Please vote this article up on social news sites! Why?

I don't have adverts on this site, and I make no income from it. I write these articles because I believe they will be genuinely interesting and useful to people. The more people get to see them, the more people I think I've helped or touched. Please help spread the word! :-)

1. 

   Ross about 23 hours later:

   Its a good list!

   Focusing on one or two technology stacks can cover a hell of a lot of technology and shouldn’t really be cause of concern. In fact constantly following trends and the bleeding edge could also be an indicator of a limited attention span / inability to master technologies. I suppose thats one to be looked at in a interview.

   So where’d you find these programmers?

2. 

   crysfel 16 days later:

   You´re totally right!

3. 

   vahid 16 days later:

   Hi Daniel . Great essay! I absolutely agree with you. Specially, I believe that most of great programmers -i.e. great hackers- have a bunch of programming languages ans technologies in their pockets to deploy them in appropriate conditions. This viewpoint avoid them to choose dump ways. And this fact leads them confidence and audacity to get their jobs done . That’s why great programmers are eager to learn new technologies and have a lot of fun with them :)

4. 

   Ashley Moran about 1 month later:

   Nice summary! I agree with the comments about developers who started programming at university, and see it as a day job (to me, they are two sides of the same coin). Until I worked in IT professionally, I didn’t even realised these people existed! I assumed that all computer science/software engineering etc graduates were previously teenage hackers.

   One positive indicator I think you missed is that most, or at least many, great programmers have excellent command of written language. This isn’t quite the same thing as being good in conversation, where being quick-witted gives some people an illusion of talent. I would certainly not hire someone who couldn’t form coherent written sentences - if they can’t do it in a natural language I wouldn’t have much confidence in them doing it in a strict programming language. (Unless you are hiring for VB work of course, in which case the reverse may apply.)

   Also, something that has stuck with me long after I forgot where I read it (possibly the Jargon File?) is that you should look out for developers with highly-developed outside interests. Generally, the ones who excel at or show extreme dedication to their hobbies are better hackers than those that don’t. I’ve got no idea why this should be the case, but I have seen many examples of it. Perhaps it’s because they find inspiration in strange places, or perhaps because they solve as many problems subconsciously as they do in front of a keyboard.

5. 

   Vitor Pellegrino about 1 month later:

   I agree with everything except by the fact i consider myself a pretty good programmer as well even though i started to program at college; i think the passion and self-learning attitude are more important than any “pre-college” experience.

   Keep the good work! Regards from Brazil

6. 

   daniel about 1 month later:

   Hi Vitor,

   As I mentioned in the article, but it is worth repeating, these are all only indicators. None of them is sure-fire. You will find excellent developers who break some of those indicators. Marcel Molina Jr, one of the leading lights of the Ruby community, started his career as a Linguist before falling in love with Ruby. However, I think that few great programmers break all of these indicators.

   Thanks everyone for your comments!

   Daniel

7. 

   Marc about 1 month later:

   >I consider myself to be a pretty good programmer.

   Well, then you don’t want to hire me!

   Confessions of a Terrible Programmer

   :-)

8. 

   daniel about 1 month later:

   Marc: Good article, quite smart :-)

   Looking at your list, I see you actually match my list pretty well: you’re opinionated, you know what you like, you’re articulate, you’ve got experience in a good variety of technology stacks… I don’t know about the rest, but that’s a good start!

   But I’m guessing you didn’t post this to start an argument :-)

9. 

   Daniel about 1 month later:

   I am a self taught programmer and this list is almost the same as the one I use when hiring on new programmers. Our team is like 50% college educated, 50% fully self taught, some guy has a biology degree, etc.

10. 

    DR about 1 month later:

    Hey, we’re all good programmers, aren’t we!

    I completely disagree with this essay.

    Basically, you took a description of yourself and said “Here is what a good programmer looks like.” On the other hand, there are plenty of bad programmers that look just like that as well. Of course, at least half of them believe they are good programmers.

    I also know good programmers who weren’t into computers when they were growing up. Most bad programmers will list all of the latest technology on their resume to make it look good.

    Anybody who tells you in an interview that they decided to be a programmer for the money, will probably fail the intelligence part as well, so at least that one is right. The young woman who I got to babysit my children once was overheard at a party asking “Give my name out for babysitting because I love money.” Needless to say, she turned out to be a lousy babysitter, even though she had all of the qualifications (first-aid certification, etc.)

11. 

    Ed about 1 month later:

    @Ashley Moran

    Some of the best developers I have ever seen are dyslexics or have other similar problems with natural language…so I think comparing written work at the lowest level may not be the best indicator - I agree that verbal statements may however be some of the best structured sentences out there however.

12. 

    Anonymous about 1 month later:

    Gott… I have all but intelligence :-(

13. 

    Henrik about 1 month later:

    I think there are many good things in this list, but… the inevitable but… i think many “hacker” types are terrible programmers. Not that they write terrible code, but from a business perspective. I have many times seen programmers use bleeding edge technologies in places where it was inappropriate from a architectural point of view but .”…but this is really cool, we NEED to use this pattern/ technology etc.” and then clutter the application with unneeded complexity, just because “its cool”. Naturally a university degree will not safe guard you from bad designers but i prefer programmers who have the big picture in mind and not go goofing off with their pet technology

    Never the less, there are many good points in this list. :)

14. 

    Jeff about 1 month later:

    Considering I’m not in a position to review the talent of all that many developers, my opinions aren’t based in nearly as much hard data as I’d prefer. However, I take exception to these two bullet points:

    Opinionated about which technologies are better for various usages

    Very uncomfortable about the idea of working with a technology he doesn’t believe to be “right”

    Knowing a lot of technologies and being able to determine which to use in specific circumstances is clearly a benefit. However, being overly opinionated on the matter could easily lead to conflict. Team dynamics and the social aspect of software development, such as being pleased with the people you work with and the environment you work in, is very highly correlated with successful software projects. If one developer is going insane trying to convince the team to do things their way, it will harm the project significantly more than using a slightly less skilled developer who won’t cause problems.

    Furthermore, such developers as you listed are prone to becoming magpies (c.f. http://www.codinghorror.com/blog/archives/000916.html )

    Anyway, as I said, I may just have read too much into what you wrote.

15. 

    suffering fools about 1 month later:

    I work for a large software company in the Seattle area…one thing I am really surprised by is how poorly managed the interview process is. They ask puzzles, even though officially they are not supposed to. So reading one of the books on puzzles is a qualification. And they ask programming questions that are either impossible in the time given or for which they are looking for a very specific answer to regardless of if there are other valid solutions. I think they want to see the solution they received in their CS course.

    Basically we don’t consider any of the ideas mentioned in this article. And since anyone on the team can sink a candidate then good candidates are sunk while candidates who have studies to the interview get through. Its frustrating to watch and its frustrating when trying to make lateral moves.

    But hey, if upper management doesn’t care what can you do? The worst situation is when upper management is happy with the products you are producing, profits are up, even though you feel, nay know, they are total @rap.

    Sometimes all you can do is plan for an exit.

16. 

    pp about 1 month later:

    I’m very suspicious of this, it looks to me like you’re projecting an (idealised) version of yourself. The first attribute you describe as passion is really just a hobbyist mentality, you even say they will do it without being paid. Number 2, love of learning, follows on from #1 and is a useful attribute for a lot of workers - when it’s focussed as initiative and not just limited to collecting new skills like a butterfly collector takes specimens. Some may even call it a lack of loyalty, as they are just using you as a stepping stone in their career. Number 3 “… if you can’t have a great conversation … ” is a bit of a giveaway about reflecting your own personality. Ever considered they may not want to talk, or have other things they’d prefer to be doing than brown-nosing with the boss? Like getting home to the husband or wife (or can’t a great programmer have a home life and a family???)

