sure the beneficiaries can pay the bill?
Depending on the anwers I could formulate an answer. And with a little hope the interviewer would start to doubt about this project before I answer. :D
I must agree that tests are a poor way to evaluate the abilities of a programmer. So, why are they so popular? I'm always thinking the world of computers is not run by computer people but by business people - not by geeks but by MBAs. I also believe the sensibilities that drive both may not only be different by complete incompatible. I wonder if this has anything to do with it.
About those facts: I think it's much better to ask "how'd you design such thing?" if respondent doesn't know the answer. Sensible answer to such question is what matters, not knowing the original fact.
Talking about tests and business people:
A friend went to a bank (for a non-IT related position) and he was asked the following:
"You are on a zeppelin with two friends and the zeppelin starts to lose height. In order not to crush smne has to fall off the zeppelin. How would you convince your friends that you should stay on it."
The guy answered:
"I would not try to convince none of my friends. I would be the one to fall first."
The manager smiled and said "Interesting. It was the first time I got such an answer."
He failed the interview (of course).
I really believe the guys in this bank are total morons. :mad:
"Because I'm a banker and, after lawyers and estate agents, if there's one thing the world doesn't have enough of, it's bankers!"Quote:
Originally Posted by panayotisk
Back on subject... I once had an interview where i had to sit a full day of tests, then at the end of it they sat each candidate down in front of a panel of interviewers. The first question the panel asked was "what did you think of todays testing?". I then proceeded to say how poor I thought it was, that the tests did not test for aptitude, and the only thing the test was going to show was which candidate had the biggest ego.
They thanked me for my time and that was it. But I thought my analysis was correct. Some of the tests were group things - as a group we had to solve a task. All that happened was that half the group started shouting. The thinkers would try to calm things down and get things organised, but would constantly be overidden by the people trying to prove what great "leaders" they were. And after the alloted time, nothing had actually been done.
At the end of the day, the organisation had spent countless dollars testing candidates, but I can't see that they actually learnt anything about 90% of them
Interesting.....Quote:
Originally Posted by Zeb
My first thought when reading this comment was that interviews are not for telling the truth, but rather, are for telling the people what they want to hear so you can get the job. ((More importantly, they are to find out if the company is one you really want to work for.))
(That is not to say you should lie-- because you shouldn't. It is only to say that answers in an interview should be a combination of what you believe, but presented in a manner that fits with what you believe works with the organization and people you are interviewing with. )
Interesting... How would the answer change depending on how the Japanese felt?Quote:
Originally Posted by cilu
If they objected to the project, then my guess is that you'd have to introduce some kind of mind-controlling agent into Japan's water supply prior to moving Mt. Fuji.Quote:
Originally Posted by Rigel
I agree that a certain amount of tact is probably called for, and perhaps that was lacking if Zeb was as blunt in his interview as he was in his post, but still, I don't think it's necessarily wrong to criticize the interviewing process if they ask you for an opinion on it. If they don't want to hear anything critical, then that question is just an invitation to spout some obsequious nothings. And of course, your point about finding out if the company is one you want to work for may very well have played into Zeb's response; if he felt the interviewing process was so ineffectual, maybe he got the impression that day-to-day work at that company would be equally so.Quote:
Originally Posted by Brad Jones
heh heh heh... :pQuote:
Originally Posted by Smasher/Devourer
I'd use Java.Quote:
Originally Posted by cilu
.NET won't work because the dirt would fall through the holes. With lots of java, the workers would by pepped up and ready to go! Obviously, java is the way to go in this case.
.... and I'd say it with a straight face ;)
Brad!
You would give such answer on the interview to Microsoft? :eek:Quote:
Originally Posted by Brad Jones
You know how I'd move Mount Fuji?
I'd take a C++ computer program and generate model of the mountain.
Separate this model into different fragments. (ie. like a puzzle)
Divide the mountain up into several horizontal layers.
Make a mold of each layer of the mountain.
Once I have a mold I'd assemble the first layer of the mold at the destination.
I'd start shoveling the dirt from Mount Fuji into the first layer.
Then I'd assemble the mold of the second layer and start shoveling dirt
into that mold ... etc .. until viola... we have moved Mount Fuji.
I'd move Mt. Fuji by driving past it in my car. Assuming the car to be one's frame of reference, the mountain is moving. (A cheap answer, I know. ;))
This was a long time ago (well, in my terms ;)). Wouldn't do that now. Of course, I'm glad I didn't get the job now, as I've learnt a lot about that organisation since then... :thumbd:Quote:
Originally Posted by Brad Jones
I can understand why they use tests in HR, to get some idea of the person's skills, and provide metrics for comparison. However, like the author of the article, I don't think that a multiple choice test is a good measuring test for problem-solving abilities, critical thinking, or creativity.
Yes, I did say creativity, because I think creativity is very necessary in a developer's arsenal for problem solving.
When I first read the title to your article, I thought you were talking about Software Testing. I thought it was interesting about Microsoft's interviewing process. I'm not in the hiring process, I'm on the other side of the desk taking the tests. One thing that really is quite aggravating is having taken classes in college for college credit, then being told I need to take tests for certification! Why! It should be enough that I have taken classes for months and earned the college credit. I understand about standardization, it's just an aggravation.
Good article, Thanks.
That's what I originally thought upon reading the question.Quote:
Originally Posted by Smasher/Devourer
Then I thought... just swap the names of Mt. Fuji and another mountain.
This way you'll be able to move it faster then light.Quote:
Originally Posted by Hermit
No no, cilu. If someone asking you these and really needs to move Fuji mountain, never will hire you.Quote:
Originally Posted by cilu
That's because, analyzing each point of your answer the deadline can be planned after one or two ice ages, it's nosense to explain in details.
However, there is a quick way to move the Fuji mountain:
Hire a dozen of interview consultants. For sure the probability that at least one knows how to do it is touchable.
Not like the tens of thousands developers who did not descover untill now a more trivial question: "Why is a manhole cover round?".
Well, I have few former colleagues working for Microsoft, but cannot ask them about the interview because, of course, they sign some papers to not reveal the secret of round manhole. :DQuote:
Originally Posted by Paul Kimmel
What I know from another one recently applied for. Each of those thumbs-ups is necessary to step to higher level of the interview.
If indeed, at the higher level a real developer can show a "thumb-down", all my respect, I'm throwing my hat.
/keep ahead and don't give up!
But the answer is simple: Because manholes are round. ;)Quote:
Originally Posted by ovidiucucu
And just in case this is followed by the question "Why are manholes round?": Simply because tubes with a circular cross-section are structurally the most stable...
The way that I also thought... For that I did't join MS. :D ;)Quote:
Originally Posted by gstercken
Q. Why AC is used for electricity transport?
A. Because, using transformers...
No, no. Just because, even Edison invented the electric chair to demonstrate how dangerous is AC vs. DC, he failed and lost the competition with George Westinghouse. But for sure in a parallel world in which Edison won, the humanity found the right solution to easily/cheap transform and transport DC. :D
As well, we can find advantages of the square manholes, and... ;)
EDIT:
Trying to construct a MFC-Extension library encapsulating the round manhole covers. I'm implementing now:
Just in case. Nobody can 100% guarantee that never a non-round one can appear.Code:class CRoundManholeCoverException : public CException
:D