Most of the comments (so far) are focusing primarily on the details of the Harvey Mudd example instead of the much broader, more important theme of mathbabe's post: She wants educators to think seriously about the selection process for young nerds and how this process is currently biased so as to weed out girls. She believes we can do much better.
She speaks from personal experience. On a different post on her blog, she writes, "when I and two other girls (and, as it happens, no boys) in the 6th grade did well enough to go into a special, advanced 7th grade algebra class, my (female) teacher brought us up to the front of the room and told the three of us 'I don’t see why you would challenge yourselves like this anyway since you are girls, and you won’t be needing math when you grow up.' I was the only one of the three of us to actually choose that class, and I was the only girl in the algebra class."
I think that speaks more to the general attitudes towards life of the teacher. The general premise is, "why challenge yourself when you coast through without putting in the effort?" It's wrapped up in sexism (girls don't need math), but boiled down to its base form, that's what she's telling those kids.
Agree -- that's a good way to summarize what that teacher told mathbabe and the other two girls. Regardless, of the three girls, only mathbabe persisted with math; the other two were unfairly weeded out. The question is, is this type of weeding-out pervasive? Mathbabe clearly believes the answer is yes, and we should we doing something about it.
I agree that those two girls were unfairly weeded out of math, and we need to prevent that.
My point was that that teacher either needs to take a different attitude towards life, or get weeded out of the system herself. Why? Because her attitude will get impressed not just on girls attempting to enter predominantly male spaces, but upon all students that she interacts with.
Mathbabe nailed a couple of key points on what people use to steer toward a career... knowledge the career exists, and belief they would be good at it. Most youth view their options in life based on what they see in their immediate social sphere. Having parents as mathemetician lets you know it's possible and probably also contributes to the belief that you could be good at it.
If we want more women programmers/mathemeticians/etc, we need to show female examples earlier on to let girls know it is possible and weed out the fools in the education system who espouse girl brains don't work as analytically as boy brains. I bet a lot of girls don't go to mathcamp not because they think it's "uncool" but because they either don't know it exists or don't think they would be good enough.
Actually there’s another reason I ultimately decided to call this blog “mathbabe,” namely when I googled it[...] there’s another site with a very similar name which simply shows nearly naked women next to cliff notes on basic math subjects. WTF?!? It is ridiculously obvious to me that math babes should be doing math, not adorning it. So I kind of had to call myself mathbabe after that.
Compare Hello, World! in the two languages, in terms of new concepts that need to be known to understand it. In C++ you need to know about including <iostream> and cout. In Java you've got all that class, public, static, blah blah nonsense.
I did a presentation at my old high school about why programming is awesome in general, and had a couple notes on this. I counted 18 unique facts that need to be mentioned in order to completely understand a Java Hello World, and 25 for C++. For Python, it's either 4 or 5 depending on whether you start with the REPL or not.
I don't want to get bogged down in the argument over strong/weak/dynamic/static typing -- but also don't think it's necessary for a beginning course.
The necessary HTML dings the number of unique things up so it's not quite that simple. But I like the idea of being able to easily share your creations, I started with PHP and a friend let me FTP files onto a subdomain he set up for me. I remember sharing a "99 bottles of beer" print-out with online friends when I learned about for loops. (I also remember spiking the server load by increasing the number into the billions...) I think a JS oriented programming class has as much merit as a Python one. We can discuss tradeoffs more (e.g. JS's style of OOP is superficially further from Java's than Python's is and AP culture still favors Java/C++ style OOP, which PHP is actually very similar to) but I think they're both better attractors into the field than Java.
When I took AP CS in C++, unfortunately, the finer details of iostream took up an inordinate portion of the class, which was particularly useless because iostream is really weird in a lot of ways (e.g. the stateful format specifiers sort of mixed in with the data being output). The part that wasn't taken up by that was taken up by memorizing some STL syntax.
Fortunately I actually taught myself programming in HS by writing mIRC scripts so I could do cool automation stuff in my spare time, and I had learned a little earlier in some kind of Logo course targeted at elementary-age kids. If it were up to that high-school C++ course to get me into computers, though, I would definitely not have done so...
The sad part is they can't realistically choose to use python any time soon, because the language is undergoing backwards incompatible changes. Until python 3 is the norm everywhere it's tough to teach to such a moving target.
I don't think this is a true concern. When I took the AP class we used Java 1.4. I'm sure you know how big the differences are between Java 1.4 and Java 1.7. The AP course is about the fundamentals of computer science, the language doesn't matter--it never touched Generics or Ant or anything a modern Java dev needs. But moving to Python allows for alternative curricula that's either pre-requisite or concurrent with the AP class that also uses Python and its abilities and easy-to-get libraries, like PyGame or Flask, to attract people into the field with shiny things.
