I Won’t Hire CS Majors: I just want rich boys

September 11, 2015 at 8:49 am 36 comments

The Wall Street Journal recently ran an article by an entrepreneur about why he doesn’t want to hire computer science majors at his start-up.  I was particularly struck by this line:

The thing I look for in a developer is a longtime love of coding—people who taught themselves to code in high school and still can’t get enough of it.

Source: Why I’m Not Looking to Hire Computer-Science Majors – WSJ

Allow me to translate: “I want rich boys. I want those boys who were in the 10% of schools that have CS teachers (which are all rich districts), or parents who knew enough to provide their boys instruction and access from their teen years on. (Nobody really ‘teaches themselves to code.’) The females who start coding in high school will be filtered out by the time I’d hire them. I want those boy to come to me with a decade of immersion in the existing male-dominant, defensive, homogeneous culture that pervades CS classes, so that my startup will be just as lacking in diversity and just as unwelcoming to women. Let’s hear it for the status quo!”

Entry filed under: Uncategorized. Tags: , , .

The narrative of grit: We need to change the system, too Students concerned about demand for CS classes at Berkeley: First of many

36 Comments Add your own

  • 1. Derik  |  September 11, 2015 at 9:56 am

    I could not disagree less with your assertion. I personally have avoided jobs where everyone there was a CS major. I find that in most ‘business development teams’ CS majors are the worst developers. They have huge egos… hey I have CS degree and there code is overly complex.

    I also disagree with the idea that the only want to learn to code on your own. I did not go to a rich school yet I learned to code in QBasic my senior year. I did get a BIS degree, but the focus was on business not algorithms.

    To me, in many, not all cases. CS degree == lack of practical skills, too much theory and massive egos.

    • 2. Mark Guzdial  |  September 11, 2015 at 11:04 am

      There’s this wonderful quote from Michael Kimmel, “White men have the greatest privilege in the history of the world. It’s called ‘the history of the world.'”

      It’s really hard for you to argue that you had no privilege.

      • 3. Hadi Partovi  |  September 11, 2015 at 5:43 pm

        If Derik “could not disagree less”, that means he could not agree more. Sounds like you agree with Mark? 🙂

  • 4. Raul Miller  |  September 11, 2015 at 10:34 am

    I am not sure I agree that your characterization is accurate.

    http://www.census.gov/content/dam/Census/library/publications/2014/acs/acs-28.pdf says 83.8% of U.S. households report computer ownership. This document does include a mention of a bias towards the rich, and there are problems for those in poverty – but those problems are not going to be just about computers but also about nutrition and presumably other problems also.

    Anyways, I kind of doubt that the usual concepts of “rich” (e.g. “the 1%”) is an adequate way of characterizing the biases and issues here.

    So, I guess I am wondering where you got your information.

    My concern is that spending time and energy trying to solve the wrong problem tends to make the real problems fester or even worsen. And the numbers on this one just seem blatantly wrong (and my initial impression of what you are suggesting also happens to conflict with my personal experiences – which include a variety of encounters with computer professional “rich boys” who just didn’t seem to have a clue about solving computational problems but who instead were pretty good about convincing people to give them lots of money for their time).

    That said, there are biases towards the rich, simply because they can casually afford to ignore some obstacles. But you might be missing the forest (for example: inadequate curriculum, but also non-educational issues such as personality types) for the trees (the systematic advantages of the lucky few).

    • 5. Mark Guzdial  |  September 11, 2015 at 10:59 am

      Raul, are you seriously arguing that access to computers is all that’s needed to develop interest and skill in programming? How then do you describe the lack of rich women in CS? There’s far more to it than that.

      The point of my post is “privileged.” I recommend Miranda Parker’s paper from RESPECT 2015 on what privilege is.

      • 6. Bonnie  |  September 11, 2015 at 11:12 am

        I agree with Mark on this one. I teach at a school where many of our majors are first in their family to go to college, heavilly Hispanic and African American. When they hit my CS1, none of them have had any exposure to computing. Not only is it not in their schools, but it isn’t part of their teen culture the way it is for middle class white boys. And yes, we have a lot of issues with them being shut out of the job market, for a whole raft of reasons. Even our top students have trouble, because they don’t look or act like the sterotypical geek programmer.

        • 7. gasstationwithoutpumps  |  September 11, 2015 at 1:43 pm

          I teach at a Hispanic-serving institution, where 40% of the incoming class are from families in which neither parent has a 4-year degree. We don’t have any problems with our CS students getting jobs, though we do have the US-wide problem of serious gender imbalances in CS. (The new Computational Media department has an even bigger gender imbalance problem, as they have the game design majors.)

          I think that one key is to look for programmers who are obsessive about updating their skills—ones who know how to teach themselves new things and continue to do so. Neither conventional CS education nor choosing the high-school know-it-all are particularly good predictors of who will continue to improve their skills.

          “A longtime love of coding” may be somewhat predictive, but only if the person’s love of coding includes a love of learning new approaches, new tools, and new languages and is backed by a solid understanding of fundamentals and good problem-solving skills.

          • 8. Bonnie  |  September 11, 2015 at 2:27 pm

            Unfortunately, we are in NYC, so the employers are largely those arrogant Wall Street types! A lot of them only look at students who come from Ivies.

            • 9. zamanskym  |  September 13, 2015 at 6:38 am

              Bonnie – not disagreeing with your experiences but I too am in NYC and it’s been a long time since the majority of my kids ended up in a Wall St. firm and it seems that the minority of my kids even interview with them.

            • 10. dennisfrailey  |  September 13, 2015 at 9:13 am

              A number of years ago I was approached by a major wall street firm. They were having problems due to their traditional way of hiring software peope and found that the best ones they had were educated while in the military, taking a distance education software engineering program at SMU in Dallas, in which I was one of the faculty and creators. In the end they hired me and my colleagues to put together a program to educate their 200+ software developers about software engineering. It was a very high quality program (still is) that teaches a lot of fundamentals. It’s modeled after the ABET criteria for software engineering and the CMMI model, which addresses a lot of topics not specified by the ABET model.

  • 11. Joe Mainwaring (@theaccordance)  |  September 11, 2015 at 10:38 am

    Perhaps in your region of the world that assertion holds true, but it certainly does not represent the entirety of the global landscape. Nobody teaches themselves to code? Untrue. I’m proof of that. I was also not a “rich kid” who lived in a district that provided any type of formal curriculum in CS. My parents? A housewife and a HR director; no background in programming whatsoever. My longtime love for coding may have some bias, but that’s towards my choices in a tech stack, not the gender of whom I collaborate with.

    Sexism does exist in tech; but the philosophy of hiring developers with a longtime love for coding is not the driving cause of that. If that were the case, then I would have dozens more male coworkers than I do today.

    • 12. Mark Guzdial  |  September 11, 2015 at 11:02 am

      My focus in computing education research is on empirical evidence. Self-reports are biased. The evidence suggests that very few people are self-taught in computing.

  • 13. dennisfrailey  |  September 11, 2015 at 10:52 am

    I wrote a letter to the editor of the WSJ (which they did not choose to print, but another letter in response was similar to mine). The gist of my comment was that although there is some truth to the writer’s contention that many of the “top notch” CS programs are more focused on producing CS researchers than computing practitioners, the author’s approach of hiring people from coding schools because they know the latest techniques and languages is short sighted. I pointed out that the languages and techniques that are popular today will be obsolete before long and that both students and employers are being short sighted in ignoring preparation in fundamentals that will still be relevant for the foreseeable future.

    In my own case, the most useful courses I took for my overall career were the fundamentals of math and computer science and the writing and reading skills I learned in the K-12 period. The programming languages and development techniques from back then (in the 1960’s and 1970’s) have been replaced about four times with newer ones.

    • 14. chardnett  |  September 11, 2015 at 3:11 pm

      I disagree. The programming techniques do change, a student that is taught current software development techniques on modern tools can adapt to new programming methodologies just like someone from a “research focused” school. There is lifelong learning involved regardless.

      • 15. dennisfrailey  |  September 12, 2015 at 5:43 pm

        I certainly agree that lifelong learning is part of the deal when you choose computing as a profession, but my point is that there’s a whole lot more to achieving success than being proficient in a currently popular language and tool set. I can think of several examples from my own career where the fundamental knowledge I obtained in university courses made a difference in my success. As a simple example, I once was given an algorithm that worked very successfully in a simulated environment but was much too slow for a real computer (it was an algorithm related to a real-time application). By combining knowledge of mathematics, numerical analysis, computer arithmetic, and signal processing, I was able to restate the algorithm in a manner that was roughly 1000 times faster and to make it feasible for use in the product at hand. In a totally different example, the CEO of a major company once asked me to help him write a speech about computers. Because of my ability to write well and give effective presentations, I ended up writing several speeches for the man and this led to all kinds of opportunities for advancement.

        One thing I’ve learned along the way is that most employers don’t know how to tell the competence of a prospective software developer. It’s all too common to judge them on whether they know a particular language or tool. I’ve had to clean up a lot of software written by individuals who were self-taught experts but whose knowledge was narrow and focused. Of course this can also happen with individuals who go to college but don’t get a suitably broad education. In one case an individual had a PhD in AI, but knew nothing about computer architecture except the architecture of the LISP machine and had no clue that the AI software he was so proud of could be replaced by a (substantially faster and simpler) decision table.

  • 16. Jens Mönig  |  September 11, 2015 at 10:56 am

    another – perhaps even less favorable – way to translate this would be: “I want serfs who depend on me because big companies won’t hire them for lack of a CS degree”.

  • 17. Bonnie  |  September 11, 2015 at 11:02 am

    Yep, and those high school hackers don’t know any actual computer science, so they will write reams of bad code, very quickly. And then we wonder why software is so bugridden. I had to work with those sorts of hotshots when I was in industry. It wasn’t fun.

  • 18. Jana Markowitz  |  September 11, 2015 at 11:46 am

    This is not a big surprise. Rich (entitled, arrogant, macho) boys are who Wall Street hires in general. Only about 6% of Wall Street finance employees are female. The whole culture there is dysfunctional. That’s why the Great Recession happened and we seem to be doing nothing to try to change their mindset or behaviors. This is not an IT/CS problem, it is a Wall Street/Finance Industry problem.

    Please keep producing CS majors – we need them in all industries and in all kinds of jobs, not just tech ones. Their problem-solving and organizational skills are needed everywhere. But please also give them people skills and writing/presenting skills. Those are a must.

    • 19. chardnett  |  September 11, 2015 at 3:13 pm

      Completely agree. Communication skills are paramount and they are the biggest gap that employers point out when they talk about recent grads across the spectrum.

      • 20. dennisfrailey  |  September 12, 2015 at 5:46 pm

        I concur. In my 36-year career in the hi tech industry, the people who lacked communication skills were the most likely to suffer from a mid-career plateau where they could not get promoted.

  • 21. gflint  |  September 11, 2015 at 1:40 pm

    I took my CS class on a field trip to a local software company. I asked the COO this exact question, who are the best hires. He said he hires only CS college grads. The few hackers he had hired lacked the depth of knowledge for the job. Their focus was too narrow and they could not deal with divers problems.

  • 22. Travis Collier  |  September 11, 2015 at 11:35 pm

    CS really should not have much to do with practical coding. It is essentially a branch of mathematics. Hiring CS majors for coding jobs is a bit like hiring economists as accountants.
    Sadly, not many institutions have separate CS and computer engineering. Introductory CS + courses in practical computer engineering is what most projects really need in the way of coders and designers… Basically a vocational education (which has many other benefits actually).

    As for the socioeconomic and gender stuff… Um… how would just hiring CS majors be better?

    • 23. gasstationwithoutpumps  |  September 12, 2015 at 12:30 pm

      Travis, you have a very narrow view of computer science (just the “theory”), which is unfortunately also true of some computer science departments. It is more common, however, for CS departments to embrace a range of subjects, including software engineering, network engineering, game design, and other hands-on fields.

      At UCSC, we have separate departments for computer science, computer engineering, computational media, and bioinformatics (in the biomolecular engineering department). Other places have merged everything from computer science to electrical engineering into one big department. The differences in structure change the office politics a bit, but really don’t make much difference to student education, which depends more on the attitudes and interests of the faculty than on how the bureaucracy is structured.

  • 24. nickfalkner  |  September 12, 2015 at 8:46 pm

    Couldn’t agree more, Mark. Thank you for posting this. As always, I admire both your generosity and restraint in addressing other people’s comments on points that you raise.

    • 25. David  |  September 13, 2015 at 4:36 am

      I understand what the author is saying it has nothing to do rich boys or a male dominated shop. The author is looking for people with drive and passion that can get things done. I am working worked as a programmer and architect for years and yes you an self study and teach yourself the skills needed. My background is electronics engineering with experience with computer hardware, radar and communication systems. Here is how I got started, I bought a computer and bought a book. Later I took courses from NRI while on border patrol.

      No one needs to attend a brick and mortar school, the concept is outdated except in a few areas. I learned software development using a trs 80 model 1 and z80 assembler later I got a degree in cs and really did not learn a thing that I did not already know.

      Later I was writing novell drivers in x86 assembler and reprogramming system bios on the early PCand PS2 computers for hardware enhancements. I worked with many CS grads and others with different backgrounds. My experience has been if they have a decent math background and have the drive and have been developing for awhile then they get a shot.

      How many CS grads know design patterns ,architecture patterns, and, javascript patterns? Not too many, to which I end up mentoring. I had one excellent programmer who’s background was journalism. I have found pure CS shops to be discriminatory against with of different backgrounds and age. While I do not fully agree with the author I understand some of the points.

      • 26. nickfalkner  |  September 15, 2015 at 2:15 am

        Some people can’t succeed for reasons that have nothing to do with their drive and everything to do with systematic discrimination and lack of opportunity. That’s what privilege is about.

        The fact that you can’t see it (or choose not to see it) doesn’t mean that many other people do not face problems because of their lack of privilege every day.

        You say you bought a computer and a book – which means you had the money to buy both of them and a safe place to store them, as well as power to run them. If you can’t see that there’s a level of privilege there, then I really don’t know what else to say. You had access to computers before you went to college. I’m guessing this means you had enough support early in life to do this. I’m glad for you. Many people do not have this option. And it’s not their choice, it’s the way that their lives are.

        There’s a lot of presumption in your requirement for people to “have been developing for awhile” because you are assuming that everyone has access to computers, the knowledge or even wi-fi to get to the Internet to find enough to drive them?

        You can have the time and capacity for drive and passion when you get enough to eat, when you’re not terrified of going home, when you live in a safe neighbourhood, when you’re not being threatened all the time, when your school is safe, when you have a school, when you have role models, when you have mentors…

        If you don’t understand what I’m saying, then I really don’t know what else to say to you.

  • 27. Serge Kruk  |  September 14, 2015 at 2:32 pm


    While I generally agree with your sarcastic translation, there is
    one point of serious disagreement I want to highlight. Your
    parenthetical statement “Nobody really ‘teaches themselves to
    code.’” is patently false. Anecdote: when I was fourteen or so,
    my best friend and I got a hold of a Fortran manual, read it in
    detail, and learned to code (badly). One of us would write a
    program that the other would ‘run’, on paper, to figure out what
    it did. We challenged each other by writing more and more
    abstract algorithms (i.e. not only computing the area of this
    particular triangle, but computing the area of any regular
    polyhedron.) And, before you ask, no, we did not have any access
    to a teacher that pushed us in this direction or even helped us
    with programming. We did not even have access to a computer. We
    were simply two kids in love with the idea of programming a computer.

    Now this was ages ago (the Fortran reference is a dead giveaway),
    but we were not outliers. Both my best friend and I ended up
    with careers in or around ‘computing’ where we met dozens of
    others just like us. Do I have empirical, verified, data about
    this? No, I will grant you that. But, before all of us old folks
    retire (many have done so already), I suggest that someone should
    survey the field and ask this simple question: ‘Did you learn to
    code on your own?’ and correlate the answer with age. I am
    absolutely certain that you will find that it used to be true to
    a very large extent.

    • 28. Mark Guzdial  |  September 14, 2015 at 3:58 pm

      As I mentioned earlier in this thread, any one’s personal experience is necessarily biased and likely an outlier. But even in your story, you didn’t learn alone. You had each other. You found a Fortran manual. You got access to a Fortran compiler. You had access, which is a privilege, which others did not.

      • 29. gasstationwithoutpumps  |  September 14, 2015 at 9:38 pm

        Mark, you are not reading closely. He said that they did not have a computer—that they simulated a computer for each other, so they did not have access to a compiler.

        If finding a discarded Fortran manual and having a friend constitute privilege, then the bar for privilege is extremely low.

        I think that most of us who have been programming for over 45 years are at least partially self-taught, but that the greater ease of getting instruction has gradually made this less common.

        I did learn some programming on my own, but I definitely had privilege—my father was an engineer, and my high school had its own computer (rare in 1970) and had a Fortran programming class. I was doing programming before my first course in it, and most of my programming I taught myself, until I switched to computer science in grad school and got some solid instruction.

        • 30. nickfalkner  |  September 15, 2015 at 2:18 am

          Yes, the bar for privilege is low but it still excludes billions of people. As I said above:

          “You can have the time and capacity for drive and passion when you get enough to eat, when you’re not terrified of going home, when you live in a safe neighbourhood, when you’re not being threatened all the time, when your school is safe, when you have a school, when you have role models, when you have mentors…”

          There is a real problem in some groups and communities about people having to pretend that they don’t like programming because it’s seen as ‘nerdy’ and people have cultural pressure to either lie about their interest or discontinue their studies. Having a friend to support you? Yeah, you’ve got more than many people.

          Once again, I stand with Mark on this and wish more people could see how much privilege we take for granted.

        • 31. Mark Guzdial  |  September 15, 2015 at 1:00 pm

          You’re right — my apologies. I did not read closely. I missed the part about not having a computer.

          Like Nick said, the bar for privilege is extremely low. It’s not a binary value — it’s a set of real values. Miranda’s definition that she drew from STEM education research was:

          Privilege is an unearned, unasked-for advantage gained because of the way society views an aspect of a student’s identity, such as race, ethnicity, gender, socioeconomic status, and language.

          I know that my being a white male gives me advantages that other people don’t have, for example, in terms of respect, trust, and responsibility. That’s privilege. I didn’t earn or ask for being male or white.

          The point of privilege research is to identify what factors are leading to success that we might not realize, factors that might impede the success of the same intervention in other conditions. For example, I could imagine being in a setting where a student the same age as Serge was didn’t know that the discarded Fortran manual was a guide to a programming language, where students were not in a social setting where two friends might realize that they were both interested in figuring out this programming language even if they didn’t have a computer. I’m thinking about Kohl’s “I won’t learn from you” or Betsy DiSalvo’s Glitch students who feared that studying computing would not be accepted by their social group. The point of identifying privilege is not to say that Serge didn’t have drive, but to say that the same “intervention” might not work in other settings with other students — and it’s worth figuring out why.

  • 32. rademi  |  September 15, 2015 at 9:42 am

    I think you two are talking past each other.

    Yes, it’s quite true that you can have drive and passion and “not succeed”. And I guess it’s also true that a person with enough privilege might be able to “succeed” without drive and passion?

    But succeeding without the drive to do so seems like it must be a very rare and unusual thing. If I understand properly, people put into facsimiles of that sort of situation tend to self-destruct.

    That said, it’s probably also important to have a meaningful concept of “success”. For example, if your concept of “success” is earning more money than anyone else, you’ll have to make a lot of sacrifices to get there, and you’ll have to be able to trust and rely on a lot of people. It takes a peculiar combination of personal worthlessness to make those sacrifices, some rather intense drive of another peculiar sort to amass that kind of funding, and a fair bit of luck and [yes] privilege to pull it all off.

    Most people, though, aren’t the sort to achieve that sort of success. This would be true even if every person were exactly identical.

    Anyways, you aren’t going to solve problems of unfairness by ignoring the need for long-term personal motivation.

    Now… there are some very real needs [plural] for education. And not all of them are obvious needs. But educational institutions aren’t going to go away… So you shouldn’t be scared if someone indicates that they look for motivation rather than education.

    What you should instead be scared of are deficiencies in the educational system (e.g. a lack of respect for factual information, inadequate curriculum, destructive administration, numerous practical issues not being addressed in class).

    But, yes, adequate nutrition, a situation which allows adequate sleep and a nurturing home environment are incredibly important also. But none of these are the issues which specifically filter out females (these criteria might filter out many females who grew up in juvenile detention facilities, but more likely females are being filtered out by parents who recognize the faddish nature of the computing economic bubble and expected it to have collapsed already).

    There’s a lot of work which goes into learning, but that just winds up being wasted effort when it doesn’t lead the student to connect it back to things that matter for them. And that, I think, is the real educational issue raised by that WSJ article.

    • 33. rademi  |  September 15, 2015 at 10:02 am

      (This reply was intended to attach to the reply which opens with “Some people can’t succeed for reasons that have nothing to do with their drive and everything to do with systematic discrimination and lack of opportunity. That’s what privilege is about.” But for whatever reason that did not happen, and I do not have the option of correcting that issue…)

  • 34. Erik Engbrecht  |  September 20, 2015 at 2:58 pm

    I think you’re missing the point of the op-ed. What’s being argued is that CS grads, particularly from top institutions, are not worth the price premium required to hire them in SV. It’s not that they are necessarily bad, it’s that they cost a ton and don’t necessarily have the skills he needs to be productive right out of the gate.

    It’s true, he appears to be filtering for people who had the advantage of someone encouraging and supporting their interest in programming prior to adulthood. You identify this as a privilege.

    But do people in this group tend to come from more privileged backgrounds than graduates of top CS programs? If they are, does this mean a business should pay a huge premium to hire someone out of the “less privileged” pool?

    How is a pool of people demanding huge salary premiums relative to other people doing them same work “less privileged?”

  • […] “play” in order to learn tend to be highly privileged (see the “rich boys” blog post here).  Nobody wants kids to just “type in pedantic command words in a programming […]

  • […] The below quote is particularly aggravating because it says that programming is only right for a certain “type of person.” For the technology industry, that usually equates to privileged white or Asian males. […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 10,184 other subscribers


Recent Posts

Blog Stats

  • 2,049,220 hits
September 2015

CS Teaching Tips

%d bloggers like this: