Hacker News Reader

FizzBuzz is famously hard, but it's not that hard

Every time I see something like this, I cringe inside and wonder at how memes get distorted.

FizzBuzz is not supposed to be hard. In fact it is supposed to be as EASY as you can possibly make an actual programming task. So easy that most programmers would have trouble believing that there are people calling themselves programmers who can't solve it in their sleep. Seriously if you know enough of a programming language to print output, run a loop, take remainders, and write if tests, then you can write FizzBuzz.

Put another way, if you can't write FizzBuzz, then you probably can't write ANY useful program. And your interview can safely stop right there because you failed.

If a middle-class white boy who literally never had a job before getting a sweet internship at some cutting edge technology company can eventually, through practice, become a passable computer programmer, anyone can do it.

-- This is not of the form you reference, though

This is an argument about barriers to skill acquisition, and thus representative expression, not one of either innate or even actual ability or qualification level. Its not "conflating ability with opportunities" anymore than it is saying "there is learning by doing" and since the "doing is not evenly distributed"[1], therefore neither is the "learning".[2] Author is in part arguing that the only "doing" that matters is "branded doing" (ie, doing it for XYZ company or school), as as that is how the next level of doing is rationed, and there is therefore a recursive effect. This may or may not be a great argument, but I think its different than the one you allude to. In particular the low-threshold state and the (implied) recursive mechanism seem to have a more interesting dynamic.

______

[1] Or the option, etc.

[2] This depends on (or implies) a certain elemental/relative triviality of the task.

"This is an argument about barriers to skill acquisition"

I don't see that argument in her post.

getting a sweet internship at some cutting edge technology

Author is arguing/implying that overcoming this "gate" was the source of [the example's] later (developed, proven,etc) skill. It might not be a perfect argument, but its there IMHO. I think if you believe it's not a good argument/ point, its best to attack it directly. That is all.

I don't think she's saying that. Do white males without programming skills get programming internships at a higher rate? Are there even a significant number of people who take that path?

Hiring is not the problem. Getting underrepresented folks to choose computer science in the first place is.

Now I think you are attacking the point directly. I can't speak for the author but it seems to me that the person does believe this. viz "it’s a pretty fucking easy job..." etc [1]

___________

[1] The whole context is as follows:

If there’s one thing I’ve learned after thirty-three years of being alive, it’s that if you see middle-class white boys flocking in droves to a particular career path, it’s a pretty fucking easy job and you should try and get yourself one like that.

I guess that’s a little mean. Sorry, middle-class white boys. I’m not calling you dumb. I’m calling you soft.

So then, are table tennis, boxing and basketball or gymnastics not meritocracies either and thus deserving of more representative demographics? Should those sports seek to find more balance in their professional profiles? Should leagues try to find and be incentivized to fill their ranks with underrepresented demographics?

Are industry responsible or is the educational system responsible for developing interest and providing guidance to the newer generations (in addition to initiative by the student body)?

Note [2] This depends on (or implies) a certain elemental/relative triviality of the task.

I think you missed that note [2] above. I'm not necessarily supporting or defending the position. The point was it was not as simple an argument to dismiss as was suggested in the Parent. Couple of points to consider, more generally: (1) Its not clear that the condition of footnote 2 holds, per-se; That being said (2) its also very far from clear that decisions of Merit (in the context discussed) are ever/always fully informed.[#] Ironically, the further away you get from footnote 2 in some areas, the less informed the key gatekeepers are.

In other areas that might not be the case. These would be characterized by non-trivial skill obviously required, easily, cheaply, and repeatedly observable/measurable by multiple parties, etc. Timed running in 100m, would be an example of the latter.[##] But this is circumventing the actual argument by changing the relevant context. So its sort of a valid counter-argument, but its not quite-central to the point the Author was trying to make.

________

[#] see comment> http://news.ycombinator.com/item?id=4658045

[##] Still, this might not be the most diverse group if the olympics are your benchmark. But its non-trivial but highly specialized, and we can presume uneven native talent distrubution cannot be ruled out.

> I think you missed that note [2] above.

I think you need to edit your posts more carefully. Clarity matters, particularly when you're discussing things in written form.

I'll take this comment at face value as a legitimate critique of my writing. In general its Clarity, Brevity, Precision = Pick 2 =D. That being said, I just want to point out one thing to the general reader, which is that the allusion to the footnote was not meant to be pedantic or a put-down.

The above poster responded to a weakness in the OP's argument, and leveled a valid counter argument. My footnote was there to highlight this same weakness in the argument. But my footnote wasn't really there as a point of exposition (either), and in hinting at the weakness it also articulated the weakness in a way which illustrated a more valid (central) line of attack. That's why I highlighted it again.

But I could probably have chosen better words in my response.

It was intended as a general critique.

> In general its Clarity, Brevity, Precision = Pick 2

One of those would be a good start ;)

Those three actually go hand in hand. I reread your initial post - I'm still not sure what you're trying to say. You and the post you reference seem to be saying the same thing.

Footnotes are supposed to be extra bits, not central to your argument, and you seem to ramble, and have complicated sentence structures, rather than cutting to the chase.

You and the post you reference seem to be saying the same thing

-- Ha ha yes, That is the chase. I was re-phrasing the OP/article argument, which was mis-understood by the parent comment.

The footnote articulated something else altogether, namely a weakness in the argument. It was an more-deeply implied (or imlicit) assumption (so I didn't spell it out), i kept it seperate.

The post you are referring to picked up on this same weakness, but attributed it to me. Which wasn't the end of the world, but his counter-argument was lacking to actually critique the original Post/article.

Hope this helps clarify, the general direction.

Its a tough call on when to clarify a weak or potentially flawed argument, and when to actually argue against it. Here, I thought it added something interesting (namely, recursive logic), which is worth noting and considering.

Without throwing the whole insight out.

If I was smarter, I could probably do that faster, better, stonger. As Arnie says, I'll be Baak. =D

... trust most people saw the logical flaw - the author is assuming that the ability to program is currently evenly distributed.

I don't see anything from the author which can be construed as the statement, "ability, is at present, evenly distributed." If you withdraw that stipulation, what you have is:

4. Ergo, opportunities to become a computer programmer [e.g. education] are not distributed according to merit [e.g. by identifying and cultivating potential in individuals.]

Which is entirely defensible as a conclusion.

The "at present" stipulation doesn't really change anything. The author is claiming that the ability to program is evenly distributed, but the plain truth of the matter is that more white English-speaking males in their 20s can program than can <fill in demographic here>. Which is because of, yes, racism, sexism, inequality, and a host of other societal ills, but it dosn't change the fact that Joe's House of Discount Programming in Des Moines Iowa gets a lot of resumes from people that look like us here on HN, and very few resumes from middle-aged black women who speak Swahili.

The HR department at Joe's House of Discount Programming may well be biased, but you can't conclude that from the evidence presented. More broadly, you cannot tar the programming profession as racist and sexist because we don't hire people who don't actually exist (even if, yes, the reason those people don't exist is because of racism and sexism and all the rest).

Um, you don't think that calling the given situation meritorious would be racist/sexist?

1. Construct a system to deliver engineering educations. For whatever reason, white males are overrepresented in this system.

2. Place those engineers at top engineering firms and startups, where they succeed.

3. Conclude that the system rewards effort and ability fairly, and therefore the current beneficiaries are the best and most deserving.

Step #3 is ignorant and stupid.

To cast this another way, this would be like going back to a pre-Jackie Robinson period of baseball history, or the pre-integration period in basketball history, and concluding that the best baseball players in the world are white.

Yes, in a very literal sense that is true, if you define basketball players as only those people who can right now play basketball professionally, in a system which is accessible only to a small subset of the population. Even if there were people playing amateur or street ball, who had great raw potential, without being able to enter programs where they could hone and refine their skill they would be inferior in many ways to the pros. And a person who would hew to that literal rendering would be considered myopic, shallow and intellectually dishonest.

Similarly, to say that the current crop of engineering talent represents the most deserving talent in the world is a backhand to all the people who are excluded from participating in the system to begin with.

So yes, you can't say "programming is a racist institution because it doesn't hire people who don't exist." You can, however, say that programming is a racist institution because it benefits from a lopsided system of selection, and then arrogantly declares that the selection system is fair, that those selected received no undue benefit in their selection, and therefore those selected are the best.

I think you're missing the point, which is not calling you racist, but opening your eyes to the fact that much of what you have achieved (and I don't mean you specifically, but the class of people the article adresses) is not a direct result of your effort but largely due to privilege. Thus, as a person of privilege you should use whatever political power you have - and you have quite a lot of it - to help rectify the situation, or, at least be constantly aware of it.

Merit does not require effort. I haven't heard anyone claiming that their ability to program is a measure of their worth as a human being. It is, however, a measure of one's usefulness to a development team.

Ok, but isn't it in the interest of all software employers to use their political power to create more capable developers? This is simple economics; I'm not even touching ethics or justice here. And isn't it the immediate interest of current developers to maintain their value by blocking others from entering this profession? Sometimes we adhere to hidden interests without even realizing. Let us first open our eyes to how society works and how we may, inadvertantly, keep others down.

Those incentives are there, but the linked post isn't about that, nor is it a particularly useful starting point for that discussion.

Probably not directly, but that's implied in the word merit, and in the higher pay and better conditions afforded to programmers.

2) Most computer programmers are white

Maybe in the USA, but in that context this is not surprising, as 72% of the USA is white. (The male part is a separate discussion) Do you have proof that when considering the entire world, most computer programmers are still white?

Most places I've worked at really do want to hire a diverse group of engineers. At one company, one of the managers was asked by his manager to hire e.g. more black women. His response was "you show me one, and I will hire her".

Personally, I'd like more women in the field as well. I wouldn't mind working somewhere where, I don't know, even 1 out of 3 people were women, instead of 1/10. But I think availability has a lot to do with it.

"FizzBuzz is famously hard"

Are we talking about the thing of writing "fizz" for multiples of 3 and "buzz" for multiples of 5 and "fizzbuzz" for both, for the numbers 1 - 100? If that is famously hard, then the definition of hard must have changed massively while I wasn't looking.

If the vast majority of people applying for programming jobs can't solve it, then it qualifies as hard by definition.

It is hard, only in the sense that programming anything is hard, if you don't know how to program. If you can program at all, you can solve it. It definitely wouldn't qualify as 'famously hard'. Things that achieve fame for being hard are usually very very difficult indeed. If anything, in the narrow circles in which it has any fame at all, it is famously easy.

I agree with your analysis. However I've only ever sat in on a few programmer interviews, and was lucky enough to get only the few that had already made it through several screens and were quite sharp. However, if so many people can't do FizzBuzz, that's baffling. If they can't do that, what can they do? I mean that literally. Do they just have knowledge of certain specific frameworks or something? Can anyone shed light on this?

I know this may be derailing the conversation, and it bothers people, I'll not mind deleting this post.

One issue I've had is that if a question is too easy, I start out thinking "this must be a trick question..." and then tie myself up in knots trying to find the (non-existant) catch...

Of course in an interview, one should presumably just say "hey this seems too easy, is this a trick question?" but ...

That reminds me of when I was first learning to program. I wondered for an embarrasingly long time what the hell 'foo' meant. I knew it must be something really important as it was absolutely everywhere.

"famously hard" was an example of sarcasm.

A lot of people applying for programming jobs can't solve it. Any programmer, pretty much by definition, can. The lesson is that a lot of people who apply for programming jobs can't program. :)