And before that it was Pascal.... I know cause I tried to take it after teaching myself Pascal a day before the exam (I new C/C++). I got so caught up in the stupid syntax of the language that think I got a 2 or 3 :(
There is a CS course taught in Java in my high school, but the prerequisite teaches Netlogo, Scheme and Python. Schools teaching APCS should definitely consider having a prerequisite intro course if they have the resources to do so.
When I took the CS AP test in 1997 it was in Pascal. It was shortly changed to C++ and then was switched to the current Java. Most of my college CS (1998-2002) courses were in C, including first year courses. As it turned out my senior thesis was a Pascal compiler written in C that targeted the early JVM. Man I should have stuck with that, but my conclusion at the time was, while it was easy to target, the JVM was just too much baggage to carry around.
Personally, I think Pascal was a great choice for those high school courses.
You can do just fine with C++. Or Java. I have taught new programmers using both. I think you can do better with Python, because there's less initial friction. I'd rather not have to tell students to ignore "public static void main" because there are several concepts in that phrase alone that I'd rather they not worry themselves with. I want them to focus on developing the ability to think algorithmically.
But, as I said, you can still do this just fine in C++ or Java. I think that people tend to place way too much importance on the first language we teach students. What matters far more is the quality of the teaching itself, and the relevance and logical progression of the projects the students have to solve.
The problem is that a High School semester is really too short compared to how long it takes to learn C++, C++ is just too big a language. I ended up thinking I knew something about C++, when I really didn't.
And the hello world program for C++ is just as bad as the one for Java, in its own way. You're writing 'cout << "Hello World"' but how long will it be before you teach the students about operator overloading?
Note that I agree with you, in that I would rather use Python, or even Scheme. But I think language choice accounts for, say, a 5% impact, while the boring stuff (how much the teacher cares, how patient the teacher is, how good they are at explaining concepts, the quality of the projects - standard teaching issues) accounts for the rest.
In terms of preparing students for a career in programming I agree with you, the choice of first language doesn't matter much. But if a student is going to take a single semester in High School and never take another programming course, its far better that they end learning to write mediocre Python than bad C++.
I think it matters more than you give it credit. Michigan State did an experiment where they replaced one of the Intro to CS classes (which were formally in C++) with Python. The result they found is that for people who went on to CSII (in C++), they did just as well; but for people who were non-CS majors, a) they did significantly better than non-CS majors starting with C++, and b) they were more likely to take more CS courses. So if you're someone already committed to programming or computer science, maybe it doesn't matter, but if you're trying to get your feet wet, it matters a great deal.
Neither of your claims is supported by either paper. Rather, they seem to support my claim: starting language isn't a predictor of success.
Conclusions from the first paper: When ﬁnal exams were compared for a CS2 course
oﬀered before the Python transition with a course after the
transition no statistical diﬀerence was found. The ﬁrst CS2
course after the transition contained a mixture of Python-prepared and non-Python prepared students. There was
no statistical diﬀerence in the performance between those
two student groups.
Conclusions from the second paper: We conclude that starting students with Python has not
had a negative impact within a C++ based curriculum. It
would have been nice to conclude that the Python students
did better in these courses, but the evidence does not appear
to support that hypothesis.
With that said, I do agree with the authors that exposure to more languages, environments and tools is a good thing, and for that reason, I'd prefer using Python, then switching to C++.
My freshman and sophomore years, I was taught Java... the rest of undergrad and grad, I was living in a world of C and Matlab. I have no idea why it was taught in that order, and I think they're in the process of fixing it, but I remember my first C course when we started doing pointer arithmetic... quite the experience!
Having tried to tutor beginning programmers in Java, there are a lot of things that bug me. There is a lot of boilerplate and cruft that must exist in order to get anything done, and the type system and visibility just gets in the way until you want to teach people how to architect large programs.
Each class must be in its own file, and the execution starts from "public static void main(String args)" inside it. Then you have to deal with complex data types very early: python's set of primitives is much richer, which is great when teaching. Also, doing IO in Java is a pain unless you distribute a library.
To those who think Python's significant whitespace is a problem, I think you need to watch an otherwise intelligent beginner struggle with an error on line 80 because they mismatched their braces on line 20. If you aren't properly indenting your code, it becomes an unreadable mess anyways, so why not enforce it at a language level? And if you think an IDE can do this for you, think again. People just complain that it does things wrong when it indents half their methods by 2 extra tabstops because they missed a brace. They don't really get it, they just get into a feedback loop where they say, "Oh, it won't compile. Let me scan the entire program counting braces" which is a waste of everyone's time. I've seen people go entire semesters without ever learning to write readable indented code.