    Pretty much all the other attributes fit any good worker in any field, you could just as well say “car mechanic”

    My measure of a great programmer would be one who does what they say they will, on time, in a clear and organised way. They will be able to integrate into a team and use his/her initiative to circumvent problems. They will be able to communicate at a professional level and have a JFDI mentality - without any tantrums about “my way is better than your way” languages/methods/coding style etc. But most of all, their code will work. It will be efficient, reliable, tested and fully documented so that when this great programmer gets a greater offer, what they leave behind can be picked up easily by the next great programmer you hire. That’s the most important attribute of all.

17. 

    MikeW about 1 month later:

    Great post. The one thing I would add is you should look for these types of attributes in employers when you are job hunting. Often if an interviewer shows no passion or interest in what they are doing, it probably is not a place you would enjoy working.

    In addition, you can apply these same principles to almost any IT job, not just programming.

18. 

    Jonah H. Harris about 1 month later:

    Daniel,

    Awesome post. In my experience, I too have found the same common criteria for all good programmers I’ve hired and/or worked with.

    -Jonah

19. 

    Steve about 1 month later:

    Wanted to mention that often a developer that was required to acquire certificates for a prior job (for whatever reason) will often gloss over them because they know how meaningless they really are, whereas a poor programmer will talk them up no end and make sure to emphasis them.

20. 

    Gordon about 1 month later:

    I think the passion indicator is entirely overrated. I see cases all the time, especially in college, where a person’s passion exceeds their abilities. They are the same people who spend day and night in the computer labs and even know a plethora of cutting-edge technologies, but are ultimately mid-tier students who are passed up by truly good programmers.

    With that said, raw intelligence is the best indicator. That’s why puzzle questions are not entirely bad. They generate less false-positive, which is the worst-case for a new hire, and are harder to fake. All of the qualities you listed above can be faked easily. They are also universal in the sense that the business-minded variety are capable of asking them.

21. 

    unknown about 1 month later:

    You must got really hairy balls

22. 

    Obiwan about 1 month later:

    The simplest way to check for 1337 programming skills: pony-tail

23. 

    Kevin about 1 month later:

    The best programmer I ever worked with:

    1) Had a life outside work and did not in general talk shop outside his 6am-3pm schedule. He had once been “passionate” about his code but after too many projects got canned he realized that in the end it is just a job. (Corollary: why do so many programmers think the rest of the world isn’t passionate about THEIR work?)

    2) Had EXPERIENCE in designing and implementing larger systems.

    3) Always wrote his code so that it could handed off immediately to someone else. Nothing was ever his “baby”.

    Those are qualities I see way too rarely these days.

24. 

    Daniel about 1 month later:

    Great article! A lot of what you said rings true with me… and of course I was flattered that most of your points are things I could say “yup that’s me” to :)

    The opening quote was also interesting because I’ve never considered what a huge task it would be to be the “business guy” who has a great idea but has trouble hiring good talent.

25. 

    sayinghai about 1 month later:

    Why didn’t you write this when I was 18, bright eyed and bushy tailed - and looking for my first programming job! :)

26. 

    David Emery about 1 month later:

    Good list and I’d like to emphasize #5. My position is that I -never- hire a programmer without experience/knowledge of more than 1 programming language. The Sapir-Whorf hypothesis applies; if you know only 1 programming language you can’t really think through alternate approaches to the problem. I’ve told people who clearly qualified based on the other criteria in this list, “Go learn another programming language and then come back and I’ll hire you.”

