The generalization is weird, but the odder part is that you know your co-worker's SAT scores. How does that even come up in conversation? It makes me miss the old silicon valley where every other person was a weirdo who never even went to college.
I don't mean to cause offense, but I find people who never went to college or did not major or focus heavily on CS or engineering generally more interesting to both work with and as people. Somehow college grads (especially in math heavy fields) seem to believe that their scores or grades are a reflection of their intelligence, drive or even personality. Not to say there are huge exceptions; both among people I know, and among famous thinkers- I find Poincare, or Mandelbrot to be great 'broad' thinkers who totally understand the fallacy of such numbers in both life and science.
I have a BA from a pretty highly ranked university in Canada, and excelled in academics. However, I always thought it was a breeze for anybody to do really well in. I always thought the Arts were stupidly easy compared to the Sciences. So, I always down-played my performance, because I always knew that, in the end, I had accomplished something very insignificant, despite being in the top 5%; and even despite various professors suggesting I could do really great things and one going as far as telling me I shouldn't be in university "since I already knew all the material".
Long story short, I graduated and, basically, failed in every other aspect of my life, including landing a good career and was getting paid minimum wage for the 2 years after graduating. I left the country, am doing OK now, but learning to program and interested in more of the sciences (lo and behold, right?).
But it is funny how now that I am surrounding myself more with science people and CS/IT people, I feel I have strong proof of how excelling in one subject or area can get to someone's head. I'm glad I never let it get to me. And I think back at the times I wasn't humble when trying to explain some esoteric concept to someone, or when someone didn't explain something properly or simply did not have the knowledge to fully make an adequate generalization of a given concept, and I was very indifferent and critical; I think back and feel bad and hope never to be like that. I'm much more of a humanitarian now, though, and feel a lot more compassion for humanity; for those who act in both bad and good faith.
Personally I find that most of the best developers I've hired were people without CS degrees - a CS degree does not make someone a good programmer, and the intersection of people who are interested enough in the science part of it and the people who are prepared to work hard at being good programmers is smaller than some people seem to think.
There are certainly exceptions, and someone who does great at CS and who at the same time manages to get their life balance right AND works hard at being a good programmer can be tremendously good, but they are exceptionally rare, and the longer the CS degree the rarer they are...
People with CS degrees have a "bump" in the jobs they can get. The ones who are really good get hired at better jobs than any that you can offer. Therefore your sample is biased towards people with CS degrees who have failed to get a better job, which means that they have some deficiencies. The better the degree, the better the bump, and therefore the worse the deficiency needs to be for them to remain in the pool that you encounter.
Conversely good people without a CS degree have a hard time getting into those really good jobs, and are therefore available for your sample.
Thus the negative correlation that you've seen between an obviously positive attribute (CS degree) and your hires is a result of your only seeing people in a certain band of desirability as a programming hire, and is not an indication that people who study CS tend not to be good programmers.
> People with CS degrees have a "bump" in the jobs they can get.
I don't believe that premise for a second.
My pre- and post- degree experience is that nobody cares about the degree if you have more than a couple of years worth of experience.
In terms of my ability to get high paying jobs, getting my MSc made no difference. I got great jobs before I completed my degree, and the same after. In fact, the only way my degree comes up in interviews is as a semi-suspicious "why did you bother, with your track record, it's not as if it matters?" For me it was a personal choice, not a career choice, and the research project was interesting.
The reason I didn't complete my degree in the first place, was that it was so trivially easy to find high paying jobs based on my skills and experience (I started programming at five, and did my first paid development at 13) and nobody ever asked about the lack of a degree. I ended up leaving to do consulting and start a couple of companies.
At the same time I also know that none of the places I've worked - including a multi-billion dollar company that certainly had the budget to pay extra well for the degrees if they'd cared to - has placed any preference on people with a CS degree. The few great programmers we've found with good CS degrees have been offered no more because of their degrees than equivalent developers with other degrees or no degree at all. By your argument these people should've gone elsewhere because we'd be offering them well below their market value by offering the same as developers with no degrees, or with say, a Linguistics degree.
That idea just does not mesh with any of my experience.
> The ones who are really good get hired at better jobs than any that you can offer.
The ones that are really good without degrees also get hired really quickly - it doesn't stop me from seeing CV's from, and interviewing, a lot of people like that, many of which we immediately pass on because we know they will get offers much higher than what we're willing to pay.
Sure, I don't see many CV's for people who'd be able to claim $1m+ base, either, but people in that range are so few overall that even if every single one of them somehow have CS degrees it would not make much difference.
It also ignores the issue of the contents of the CS degrees. See below.
> Conversely good people without a CS degree have a hard time getting into those really good jobs, and are therefore available for your sample.
Junior developers or mediocre developers without degrees might face more of a challenge without a degree, but for experienced developers, my experience on both sides of the table is that years of experience matters far more when considering whether or not to give an offer, or when discussing remuneration - both when negotiating with technical managers and when negotiating with non-technical HR people.
> obviously positive attribute (CS degree)
Here you let your bias shine through. Why is it an obviously positive attribute in this setting?
Software engineering is not computer science.
A big part of the problem I see with CS graduates is that most of them don't have the faintest idea about engineering discipline. Many of them have not had any exposure to testing practices; many have not had any exposure to software design methodologies; many have hardly written any programs, and if they do they often only know one language and don't understand how to generalize the concepts (sure, they know the theoretical underpinnings, but often don't know how to translated that into writing code). Far more don't have much domain knowledge in any fields outside pure CS - it doesn't help if they can write code if they need a business analyst to babysit them all the time.
The issue is not that a CS degree makes people bad programmers, but it does very little to make them good programmers and even less to make them good, well rounded software engineers. Many places you can pass through a CS degree without having had more than a passing exposure to programming.
At my university, you could've easily get as much programming experience while completing a biology degree (or pretty much any other science related degree; for humanities/liberal arts you'd have to work a bit harder to get there) as in the CS degree, and that's not unusual.
Personally, I passed through my entire degree with the equivalent of less than 2-3 months worth of actual programming and I picked a few programming courses on purpose because they were interesting to me (e.g. a class on Smalltalk, and a course on compiler construction), but it would've been trivially easy for me to opt for a more math or theory heavy degree and avoid pretty much all practical programming exposure.
And my experience - both from hiring, and from talking to people in the business - is that a lot of people do, and very few go the other way and try to pick subjects that makes them better suited as software developers. Most people that go that way end up picking other degrees - such as going to schools that actually offers software engineering as a subject.
But I'm not hiring mathematicians or logicians or people who are meant to run research projects where undergrads do the programming - I'm hiring software developers.
While it is perfectly possible to combine the two, the important part to consider is that a CS degree teaches you computer science, not software engineering, and while there is varying degrees of overlap at different schools, if you spend your time doing a CS degree you spend a lot of your time 1) not programming or learning about software development skills, 2) not developing communications skills or more business oriented skills that are generally far more valuable most places that needs software developers than the additional skills CS does give.
Over the years I've had far more use for my (natural) language skills and communications skills than I have had for my CS degree. And I've had far more use for programming experience I've built up than either. I don't regret my CS degree, because I get a lot of pleasure out of using those skills on a hobby basis, but if I'd gone back and were to pick a degree specifically to be the best possible software engineer, there's just no way I'd have picked CS degree.
I've had situations where we genuinely had use for someone with a CS PhD, but for most development the CS skills are not that interesting and what you need are software engineering skill sets that most CS degrees simply don't teach. In the cases we've needed someone with in depth CS knowledge, we've in some cases found it easier to pair up someone with the CS knowledge with a good developer rather than trying to find the perfect combination.
These aren't Googlers I'm talking about, except the boyfriend. Most of them are friends from college, along with a few online friends and a family member mixed in.
As for how it comes up in conversation - look at this thread, it's got 3 people's SAT scores already. It's not like it's particularly taboo. Usually it starts with somebody mentioning how much they hate/miss high school over dinner or a few drinks, and then soon everybody's offhandedly mentioning their numbers.
(Edit: Also, I hope that everyone reading this understands that the numbers, while inspired by real people, are completely arbitrary. I don't really care if the general HN community dislikes, misunderstands, or downvotes the post. I care that the people who would see themselves in the 1590 category, whether or not they actually scored a 1590 on some standardized test several years ago, understand the point.)
No, it's really quite strange. This thread has got me thinking, until this thread, I can't even recall the last time somebody mentioned their SAT score. Probably more than 15 years ago? I've heard more IQ scores, GRE scores and even TOEFL scores than SAT scores.
Though remember that most of these conversations happened when I was actually in college (a few even before then, while applying to schools), and I just have a really long memory.
(The conversation with the friend that sparked this musing was last night, but I've been friends with her since she was 14, and so our conversation topics include anything that was fair game back then, which is basically everything.)
The generalization is weird, but the odder part is that you know your co-worker's SAT scores.
That's what I was also thinking. I don't think I've talked about SAT scores since I was a sophomore/junior in HS and was taking the SAT. My friends and I have occasionally talked about GRE scores, but that's only because one of my friends tutors/preps people to take it.
Your undergrad degree is just a noisy and expensive proxy for your SAT score anyway. And that's not taboo to discuss. Paradoxically, it's less declasse to mention the subjective measure that takes four years and costs a quarter million bucks than the objective measure which takes two hours and a number two pencil.