I "got" indentation after spending 2 days hunting down some bug in calculations only to have the teacher take 1 minute to reformat the code and another minute to point out an instruction that was sitting in the inner loop. With proper formatting it was a glaringly obvious mistake...
Because the hello world program has so much boilerplate. It means from the get-go you're telling students, "Don't worry about this, and this, and that; ignore it and we'll explain it later." You could go months into learning programming before you need to understand what a static method is, but it's right there in your first program.
I think a great deal of societal problems are caused by social 'pressures' and 'norms' (ie. bullying). Remember how terrified everyone was about being perceived as uncool? I remember when I developed an interest in classical music, i pratically had to keep it a secret.
One more point i'd add about the general immaturity and unfairness of school pupils, is that the smart boys are probably unkind towards to the smart girls, out of some sense of competitiveness. I don't have evidence of this but it seemed to be an abundant sentiment when I was at school.
Python is great, but I recommend that schools should start programming classes with Unity3D. Within the first class you can have objects with physics bouncing off of each other, and by the end of the course you can have a legitimate game to play (that can even be sent to your phone or tablet).
I'm a full time Unity developer now (see http://www.underthegarden.com), and while it's a really powerful toolkit, some of the bugs and problems I've encountered through no fault of my own have been pull-your-hair-out territory for me; they would literally be impossible for a beginner to overcome.
While getting something running quickly is great, I'm not so sure having your instructor fix stuff up in your project every couple of days is the best way to start a programming for beginners course.
It sets up the wrong mentality; that programming is complicated, stuff breaks even when you didn't touch it, or when you touch something completely unrelated, the best way to solve problems is to Google them or ask your instructor instead of think, and that learning the API more is the best way to get better at problem solving.
This is nothing against Unity in particular; it's amongst the best of its competitors, and I don't regret choosing it for my work; it's just the nature of the beast when dealing with massive existing code bases.
That said, I think getting started with graphics and sounds is a great idea; I'd just use something a lot closer to the metal and less complicated. Blitz2D would have been my choice 5 years ago; I'm not so sure now (how is PyGame?).
You need to walk before you can run and it is of NO BENEFIT to the student to hide programming behind a fancy UI with fun results before they know what is going on.
Let's be clear here, I am not against Unity. But I'm definitely against Unity for teaching how to program.
I know this because at my university they taught first year programming with java and used a book written by the instructor that showed you how to load an image with swing before teaching conditionals. The boilerplate code that allowed the students to load images did nothing but completely confuse them and negate the teaching aspects as he slowly introduced programming fundamentals. It was a complete disservice to try to "fancify" basic programming and all it did was confuse anyone who hadn't previously learnt to program.
Python on the other hand is a fast, interactive, forgiving interface that boils programming down to the basic nuts and bolts that you need to understand before interfacing with fancy UIs and physics.
Frankly, I think that getting people excited about programming and learning that they can create something actually cool is the most important part of their first introduction.
I took a course in high school that started with the nuts and bolts. I did fine because I already knew some basics, but most of my classmates struggled. They learned that programming is hard and no fun and in the end you've pushed some lines of text around. Most of them barely passed and will never touch code again.
For a serious university level CS course? Sure, start with Python and explain how the basic nuts and bolts work. Trying to get young people excited? I think it's worth considering other options.
How many of us really started out programming learning the basic concepts in a structured environment? I understand your point, but I think that to be so dismissive of the enjoyment and excitement factor is inappropriate.
I think Unity is perfect for an introduction to game development or game programming class, but not so great for a general purpose programming course.
The issue with Unity is that it is a complex, fully featured 3D game engine. A lot of what you do and learn, as a beginner, has nothing to do with writing code and is more about how game engines work. Like what a mesh is, what a rigid body is, how to set up a particle system, etc. I can see students getting hung up on these details to the point where learning how to script is a secondary goal.
This comes from experience. I've worked with a game designer whose only real programming experience was inside of Unity and he was difficult to work with because he didn't have the ability to separate game engine (Unity) features\concepts from programming language concepts. So I'd get weird questions like "why doesn't Lua have 3D object parenting".
The same thing can happen in any complex environment where scripting can be secondary to the idiosyncrasies of the environment (Unity, Flash, web browser, etc.)
To be fair, heapsort isn't what real programming is about either. (In fact I'd argue full-stack game programming combines more of technology principles than almost any other sub-field, not to mention non-tech principles in the art and design and story.) The GP's general point however is if you're trying to attract people into the field in the first place, dangling shiny objects isn't a bad way to go. I started with web sites with PHP and I liked the interactive nature of the browser. It wasn't until quite some time after that I stumbled upon Project Euler and had fun with those sorts of problems that I probably would have avoided prior.
I still don't like heapsort, but that's more of a general dislike with the "let's teach everything with sorting algorithms!" culture of CS curricula. I like other more-hardcore-than-sorting CS things much more. Edit: and I also have fun with them. Programming should definitely be fun.
> Not only do they help the student with writing code, the introduce important parts of programming as well as Big-O notation.
Are sorting algos the best methods for helping students write code and learning about Big-O? I don't think so. Learning the fundamental data structures are a lot more useful for that purpose, I think. Of the standard crop, I think the binary search is a good algorithm to teach (though one must note its historical implementation problems in languages with fixed int sizes), and if you're going to do sorting then quicksort + insertion sort when N < 16 wouldn't be awful. At least mention the language's built-in sort so that open source projects like gnu flex don't inherit a DIY sorting function in the code (in their case, the evil bubble sort which should never be taught).
>…"Lets learn Unity3D" course in front of me, I would never ever attend that university.
But you would probably still be doing programming and CS elsewhere. You're already set. This post is about trying to attract people into programming at the HS or college level who otherwise wouldn't be. But from the other comments it seems like I might agree with you specifically on Unity3D; I haven't used it myself but it doesn't sound all that rad as a teaching tool. I fully support mixing PyGame into a Python course, though.
Is the article seriously advocating increasing the number of women in CS by watering down their college courses? An easy "gold" course for the girls, a harder "black" course for boys (and the few girls who would have done CS absent any intervention)?
I also found this very annoying: "I have found an amazing correlation between women with math Ph.D.’s and women whose fathers are mathematicians...you wouldn’t be able to explain the statistics I have."
If you have this data, why not post it? Will mathbabe.org run out of bits?
No, it's advocating increasing the number of women in CS by proposing a separate 101 course for people (like, the article assumes, most women) with no prior programming experience, while retaining an "introduction" course for those who do have that. I didn't see any implication that higher-level courses should have separate tracks.
Indeed, and let's not forget the other half of her point, which was to have introductory level courses focus on solving examples from across science, rather than CS esoterics. From personal experience I agree that if you are trying to get someone interested in a tool, show them what the tool can do before showing them how the tool is made (or how to make their own).
This subject is aimed at students with little or no programming experience.
It aims to provide students with an understanding of the role computation
can play in solving problems. It also aims to help students, regardless of
their major, to feel justifiably confident of their ability to write small
programs that allow them to accomplish useful goals. The class will use the
Python programming language.
The implication seems to be that students with no experience can catch up with experienced students in one semester. Or perhaps that the experienced students have an advantage in perspective but not in actual knowledge, and the perspective can be taught in parallel with the knowledge at little cost.
These implications strike me as implausible. I think it would make more sense to create the for-true-beginners course with the expectation that the regular intro course comes after it.
No, it's advocating not telling people (girls and boys) that there are a few elites who are "good" at math and that the rest of them are doomed. It's not about changing the courses so much. It's about influencing the perception students have of how well they're doing in the class, which affects decisions they make about where to spend their energy.
Totally agree. And it isn't amazing at all that higher education parents breed higher education children.
I went to high school a town over from Lexington (Wellesley) and I will say everyone's parents are hi-end professionals or Ph.D.'s (actually It was a culture shock for me caused moved from Maine to MA when I started high school). High expectations bring high results.
> Is the article seriously advocating increasing the number of women in CS by watering down their college courses?
Yes, but not by this particular change. Black isn't harder, it just covers more fun side topics, by skipping some intro material. This is better for everyone--more prepared students don't get bored, less prepared students don't get intimidated.
Mudd's general new philosophy of acquiring female students at any cost, however...that's another thing. Not to mention people who should know better subsidizing "research" positions for freshman women to build glorified websites.
It is obviously not proposing 'watering down' anything. All it recommends is a recognition that people come to college with very different degrees of experience and that it can be intimidating and offputting to the less experienced. As there is a gender difference operating here best recognize it and address it at the curriculum level.
I have to agree completely: math and science just hasn't been cool and it's ostracized not just girls but I bet a hell of a lot of boys as well. I dropped out of my extracurricular science/computer classes as a teenager to maintain my social reputation, and I know I'm missing huge pieces of knowledge because of this.
What's changed? Well, computers are pretty cool now. The high school I mentor at has the hippest computer class I've ever seen, the teacher is an enthusiastic woman who drives the energy of the class and lets them focus on what interests them while directing them towards genuine learning towards what should be amazing careers for all of them. It's seriously encouraging and I have high hopes for a better gender balance in the years to come.
Processing (http://www.processing.org) is a good environment to teach programming in. It addresses most of the concerns about Java (removes boiler plate , further simplifies the language), while giving people access to all of Java as the class moves to more complex concepts. It also provides a fun, visual programming environment without the complexity of a complete modern 3D game engine or web browser.
Python is probably a better introduction language, but Processing is a close second.