Heartily agreed - I see that one of replies below already demonstrates your point all too well, the unfair (and in my experience grossly elitist and simplistic) characterization that anyone who doesn't have a high GPA simply didn't try hard enough. This is especially relevant in our field.
For one thing, I've only run into a handful of software companies who have a hard grade cutoff anywhere. None of the big names (MS, GOOG, etc) do it, and IMHO if you understand what makes a great hacker, you wouldn't do it either. NVidia was the last company of note I dealt with that had a hard grade cutoff. I didn't make it - and many other talented programmers I know didn't either... their loss I suppose. I do know the people who were selected, and none of them can really be called good hackers - at most competent coders.
I've seen this pattern time and time again - in college the ones with the highest grades were the least employable. I had the pleasure of being in a program that required internships, and invariably the people with middle-of-the-road marks (3.5 is well above middle) had a much easier time finding employment than the ones at the top. The amount of effort it takes to stay at the top precludes one from doing the side projects that invariably makes the main difference between a merely competent programmer and a talented hacker. The top marks had no problem landing interviews, but had trouble converting interviews to offers.
Sure, you can claim that a truly elite individual can make the grades and do all the side hacking projects and make it home to cook dinner. Reality of the matter is, these guys are exceptionally rare, and odds are you're not paying half as much as you would need to hire one of them ;) By hiring in that range you're getting an unproductive mix of academic overachievers mixed with a handful of pure geniuses. By targeting a lower grade bracket you're getting a much more even mix of people good at the job you want them to do, but aren't so good at academics.
I'm just being honestly direct. You cannot completely throw out GPA because it is a measure of something. And you cannot completely rely on GPA because it is, as in all things, somewhat biased.
The people with the low GPAs will continue to say GPAs do not matter. The people with the high ones will continue to say it does. All I'm saying is that, at a target of 3.5, it doesn't preclude you from accomplishing other endeavors during your university years.
> You cannot completely throw out GPA because it is a measure of something.
Of course you can. Nobody is saying the GPA doesn't measure something, the question is whether or not it measures something relevant to the hiring decisions of your shop. From what I've seen, there is actually a negative correlation between grade and employability.
> All I'm saying is that, at a target of 3.5, it doesn't preclude you from accomplishing other endeavors during your university years.
Yes it does - from personal observation to boot. My peers in college with the 3.5 GPA were almost all considerably less capable than the people who stayed up late hacking and sacrificed marks as a result. There were a couple of outliers (who were an absolute joy to work with), but they are also exceptionally rare.
The question is whether or not believe what it takes to get the grades has relevance to what you do after graduation. I'm firmly of the opinion that it does not: every major skill I've ever learned that has helped me become a better programmer has not come from a classroom or the lips of a professor. Classes were really just a nice supplement to the hacking I was already doing, not the other way around. Like a great many other crafts, practice makes perfect, and while a small minority of the population can make the grade and hack a lot, the vast majority cannot. Deliberately placing yourself in a pool with a very few number of good candidates, and a flood of academic overachievers, is not an optimal hiring strategy.
In hindsight though, this filter is not precisely a bad thing. From past experience, the companies that have hard grade cutoffs are also the ones I do not wish to work for - so perhaps it is a good idea that the industry seems to have segmented itself this way.
Haha, hey potatolicious, I just realized you go to Waterloo and I go to CMU. Those are two schools who are bitter rivals in the regional ACM ICPC programming competition.
I just found it ironic and actually quite amusing and figured I'd let you know ;)
Of course you can. Nobody is saying the GPA doesn't measure something, the question is whether or not it measures something relevant to the hiring decisions of your shop. From what I've seen, there is actually a negative correlation between grade and employability.
Sigh. I could try to convince you that your personal observations are not representative, and you could probably tell me the same thing, and we would end up yelling past one another. We could both scream for data, and probably disregard the data that doesn't back our viewpoint and continue yelling past one another.
The problem is that we both have a vested interest in believing what we do. You've already admitted that you have a GPA that probably would not satisfy most hard-line GPA targets. I'll admit up-front that my GPA likely would satisfy most company's targets; this means I already lose a gold star.
This is a shame. I have not drawn any conclusions just from your self-confessed low GPA, but you have already prepared to consider me irredeemable. I'd probably ask some probing questions, realize that you've been bloody busy outside of classwork, and not think too much of it. This strategy could work both ways, for what it is worth.
I could just decide that if I were rejected at a company where you were vetting me, I am perhaps better off not being there. If I were hired, I'd have to deal with the chip on your shoulder about how I have already shown I can't hack because of some single arbitrary metric. But that would be out of the spirit of this site, and would assume things that probably aren't true about the users. So, instead, I'm going to try to provide a counter-argument.
Here's what I can say in my defense: I got that GPA without really giving a damn about my GPA. I found the topics I was working on fascinating enough without the motivation of grades, and got good enough at them that I was able to perform well on tests and assignments anyway. I guess I could've gone farther and just not done the assignments and skipped the tests, but requiring that much disinterest seems petty.
I wasn't trying to prove myself to anyone. And I still found time to play with things and hack on things of my own outside of the classroom. The only big strike against me, perhaps, is that I didn't get involved at all in open source until later on; I never published most of the things I was tinkering with in college. It doesn't mean I wasn't tinkering though, just that I wasn't public about it.
On top of that, I had an entire network of machines in my dorm room that the administration would've thrown a fit about had they known; I can only imagine what the power bill looked like. Having this taught me a bunch of things a networking theory class never would.
On the other hand, my networking theory class allowed me to make connections from experience to theory that made a whole lot of other things become obvious almost as soon as I was exposed to them. I drag this quote out from Deming all the time, because I think it is the best argument for not just focusing on the immediately practical: "Experience by itself teaches nothing... Without theory, experience has no meaning. Without theory, one has no questions to ask. Hence, without theory, there is no learning." FWIW, I haven't stopped learning just because I left college; to be honest, I think I would go stir-crazy if I did.
Hopefully, what I've written above makes some sense. Unfortunately, I've been shot by your generalization, and would not have been able to explain any of this had you said this in a group where you also controlled admissions. I think hard grade cutoffs in both directions suck, because they make a universal predictor from a weak indicator of arbitrary performance through unknown motivations.
Please, for me, and those like me who arrived where we are without the motivation of sucking up to our superiors and fitting the mold, don't be so hasty at nailing us to the wall. You'll just end up cultivating bitterness, and groups of people who don't want to work with one another based on something completely arbitrary.
They interviewed me :P (no, didn't get the job), My GPA was about 2.8 at the time. Also, if Google has a mark cutoff, it is not made public (suggesting that there are circumstances where they will ignore a low grade).
NVidia on the other hand made their intentions very clear, and basically said to not bother if your GPA was below 3.5, right in the job description.
You would be highly incorrect. Most employers have a border of a 3.5 GPA or higher as a cutoff. Beyond that, they really care little. If you have a 4.0, you might get some extra "points" in their eyes.
And for those of you who don't believe that 3.5 is a fair cutoff, I would challenge you by saying this: I attend one of the most difficult universities in the United States, and I do not know of a single person at the university who could not achieve a 3.5 if he or she only bothered to focus. Alas, our average GPA is closer to a 2.5.
Yes, GPA is an easy filter, but for good reason. Companies want the very best. And the very best can easily achieve a 3.5 without trying and still have time to do everything this article suggests.
"I do not know of a single person at the university who could not achieve a 3.5 if he or she only bothered to focus."
This is total bullshit. First, there are many people that barely made it into college (most of whom aren't in computer science/engineering), and no matter how much they worked, could not get a 3.5. Second, by saying "if he/she only bothered to focus," you're claiming that everyone who doesn't get a 3.5 is careless, lazy, or both.
Everyone on this forum could achieve a karma rating of 3,000. And for those of you who don't believe that 3,000 is a fair cutoff, I would challenge you by saying this: I have been on HN for over a year, and I do not know of a single person on HN who could not achieve 3,000 if he or she only bothered to focus.
The reason I don't have a 3.5GPA is the same reason I don't have 3,000 karma: I enjoy both school and HN, but there is a diminishing rate of return for effort. If I do the minimum amount of effort in school to get a 2.75, then work for 40 hours a week at my company, and still hang out with my friends, I really have no regret that I never "bothered" to get a 3.5.
Grade point averages are a result of intelligence, commitment, and personal priorities. Simplifying them the way that you do is both crass and incorrect.
It depends on the school as well. The "3.5 cutoff" is the result of massive grade inflation in recent years. By comparison, my school, Harvey Mudd, has it such that 3.0 is the cutoff for the Dean's list. This seems absurdly low by comparison to most schools, yet doesn't it seem ridiculous for everyone even mildly competent to graduate with a B+ or above?
The current system encourages schools to raise grades solely to get more people to pass the 3.5 cutoffs, allowing them to say that more graduates got good jobs, creating a grade inflation feedback loop.
Perfectly valid comparison: both karma rating on HN and GPA are largely irrelevant numbers that have little bearing on how good of a programmer you are. Both numbers would, to the layman, suggest some kind of authority, but in reality do not.
Both numbers are useful for theoretically filtering for the best, but in reality... do not.
Wow, the more I think of it, the better this analogy gets!
Are you really saying that the grades you get in college courses are even remotely equivalent to a measure of how much random anonymous people agree with you on the internet? Granted, my GPA is much better than my karma, but I still don't see how thinking about math or CS problem sets all day is anything like filling the internet with your opinions/favorite links?
I've heard of instances - specifically in legal recruiting - where there's also a top limit filter. That is, some firms (top level of the second tier) won't interview graduates with a GPA over 6.0 (I'm in Australia - GPA ranks to 7.0).
I spoke with a legal recruiter about this once. The logic is that the very high GPA students are almost certainly too interested in academia, likely to go back to academia (and so leave the firm), and those that don't want academia probably worked hard for that GPA because they only want to work for a soul-crushing top tier firm.
When I pointed out exceptions to this (including my beautiful, smart and focused lawyer wife), the recruiter just shrugged. It's much easier to miss the occasional exception, than waste time on the majority where the rule applies.
Because it doesn't take time. I have a friend here at Carnegie Mellon that has a 4.0 GPA, got 3rd at ACM ICPC World Finals, and also took the 3 hardest classes this university has to offer, as well as double the average course load, all in the same semester. And he still got 9 hours of sleep a night.
By all means, I'm nowhere near as smart as he is and I personally can't do that, but if you plan your semester well, you will have time to do "more productive things" and still have a 3.5
The point isn't to "have time" the point is to have the most time possible. If you shot for a 4.0 that's going to take more time than shooting for a 2.0. In my mind it was always better to take the time saved by shooting for merely passing my classes and invest it into side projects than it was to invest that time to get a 4.0.
That didn't stop me from taking over the average course load and piling on difficult CS classes concurrently after having professors sign me out of pre-reqs, but since I shot for a 2.5 I didn't just "have time" for other things. I had lots of time for other things.
You keep asserting that it's important to invest time in high grades and you can do that while still having time for other things. I agree that one could do that, but I don't agree that investing time in a high GPA is really necessary or desirable in most circumstances.
There's a fairly wide variation in grade inflation between schools, which is why a lot of the top grad schools will adjust GPAs for students from schools with grade inflation, or lack thereof, compared to the average.
Companies would do well to do the same, or to simply work on a case by case basis. Sadly, most companies recruit from a fairly small set of schools, and have little knowledge of the grade inflation differences.
If GPAs were the worst of it, then it wouldn't be so bad, but usually the ignorance runs much deeper. I once saw a recruiter for IBM Research's Extreme Blue internship program complain to the audience that he was wasting his time talking to a room full of students, because he could be spending it talking to MIT kids instead -- he was addressing a room with 100+ CS majors from, among others, Caltech, Harvey Mudd, USC, UCLA, UCSD, Cal Poly SLO, and Pomona College.
(As for Extreme Blue, thats interesting because I did their internship last year and the people at my lab were from UCLA, Cal Poly, USC, (and obviously CMU) among others. No MIT though.)
By design, the average GPA should be close to 2.5. Most schools try to modulate the difficulty of their courses to maintain a C+ average year-to-year. In my engineering school, they had a schedule of expected class averages for each year in the program; professors that wanted to assign significantly higher or lower grades had to justify their decision to the department head. After all, if the class average is 3.5, there's no headroom for above-average students to distinguish themselves.
However, I agree that any individual who wants to put in the work can probably achieve a 3.0 or better any most schools. Most people just don't want to do that much work.
But in Singapore, maybe in other Asian countries too, getting a good grade is not that hard, but it is very time consuming. For me personally it is a choice between high GPA or personal projects, achieving both is almost impossible.
That's true. I had a near 4.0 (one A-, gah) as a senior before I said "fuck this noise" and dropped out. I also had a family, was active in a couple campus clubs, and held a full time job as a developer. I was busy, but it's not impossible.