    See also this article: http://www.stsc.hill.af.mil/CrossTalk/2008/01/0801DewarSchonberg.html

    With respect to #4, some of us are old enough that our first chance to do any programming was when we got to universities. And that’s a point I’d like to make: Experience counts. Someone with 20 years experience, even if it’s not in the hottest technologies, but that otherwise meets these criteria, is often likely to be more productive in the middle to long run, than someone with 1-2 years of direct experience but no “time-in-life”. A tremendous amount of stuff I experienced 20 years or more ago in some “antique” technology is still relevant; part of the skill to look for is an ability to apply life experiences to the new technologies. Remember that 1 good programmer significantly outperforms several mediocre programmers (based on both skill and Brooks’ Law about team sizes.)

    dave

27. 

    Michael E. Burke about 1 month later:

    A very good article! I started programming back in 1965 on IBM 1401’s, doing cryptology for the Army. (When other guys were working on their cars I was building Ham Radio sets and Stereos…Vacuum tube technology.) In today’s market I’m almost unemployable. Why? Because I’m 60 years old and have had over a hundred jobs (contracts). But I do program for myself. I love assembly language programming, I can write competent programs in dozens of higher-level languages, I love to THINK about how to solve difficult problems, and I love the fact that there is a wealth of opportunity in the form of unresolved, undiscovered, un-articulated possibilities. In my lifetime I have seen programmers hired who had a lot of paper credentials and could not think their way out of the lunch room.

    Something that is implied in your article, but not mentioned explicitely, is a system for describing problems and then a system for resolving them. At this stage of my life I think that coding gets in the way of real programming, and a good programmer solves problems ‘way upstream from the actual code.

    Another good indicator is a person who gets so involved in the process that their coffeee gets cold, they lose track of time, and don’t get up until they absolutely HAVE to go to the bathroom.

    Mike

28. 

    Richard about 1 month later:

    As someone that has programmed for nearly 15 years (hobby and professionally) with a background in psychological research I hate to say I find this “essay” not very scientific. This is all based on anecdotal evidence, stereotypes, and conclusions were reached without a solid methodology.

    Can I find people that fit these bullet items that are good programmers? Sure, but I can also point to some really bad programmers that fit this model in nearly every way also.

29. 

    macc about 1 month later:

    prefer prgrammers that can use a soldering iron or have other hardware expertise. “pure” programmers are often unable to “jump”.

    This meshes into the no single stack argument G! MACC

30. 

    Anonymous about 1 month later:

    Yup, I agree with you. I would add another point:

    Values simplicity.

    This is the rarest of things to find.

    We programmers tend to have rivers of thoughts flowing in our brains all at once.

    A good programmer will find an elegantly simple solution in a word of complexities.

    Ask them about the K.I.S.S. principle :-)

31. 

    Franky about 1 month later:

    Great! Now I have a list of things to act out when I am applying for all those crappy boring jobs. Come on folks we all know that 95% (or more?) of the work out there isn’t what we’d rather be doing.

    What percentage of you are going to even get an interview at Google, Apple, Microsoft, or some sexy startup? (OK, I did at two of them…but that’s not my point.)

    My point is it can be hard to fit this list when you are interviewing at Billy Joe Bob’s Insurance Co. for some churn and burn IS programming job.

    Oh well, at least we all now have a guide about how to act to get a job ;-)

32. 

    flambard about 1 month later:

    Your list covers only half of the requirements of a good programmer. You might say its the basics. Whitout that you cant be a good programmer. But I have know a lot of programmers who got all the skills, abilities and markers you list and still where lousy programmers. By a lot I mean 60 - 70 %. Tinkering with code and technology has only little to do with writing clear, consice, maintainable code. Picking a new sexy technology is probably the best way to screw up a serious project. Jumping from one interest to another does not show the disipline you need for good programming. A good programmer can balance his skills and eagerness with the need of the programm/employer. Not to mention the social skill he needs to understand his employer.

33. 

    James about 1 month later:

    Your list is completely wrong. I am a great programmer and you have missed on every point…

    Actually, I was reading through this and found that I fit in pretty well with many of your points. While I would like to think that your list is good because of this, I do not consider myself a good programmer.

    One of the issues that compelled me to comment, was about learning a first language in a college course. I am mostly a self taught programmer, which shows determination and interest in the subject. It doesn’t really mean that I am a good programmer. Because I am self taught, I may have sloppy code that gets the job done, though horribly inefficient. A more fitting title for this article is “How to recognize an enthusiastic programmer”.

    An enthusiastic programmer can very likely be a good programmer, but not always. A good programmer should have a more intimate knowledge of programming procedures and logic, then of any particular language itself. Most any programmer knows that there are many ways to construct a procedure, but a good programmer knows how to create the most efficient procedure. Knowing the difference between and the proper use of a ‘for each’ loop verses a ‘for i=0-10’ loop is only one example (and one that I actually learned more recently in my career).

    Though an enthusiastic programmer is a good one to have: Enthusiastic != Good Good != Enthusiastic

    I have no problem calling myself an enthusiastic programmer. Instead of learning new languages and technologies as I have done in the past, becoming a good programmer has been my ‘hobby’.

    Your article was great, and I very much enjoyed reading it. You also warmed my heart a little. :)

34. 

    Raymond about 1 month later:

    I am not sure how I feel about this list.

    It is a good overall pattern to follow for someone who does not have strong development experience, such as business manager. Employment is one area where trial and error is not a good method for success, so examples from experienced people are very helpful. Still, I have to strongly disagree with two items on the list. At one point in my career, I would have agreed with them, but additional experience working in the enterprise has changed my mind.

    1) someone who suggests bleeding edge technology does not know what it takes to develop real world solutions and maintain those solutions. I consider this to be a major indicator of inexperience. Good developers have lots to do, so they cannot afford to spend large amounts of time evaluating things that might or might not apply to completing their tasks. Good developers know how to use the tools they have to get the job done. New tools tend to waste time. I am not saying that an interest in new technologies is bad, but I am saying that someone who suggests AJAX as a solution rather than just a potential method for achieving an end probably does not know much about real world web development.

    2) Good developers do not spread their time all over the place learning every technology under the sun. Good developers understand that solving problems is the name of the game, and to be efficient you have to focus. I evaluate developers for new hire all the time, and I am suprised by how many different technologies they put down on their resume. How do you obtain expert experience in a technology within 6 months? You barely understand syntax and overall theory in 6 months. It takes years with a technology to understand what it does or does not do well, and where all the gotchas are. Focus is much more important than spreading it around.

    I usually spot good developers through one easy test. I mention a problem that we are having to them. I look to see how they approach it. If they start breaking it down and asking questions that get to the heart of the matter then they are good. If they start asking if I have tried this technology or that product or tell me about their agile methodology for tackling problems then I politely escort them to the door. A good developer knows how to see what is important in solving the problem and knows how to get there quickly.

35. 

    Jared Parsons about 1 month later:

    I think you missed one key factor that I would put at #3. The ability to systematically break down large problems into small, managable ones.

36. 

    programmer about 1 month later:

    …and then there are the morons who look to get someone who is a great programmer/thinker/developer and then “test” you with arcane references within their required language stack that have no bearing on being able to create or develop and then rub it in your face that you must not know what you are doing…because his onscreen definition that he found on www.howcomputerswork.com is not what you just said….

37. 

    unknown about 1 month later:

    “If your potential programmer didn’t do any programming before university, and all his experience starts when she got her first job, she’s probably not a good programmer” - this is not a valid statement. True, if you start programming earlier and continues to do so, you would probably be a good programmer. But I am a good programmer who doesn’t start programming before university. Other than that, I agree with you mostly.

38. 

    macc about 1 month later:

    After reading some more recent comments:

    Avoid programmers that think of themselves as “great programmers”.

    And yes, KISS is allimportant.

    G! MACC

39. 

    Nelson about 1 month later:

    I don’t think someone that goes home from a long day of programming and wants to do something other than program is a sign of a bad programmer.

    My main criteria are:

    1) ability to think logically

    2) works well with others - knows when to take a stand and when to be pragmatic

    3) a certain creativity and intelligence that can take a complex problem and break it down to its fundamentals (writes solutions that can be understood by “the next guy”)

    4) curious - wants to know “why” - solutions can often be improved with a deeper understanding of the problem or task desired (or of the target system/environment)

    Disclaimer - My list might be more relevant for software engineering than just “hack ‘n slash” programming.

40. 

    Peter Rowell about 1 month later:

    Good article.

    One error I’ve seen managers make (myself included) is finding a programmer that is good in one area and then later trying to fit them into something entirely different. Crossover is possible, but the programmer has to be interested in the new area and self-motivated to explore it.

    E.g., good in compilers != good in computational linguistics, even though on the surface they may look related.

41. 

    quantum kev about 1 month later:

    I agree with Pp (way up there). Although I agree with some of your assertions, I have found in my experience that many great programmers are not necessarily great employees. I have worked primarily for corporations that deal with sensitive data, and they have all been very hesitant - even somewhat apprehensive - about implementing the “hottest” technologies. Often times, security holes or other vulnerabilites are only truly exposed until after a product has been out for awhile. So it is important, in my opinion, to have a solid foundation in the basics, like VB, Java, C++, or even good old ASP, HTML or JSP. Sure, it’s great to know Ruby, AJAX, and all the hottest stuff. But I can tell you from experience that those technologies are not to be found in banks, hospitals, insurance companies and other large organizations - some of which are still running Windows 2000. The other major objection I would like to voice is your point that “most average or poor programmers started programming at university” - I don’t think I need to beat this dead horse, as several others have addressed it. The best 2 programmers to come out of my class in college were myself, and my buddy - both older guys just learning to program. And the reason we were the best is my final point : to me, the one most important quality of any programmer is DEPENDABILITY. When given an assignment, is it completed on time, and as requested? Many of the young guns I went to school with were avid hackers, and very talented. But they could never complete anything on time, or as requested - if it wasn’t something illegal, or “fun”, they simply lost interest. It doesn’t matter how many technologies a person knows, or that they programmed their first OS at 10 years old, or that they cracked their own iPhone - in the business world, if you cannot be depended upon to deliver requests on time, and within required parameters, you typically don’t last long.

42. 

    PhotoTherapy about 1 month later:

    Well, I am happy to know that I am almost a good programmer ;)

43. 

    Tarakava about 1 month later:

    This is the first time I ever read an article that assumes the meaning of “smart” is great social skills and the meaning of “dumb” is lack of social skills, am I losing something?

44. 

    unknown about 1 month later:

    You obviously don’t have your SCJP to say that it is completely irrelevant.

45. 

    unknown about 1 month later:

    Daniel is someone who makes stupid decisions.

46. 

    piedoggie about 1 month later:

    crap, crap, crap. I hit every one of those points. Self driven learning, always exploring new forms of technology and then applying it to the current business, etc. etc. Maybe if I hadn’t been so good, my hands would still work and I wouldn’t be living with constant pain up to my elbows. What’s ironic is I left my father’s rigging business (machinery moving) because I saw so many people around me losing fingers, damaging their limbs and back etc.. and I wanted to go do something where I wouldn’t be injured on the job. Since I had been programming in high school, I thought hey, this is not a bad career. You work indoors, you’re not covered in grease, you don’t breathe toxic chemicals and you aren’t going to get injured by heavy machinery falling on you.

    Ha.

47. 

    Chip Uni about 1 month later:

    I slightly disagree with one bullet point:

    • Happy to work with whatever technology you’ve picked, “all technologies are good”

    I think that you need to discover what motivates the programmer.

    If the project’s goal motivates the programmer, then she might be happy to use any (reasonable) technology to reach that goal.

    If the programmer is coming from a different platform, she may choose not to have a strong opinion about which particular technology to use.

48. 

    Steve about 1 month later:

    WOW! Sorry to have arrived at your article so late! I really can say much of what you have written sounds similar to what I do when sourcing/reviewing/hiring new candidates. I think that your bulleted list gives clear enumeration to what we all should be aware makes a good startup candidate, period! :)

    Love the info and perspective. I’ll be relying on this piece from now on!!

    Thanks and take care,

    Steve

49. 

    Merijn Vogel about 1 month later:

    about new vs old technology

    In programming-land I am a rare case. I like to take on ‘challenges’ on old software code that other people created, modernize it where necessary and make something useful of it. In my company people always are complaining when some other colleague created something, left the building and they have to figure out what and where, however I do like that.

    It gives a bit of a base, you’ll have to work around things that are weird/hacked in, etc. The fact that I do like that, makes that I may become stuck in “Old Tech”, like java 1.4 whereas java 1.6 and such are more current. That’s where home-hacking comes in; the company is in web-space, so at home i try to develop a suite of gui applications, etc.

    And trying to learn a programming language you do not know each year, in a few weekends, is always a very nice experience I my opnion.

    Nice article, and nice discussion afterwards, which amazingly is still going on.

50. 

    Jana about 1 month later:

    That’s real good insight into the process of hiring. Except maybe for one thing. Before you start interviewing people, you as an employer need to cut the list of applicants short by about 90 %, unless you have unlimited time. And most of the criterias you mention do not show up in a CV.

    Still, a one-in-ten chance you happen upon a great programmer!

    But really, what the CV-bound hiring process does is it swings the pendulum at least 60-40 in favor of people with great communication skills. And if you know your DISC, you’ll understand that is not the main trait to look for in an employee with a problem solving profile.

51. 

    mike about 1 month later:

    @gordon

    how many of those “mid-tier students” go onto become great programmers b/c of their work ethic and desire; whereas, how many of those “great programmers” at school, who do not have this work ethic/passion, inevitably flame out and end up hawking some “mid-tier students” tech in pre-sales calls???

52. 

    Anonymous about 1 month later:

    I don’t get the bit about “started programming before University”. It doesn’t make any sense to me. Your other points are equally unreliable.

53. 

    James Priestley about 1 month later:

    “I usually spot good developers through one easy test. I mention a problem that we are having to them. I look to see how they approach it. If they start breaking it down and asking questions that get to the heart of the matter then they are good.” – Raymond’s post, Jan 11

    And if they offer a solution, they are worth hiring. Which is how my oldest son happens to be writing code at FaceBook.

54. 

    smith about 1 month later:

    I think your comment about the “exotic” ruby libraries is completely wrong, and perhaps leads to the problem with your argument. Writing a website is simple. Any douche can do it. Rails code is so simple that even NON programmers can do it - so it shouldn’t be held up as a sign of greatness.

    When I look for great programmers, I look for the following:

    1. Fluent in lisp or prolog - there is no prolog/lisp programmer who cant pick up ROR in an hour.
    2. Understand to at least an undergrad level another field (soft sciences).
    3. Can design complex software (database indexing, data mining, distributed memory) with
55. 

    Paul about 1 month later:

    You’ll catch flak but this is all essentially on target. In particular, you’ll catch shit about the ‘formal qualifications’ thing. But it’s closer to be being true than not.

    Most non computer people don’t realize that coding is far more of an art than either a science or a craft, much like painting or writing, I don’t think you can teach someone to be good at it. I think you need to be ‘wired’ a certain way to be good at it. In other words, born with the talent.

    As such, hiring is difficult, it becomes almost more of a ‘tocuhy feely’ thing.

56. 

    John about 1 month later:

    I think it’s all about time. The best programmers can do the most work in the least amount of time with the least amount of errors. The gurus can do that and make it easy for other people to understand. “They make it look so easy”.

57. 

    Programmer and former software development manager about 1 month later:

    I think you forgot couple. These come to mind: