"The easiest way to show off your brains is to have had good grades in school. Just put your GPA on your resume and you're done."
Real bad litmus test.
I know lots of people who had a 4.0 who, if you dropped them off in the woods, would have just died.
OTOH, I know lots of others who were brilliant but too bored to care about grades.
Coding a one loop bubble sort in hex blindfolded while juggling eggs, playing 12 games of chess, solving a 100 x 100 sudoku, providing an alternate solution to the 4 color theorem, teleporting tonights winning Powerball numbers, and cooking dinner would be a more applicable barometer.
Unbelievable. The thing about fog creek is that they want highly competent technical engineers without any entrepreneur spirit. Joel is the leader and he needs code monkeys to carry out the work. People who make 4.0s are the exact kind of people they need-- discipline, reverence for authority, people pleasers-- people who won't rock the boat. For Fog Creek this is a great litmus test.
Think I'm full of it? Watch their little movie they put together then tell me I'm not spot on. (I hope you enjoy pimple faced programmers verbally fellating joel spolsky)
Incidentally, Google has a similar aesthetic, although not to the same degree. People who are drawn to startups are not usually the kind of people you want in a corporate structure.
Summary: Intern comes up with a job sales board for Fog Greek, grosses $1m. Joel decides not to compensate him, arguing that it is difficult to measure the contribution of each individual.
The whole article is sickening, because Joel tries so hard to contort reason and make it sound like he was being fair.
Well, not quite so--they did decide to offer him stock conditional upon him joining. While I don't agree with not coughing up a cash bonus, he is making a point about their culture, and that is always a dicey thing to fiddle with. And he isn't really in startup mode.
Stock in a company that isn't public and has no plans to IPO is literally worthless. Joel could have given him ONE HUNDRED BILLION shares if he'd wanted to. Well, almost worthless: it's worth exactly as much as one sheet of paper. Or half a sheet, since now you can only write on one side of it!
At least the t-shirt you could wear for yardwork or in the gym...
There're lots and lots of former startup founders currently working at Google. Andy Hertzfeld of Radius & General Magic. Joshua Schacter of del.icio.us. Brad Fitzpatrick of LiveJournal. Bunch of others whose employment and/or startups might not be known publicly. That doesn't even count the ones who end up at Google because it buys their companies.
It's not just famous people too: at least 2 or 3 folks I work with regularly are former startup founders, not including myself. Seems to be the new career path for bright aspiring programmers whose startups fail: graduate from college, start a company, fold up company, work for Google.
I would suppose that these ex-founders (especially the ones with some fame like Josh Schacter) belong to an uber-class minority within Google, where they are given more freedom and respect then the average Googler.
People who make 4.0s are the exact kind of people they need-- discipline, reverence for authority, people pleasers-- people who won't rock the boat.
Wow, really? High marks is an indication of a person's servileness? It can't be that they were fascinated by what they were learning, and as a result to perform well even though they didn't care that much about getting a high GPA, or have that much discipline or deference to authority? It seems there are two options here: do well and be branded as a servile, unmotivated toady who can't learn a damn thing on their own; do poorly, and be branded as an independent spirit who just might be brilliant depending on the results of further probing and research.
From this comment and its parent, it seems that colleges are being paralleled to boot camps, where the real goal isn't to develop knowledge and/or body strength and endurance, respectively, but to develop a psychology of doing what you are told instinctively. That you learned something along the way is completely immaterial, because that has nothing to do with the average you came out with. If you are capable of any independent thought, you would've known the material before taking the class, and scored low out of sheer boredom and apathy. Doing well shows you both didn't know the material before, and you were more interested in impressing upon someone else your knowledge of the material than actually learning anything at all.
If this is the case, I did it completely wrong. I suspect, however, that this is not the case, and that people are taking the metaphor of GPA to Obedience Average way too seriously. It can mean that, but it very much depends on the institution and the student, and probably a number of other variables.
In the spirit of full disclosure, my GPA in college was rather high. I will not say how high, because then the replies to this comment will be about that instead of the larger point. I worry just mentioning this much will dilute what I am saying, but not doing so seems disingenuous.
Conversely, I have been prodded by managers, coworkers, and friends repeatedly to be more disciplined, more respectful of authority, and more interested in pleasing people. I suspect I was this way in college as well, given the similar comments friends made there. Being too deferent has never been a problem for me.
Maybe I'm one of the outliers people discount to show a nice, clean correlation between two groups; it is merely misfortune that I am then rolled up into the correlation to suggest that I am as all the evidence I've gathered so far suggests I am not.
In all honesty, I'm not totally sure how I ended up with a high GPA. Maybe the institution I went to was too easy. Maybe my curiosity for the material I was learning was motivating in the same degree as the need to please authority was in the other high performers. Maybe that I had a curiosity for the material is a mark against me; it shows I was either curious about the wrong things, or not curious enough to learn it on my own without "structure" there to force it upon me. Maybe one or both are true. Maybe I, and those I know, are completely wrong about me.
I'm not saying either of you are wrong about what Joel is looking for by taking GPA into account. I'm not saying your right either. This is just about what high marks "mean" on their own. They can have more than one cause: it could be a sign of obedience, or it could be a biproduct of some other force that had nothing to do with the marks. The other metrics being taken into account will probably clarify which case it is.
Oh, and one other thing:
Coding a one loop bubble sort in hex blindfolded while juggling eggs, playing 12 games of chess, solving a 100 x 100 sudoku, providing an alternate solution to the 4 color theorem, teleporting tonights winning Powerball numbers, and cooking dinner would be a more applicable barometer.
Please, can we ease up on the machismo when discussing what makes for a great developer. The best metaphor I can come up with is the overly muscular guy at the gym claiming to bench some obscene amount of weight, and staring down everyone else to intimidate them from benching anything at all. Yes, it may be part of the culture to be over the top like this, I've always considered it one of the more unbecoming traits; it just seems excessively primal.
...high marks is an indication of a person's servileness?
It's no less preposterous than taking high marks as an indicator of intelligence.
Please, can we ease up on the machismo when discussing what makes a great developer [?] The best metaphor I can come up with is the overly muscular guy at the gym...
Gym Guy might not be all that bright, and has some glaring interpersonal deficiencies, but he does seem to be the polar opposite of servile.
"I know lots of people who had a 4.0 who, if you dropped them off in the woods, would have just died." -- It's software development, not the army.
"OTOH, I know lots of others who were brilliant but too bored to care about grades." -- Hence the whole second paragraph:
"Of course, that's not the only way to demonstrate your intelligence. I know many incredibly smart people who just didn't do well in school. Maybe it just wasn't their thing, or they were working two jobs and didn't have time, or they dropped out to start their own company. If that's you, then you'll have to show me how bright you are. Do you solve really hard problems on a regular basis at work? Tell me about it. Did you apply to MENSA for the fun of it and get in? Impressive. Hold the record for creating the Sunday crossword puzzle with the fewest black squares? Awesome. Exceptional standardized test scores will even do in a pinch (though that might not be as widely accepted)."
We're just screening applications, not making hiring decisions on the spot. If someone has great grades (and four other positive indicators) but are truly not that smart, we'll find out pretty quickly in the interviews. For people who are really bright but didn't have the grades, the burden is on them to show us how bright they are. If they don't, but pass on the other criteria, then they'll move on to the next stage. It's anything but a litmus test.
Did you apply to MENSA for the fun of it and get in? Impressive.
Egads. I was curious about MENSA and went to some meetings. A more self-congratulatory group of people I'd never met. I was offered membership but turned it down!
I think GPA is a better measure of "I can work inside a system and play by its rules"
Like you I think it's a poor measure of "brains" but you can't get a high GPA and be a total slacker. I know plenty of people who were too bored to care about grades...and also too bored to care about the work assigned to them by a manager at their work.
I think GPA is a better measure of "I can work inside a system and play by its rules"
Does having a high GPA indicate that this is what the person was trying to prove? Does it even suggest that following rules was the motivator that got them where they are? As I pointed out in my earlier comment, maybe the graduate's GPA was high due to motivations that had very little to do directly with wanting to achieve a high GPA or wanting to please other people.
The meaning is ambiguous; as are the meanings of everything someone might rely on to assess someone's ability. You need context to figure out what each individual indicator means, and this includes the GPA; it might mean absolutely nothing, or it might be the thing that pulls two or three other indicators out of ambiguity and makes the true ability of the person become more believable.
This is very true. Perhaps, though, startups in their early stages should hire people with low GPAs, but who have some other proof of intelligence? If I want to really innovate, I'm going to hire people who not only think outside the box, but abhor going anywhere near it. 'course, these are the people that don't stay on as the company transitions from R&D to sales and maintenance.
In my (limited) experience, running an early stage startup is less about innovation and more about drive to get things done. Then again, I work with people who are good at what they do and fully invested in the project. Nobody is getting a salary (and probably won't for at least a year), so that tends to weed out the mediocre players.
We're early phase, and I realized recently that if I could hire any developer that I've worked with in the past that all of my top 3 (one of whom we recently hired) were college drop outs, and in general, I'd actually bias my decisions slightly towards those who had completed degrees. This surprised me a little.
Damn, I don't have a GPA! I could send away for a copy of my transcript: ~35 pages of course descriptions and narrative evaluations of me by my faculty, along with self-evaluations (missing would be my evaluations of my faculty, which go in their transcript)
the 'relevance of grades' argument never seems to die here. like any other measure of ability, the use of GPA has its limitations. the reason the debate continues, i suspect, is that a lot of people here got bad grades.
Anecdote: I went to a $40,000 / yr college and got a GPA-based "Presidential Scholarship" every semester. A friend of mine, who didn't really know what to do with college, got a 2-year degree in general ed from a state college. I can tell you that he's at least as smart and about 3x wiser than I am. He gets more out of his job, has a better social life, is in better shape, etc etc etc.
So from my perspective, GPA is about 95% useless in real life.
:)
I did say "in my experience"! I also called the story out as an anecdote, not anything statistically useful. The 95% figure is an estimate based on lots of people I know, not just the one story.
I think we agree completely that the use of GPA as a measure of ability is limited. To be fair, though, I never asserted that GPA would predict who would 'have a better social life' or who would be 'in better shape.' In fact, I might predict that these would be negatively correlated with GPA. Also, I'm not sure I understand your story: did your friend actually have low grades, or are you just saying he was highly successful after attending a less famous school?
In any case, I hesitate to offer my own examples: I don't have a sufficiently large sample size, as I don't hang out with people who had low GPAs.
Not unless you had time for both. I definitely didn't have to spend as much time as my peers on Comp Sci homework, and had plenty of time left over for my own pursuits.
The great programmers I respect easily got CS degrees without working too hard; you show up and submit your tests/projects using your self-acquired knowledge. You should also have a natural curiosity for the subject matter. I think this is why Google's strict GPA policies work for them.
If you're an undergrad then your full-time job is to get your degree. If not, then you probably shouldn't have gone to college in the first place... Why take on all that debt if you aren't going to take it seriously?
Or you could be me and work close to full time the entire time you are in university so you aren't in ridiculous debt when you leave.
(that being said, I would have done much much better in my undergrad if I hadn't of done that)
I chose courses that interested and challenged me. I had semesters that were nothing but computer science, math and physics. Between implementing a compiler, figuring out how to display a polygonal topographical map and learning the life cycle of stars, I didn't have time for side projects.
I submit that if your side projects interest you more than your studies, then you are in the wrong place.
The question presents itself: why would someone that good want to work on project management software with such a funky user interface and oddball feature set?
That's the same sentiment I had. The software guys I've worked hated FogBugz so much they scared the crap out of our CEO. I watched Aardvark'd and once longed to work for The Great Fog Creek, but I've already moved onto bigger and better things.
In 15 years in the industry I have met many people who like to quote Joel Spolsky, but I have not come across one single person who actually uses a Fog Creek product. Kinda like lots of people have read ESR, but no-one is using his code...
1. They have a stable base of corporate microsoft shops using fogbugz -- which is why none of us knows anybody using it. My guess is their other products make little if any revenue.
2. They don't pay most of their people very much. They probably only have a dozen or so people on staff full time and everyone else is an intern.
3. Their job board does pretty well. Quite a bit of passive $$$.
Fog Creek seems like a stable and profitable business. That said, there are thousands of profitable small boutique software companies out there.
Why not? I've come to believe that the actual field of programming matters much less to job satisfaction than the way it is done. Some "sexy" areas of programming (data mining, high scale apps) can be an enormous grind. Some CRUD business apps can be very rewarding. What matters is having a very good effort->reward loop. You can get that loop from building project management software. The specifics of the organization and the development process matter more than what you are actually building.
Indeed. The interesting programming problems are almost always independent of the business problem domain. I work for a company that writes HR applications. Very boring. The programming part, though, is very interesting.
Things like Moose, KiokuDB, and so on all came out of this work.
"I've come to believe that the actual field of programming matters much less to job satisfaction than the way it is done. .... The specifics of the organization and the development process matter more than what you are actually building."
This contradicts my personal experience.
I worked for a great organization (ThoughtWorks) which had a great development process (Extreme Programming with "lean" elements, with a lot of tailoring and flexibility, the team could do almost anything really) and I worked with great, awesomely talented people, but I got bored to death with the enterprise apps we were building (the domains were all incredibly boring to me, and I could switch off 90% of my brain and still deliver successful projects) and so I left.
Now I don't have an "organization" (I often work alone or with small groups of people assembled for a project and that too virtually), but my domains (AI/robotics/ Computer Vision/Planning software) are so much richer than what I used to work on, (and so is my toolkit - C, erlang,lisp etc vs java/ruby-on-rails) that I wouldn't exchange the "grind" bits of these projects for the "cool" bits in my old work.
I suggest that just as working in lisp maybe in general more pleasant than working in COBOL, (there will always be exceptional situations where COBOL is more pleasant, or the odd person who prefers working in COBOL), programmers would in general prefer some fields of programming to others.
You say,"What matters is having a very good effort->reward loop. You can get that loop from building project management software." .
I agree totally, but I'd also say that unless you are running your own startup making such software, this would be a rare experience.
In my experience the "reward" in enterprise software comes from understaning customers in depth and satisfying them. In a non startup enterprise coding effort, this is hard (but not impossible - see jrockway's example in his answer - I am just claiming this is more the exception not the rule) to do?
What Steve Blank calls "Customer Validation" remaining equal, maybe there are levels of "blub" in domains and/or projects too?
EDIT: This quote from PG expresses this idea much more clearly than I could
"
It's pretty easy to say what kinds of problems are not interesting: those where instead of solving a few big, clear, problems, you have to solve a lot of nasty little ones. One of the worst kinds of projects is writing an interface to a piece of software that's full of bugs. Another is when you have to customize something for an individual client's complex and ill-defined needs. To hackers these kinds of projects are the death of a thousand cuts.
The distinguishing feature of nasty little problems is that you don't learn anything from them. Writing a compiler is interesting because it teaches you what a compiler is. But writing an interface to a buggy piece of software doesn't teach you anything, because the bugs are random. [3] So it's not just fastidiousness that makes good hackers avoid nasty little problems. It's more a question of self-preservation. Working on nasty little problems makes you stupid. Good hackers avoid it for the same reason models avoid cheeseburgers."
from PG's "Great Hackers" essay.
PS: I don't claim to be a great hacker but I am probably a good one.
So that they don't get abused and mistreated like they do at most startups and large software companies?
Depends on what you want from your life really, but if you're given a challenging environment where you get to perfect your craft and learn from other equally intelligent people - what more could you ask for?
I've been on both sides of the hiring table in different professions (staffer, co-founder) and that article made my stomach turn. Anyone who puts up that much of a fuss lacks the courage to lead and manage people the right way - through inspiration.
To the youngsters out there: the more difficult they make it seem, the less they actually have to offer.
I'm curious: how would people compare the prestige of working at Fog Creek vs. the prestige of working at Google? Or one of the other big Internet companies - Microsoft, Amazon, Yahoo, EBay, etc...?
I remember Joel doing similar how-to-get-hired essays when I was in college. I was always like, "Pffft. Why would I want to work for a company that does bugtracking software?" Ended up taking a job at a financial software startup instead. I'm wondering if other people feel the same way or if I'm just a gigantic snob. (The two may not be mutually exclusive...)
I think that working at a company like Fog Creek or 37 Signals would be seen as highly prestigous, but only by a pretty limited audience: those that follow software development blogs.
Whereas working at Google, pretty much everyone who has touched a computer is familiar with who Google is.
So I guess it depends on what your intended audience is.
Twain once said something like "The most sought-after prize is the envy of other men". It's natural to want "prestige", but chasing after it kind of misses the point of life, no?
If you haven't done so, I recommend working for a small technology company with a strong engineering culture, staffed by smart & competent people who have nothing to prove. Worthwhile problems are fractal -- there's almost always more there if you keep digging.
It depends on what you do. Google definitely has more high profile jobs than Fog Creek, but also probably more grunt jobs than a typical small company.
Real bad litmus test.
I know lots of people who had a 4.0 who, if you dropped them off in the woods, would have just died.
OTOH, I know lots of others who were brilliant but too bored to care about grades.
Coding a one loop bubble sort in hex blindfolded while juggling eggs, playing 12 games of chess, solving a 100 x 100 sudoku, providing an alternate solution to the 4 color theorem, teleporting tonights winning Powerball numbers, and cooking dinner would be a more applicable barometer.