By the way Harvey Mudd seems to have solved the issue.
"With leadership from the college president and college-wide support, Harvey Mudd increased the percentage of women graduating from its computing program from 12 percent to approximately 40 percent in five years. This dramatic increase was accomplished through three major changes: revising the introductory computing course and splitting it into two levels divided by experience, providing research opportunities for undergraduates after their first year in college, and taking female students to the Grace Hopper Celebration of Women in Computing conference."
https://www.hmc.edu/about-hmc/2015/03/26/new-report-on-women...
You need your best professors and resources on the first 2-3 courses and everyone who comes out of those courses needs to "get it." Otherwise, everything afterward suffers and waters down. And without that stuff after (algorithms, operating systems, OOP, etc.) most students leave completely unprepared and barely understanding the stuff from those first few classes. And whoops, it's 4 years later.
Harvey Mudd is tiny. People always cite the percentages it has regarding gender balance, but the fact of the matter is that Mudd's incoming classes are smaller than 200 people and every single person applying has self-selected to go to an exclusively STEM college. This makes it very, very easy (relative to other colleges) for Mudd to basically pull a disproportionate number of already CS-inclined women from the pool of applicants with a bit of heavy marketing and effort. Mudd spends a large amount of resources trying to convince girls to attend, especially ones interested in CS, a personal interest of the president. Additionally, about 6 years ago they did something to revamp the admissions process which took the incoming classes from about a 2:1 male:female ratio to an even ratio (literally the incoming class for 2013 and every previous year was >60% male, and 2014 was magically ~50/50). This is remarkable for a school catering only to high-performing STEM students.
But it doesn't apply in a general sense. This model works for a tiny niche school in a big, big market. It will not work for every (or even many) schools in the market. And when you are pulling in classes where there are nearly as many CS-inclined women as men, it makes it a lot easier to have an environment that doesn't drive them away. But not every school can do this, it is literally impossible.
I have an incredibly smart friend who went to Oxford to do maths. His "competition" at Eton had sat the maths A-Level a year before normal, and had spent the last year studying university-level texts, and he was playing catch-up. Not sure who that benefits.
While I agree that people might have a good amount of experience of writing software before university/school, I'd be surprised if they were well versed in Computer Science (which I think is more focused on the mathematical/algorithmic side at first).
(I studied CompSci at Cambridge)
If students have extensive previous programming experience, they have had a chance to develop a mental model of computing that complete beginners simply don't have. They know how to debug. There are broad skills which aren't specific to a language or even a paradigm.
Putting all students in CS 101 together makes about as much sense as including Cantonese speakers in an introductory Mandarin course.
That said - they'll leave that course with vastly improved mental models while the other students will still only have a basic understanding.
But it's still a far cry from knowing nothing about programming in the first place. It definitely took my less time to learn and understand my first functional language than it took for me to learn my first language.
The biggest gap I observed was in motivating _why_ we were doing anything. The whole theme of the course is using abstractions to tame complexity. But beginners had never encountered real program complexity before, and it's not an easy thing to simulate in an introductory course as they're usually done.
Do you have a link to this? A quick google didn't seem to turn up anything that was obviously what you were referring to.
The good thing about scheme is that, after understanding the basics, it seems more natural to most students. I help one of my class study for his final exams and he understands recursion way better then for loops and iterators in Java. The syntax is also more minimalistic compared to Java which helps a lot (I did not think the syntax would be the hard part in Java). I don't even bother teaching him when to use private, public and protected. I just tell him to use protected for the variables (because it is often used in the final exams despite the fact that l variables should always be private in Java) and public for the methods
When I was graduating, they were just changing things around so that you had theory classes before programming introduction classes. I kinda did this by accident (took Discrete Maths/161 before I started 201 programming) and I think it's a really good idea. You get students use to the idea of sets, basic logical operators and the foundations they need before dropping them straight into programming.
I just tell myself it, over time, subconsciously changed my thinking to make me better at grasping software.
I make the same justifications :)
In a situation like that, I would be willing to assume that those who you thought were experienced were almost as lost in any event as the beginners; and no amount of language familiarity would help them.
But yeah, on the first day of school, children tend to puff up their egos and put on airs, and some people might've exuded a soon-to-be-deflated sense of overconfidence, when they weren't actually very experienced after all.
Both are possibilities. BTW, just as those supposedly inexperienced students found out, you know what happens when you assume... right?
You have to be okay with being "behind" compared to your peers, but if you'd like to be successful long-term in anything, you need to become okay with that anyway. Fashions change, whatever you were good at before becomes obsolete, and you relearn things. Slope is more important than Y-intercept.
That's nice :)
https://www.quora.com/What-are-the-most-profound-life-lesson...
The trick is being able to support those with decent slope while they're still under those who had higher y-intercept so they continue.
This is probably the biggest killer of startups: it doesn't feel like you're making any progress at all, until you have a sufficiently deep picture of your marketplace to know exactly where to hit it that your competitors aren't, and a sufficiently solid product that your customers realize this.
It really seemed like there was a vast chasm between the two groups. While I did see complete newbies making A's, it took them considerable effort. I'm not sure they really groked the concepts at that point in time.
It's sort of like deciding to be a music major when you've never played an instrument before. How do you know if you even enjoy it before starting down the long path to getting a degree in it?
That's highly debatable. A lot of people including my old compsci professor seem to believe it's largely something someone either can learn to understand or can't. I've noticed myself that students seemed to either "get" programming or not, both in a class I took in high school and the compsci class at my first college. This article on the issue is worth reading: http://blog.codinghorror.com/separating-programming-sheep-fr...
http://retractionwatch.com/2014/07/18/the-camel-doesnt-have-...
Yeah, but most people don't know that. Sure, us nerds figured it out, but most people are shocked that I learned to code at 13. I thought it wasn't that difficult. I don't think I'm smarter than most, I just wasn't scared off by the technical details.
These days you can be be learning to code using any number of slick web sites in about 5 minutes.
I was too focused on my high school academics (and goofing off) to devote enough time to properly learn to program. Once software became a part of my academics, I found that I got much more better explanations of how things fundamentally work. I also finally had more time to practice.
What you'll find with many computer science curricula is that the science of computing is so abstract that the simple excel macros and other stuff you've already done and perhaps dismissed is actually "programming."
No joke. Don't leave even one class not understanding what was covered.
The fact that so many high schoolers can pick up and become productive at programming is in part evidence that programming is not the most complicated thing in the world.
Everyone starts at the same place.
And nowadays you don't even have to wait to go to college to have your first brush with programming.
The best intro to programming online in my opinion is Harvard's CS50x on edx.org. It's meant to be an introduction to programming for people who've never even written a for-loop.
CS50x uses C, but there are plenty of good introductions with other languages. Google Think Python for a free pdf of an introductory programming book that teaches Python and how to think like a programmer. Following that up with Udacity's Intro to CS course is a great way to get your feet wet.
There are so many ways to learn and so many places to ask questions when you get stuck. And it's such an intellectually rewarding pursuit! Come join us!
It was rather disheartening.
In fact, outside of the introduction class, there was never actually a place where they sat down and said "here is where we teach you how to program". The professors mainly taught theory and had the assumption that we'd be able to put it into practice if needed.
Luckily we did have lab assignments, and multi-month group projects, so you would pick up some vestiges of professional programming within the 4 years you were there.
Heck, while I loved computers, I'd never been dedicated enough to teach myself a programming language. 4 years later I graduated with Highest Honors, and worked for the DOD.
I've learned to not ever say "this is easy" to anyone when I am teaching them something. That sounds perhaps like an encouragement -- like, "you are almost there" kind of thing. but it often ends up as an insult to their effort or intelligence.
To go back to your point, never stating the difficulty but focusing on describing the concepts and aiming at the student blind spots and/or communication issues (how many times an issue ends up in a simple "oh that's what you meant..")
I (unfortunately) did this as a T.A. in college. Despite being well versed in the material and caring quite a bit about trying to teach I used the "this is easy!" line as well as falling into many other teaching traps. I think part of the problem was that I never had a problem learning the material (I came into college with a fair amount of prior experience) so it was hard to relate when people had trouble. The other problem was without being formally taught about how to teach, it's hard to get feedback on what's helpful and what's counterproductive. For instance, it wasn't until I took a class with my girlfriend that I realized how much I interrupted or talked over other students when they asked questions. Without direct feedback I would probably still be doing it.
It's slightly paradoxical that most TA are people who had it easy with the subject, thus rarely able to explain things to people who struggles, thus not really teachers. I said this here somewhere before, this system crafts colleagues who will discuss a bit with you as pairs if you're on similar levels, not people able to explain. But maybe the school comb is balanced this way, only people able to grasp things on their own gets to advanced through it.
What book? I'm the same way (formal treatments usually aren't sufficient).
http://www.amazon.com/Linear-Algebra-Applications-Gareth-Wil...
When I asked the person why it was closed, I was informed it was too broad and not a 'good fit' for SO. Of course it was broad, beginners don't know what they don't know. I then looked up this persons history to the first questions that they had asked themselves. I was not at all surprised to see that they were all very similar types of questions to the one that was just closed.
I pointed this out and how their questions had gotten 50+ upvotes and multiple answers which then led them to help others and become a mod vs this poor person that will never bother using SO again. His response? SO has changed, it's not for beginners anymore.
A contributing issue might be that the moderators are the result of elections. This would weight moderator personality types towards people who crave recognition, perhaps even more on the narcissistic end of the scale. As opposed to more "level 5" type leaders who are probably less likely to run in an election and talk themselves up.
I spent a day or two trying to get karma on SO and then gave up. SO is still a good resource for some types of questions, but for the most part, it's not worth hanging out there to answer questions.
I realize that they are trying to "fight the people gaming the system." But I just felt like asking questions wasn't valued well enough. My participation dropped on SO like a rock after that.
I wrote an article about SO 4 years ago, and many of the points still remain and there are newer issues: https://theexceptioncatcher.com/blog/2012/09/stackoverflow-i...
My participation in SO now is primarily focused on asking questions only when I have something I've been struggling with for quite a while. By that time, I won't get an answer on the questions because it's so obscure.
The value of just the chance that someone will clarify things and suggest things that I wouldn't have known about is worth the inevitable online beat down.
I've deleted some of the questions afterwards, after receiving a lot of down votes or having it closed, but generally within 30 seconds there'll be a two line answer to a question that I spent several hours trying to figure out.
It's still an amazing resource, just a broken points/closure system.
And I think it's a fairly elitist attitude that's prevalent in programming that leads people to treat the field as if there is some innate quality that makes for a good programmer that we want to filter for. It's a similar fallacy that leads people to believe that they're "just not good at math," I think.
Once they have that under their belt, their knowledge/experience has been "bootstrapped" enough such that they know what they don't know, or where they need help, etc.
(An online course in what, exactly? Web design? C? JavaScript? Scratch? CS fundamentals? "Coding"?)
The SO approach is just nasty - unforgivably so, IMO.
SO could at least have split into abs. beginner/more experienced beginner/journeyman/professional/expert levels.
Currently it just seems to provide an excuse for not very interesting people to bully beginners instead of helping them.
SO does account for this though. Questions can (and regularly are) closed as 'already answered'. Those questions then link through. This means that all those questions hang around as pointers with all the ways of describing the issue people tried, hopefully increasing the search term surface area.
There exists a category of programmer who has so little knowledge that SO is not the place for them. They can be, should be, and are directed to Google and the documentation.
My point was that directing to Google or documentation is a basic misunderstanding (or simple forgetfulness) of what it means to be a beginner in the field. It is that telling someone to "Google it" or "read the docs" fails to understand the fact that beginners often simply can't comprehend documentation and can't use Google effectively because they do not have the foundation necessary to understand the problem they are facing.
This isn't an existential question of "what is a programmer." It's the statement that if SO is exclusive to "beginners" then it is doing the field a disservice. And it's a statement that directing beginners to Google or documentation is elitist and inherently exclusionary in practice.
Expecting people to be able to learn a language from nothing by going on SO and asking questions just isn't going to work well. It'll burn out answerers and fill the site with mostly useless content. I respect the decision made by the SO guys to focus the site and avoid that, because a useful resource for one group is better than a bad resource for everyone.
Over the past 5 years, I've probably increased my skills 10 fold by actively asking questions on there, its really sad that other people will not have this opportunity: http://stackoverflow.com/users/84131/chris-dutrow
SO is regulated because it means the content that is there is quality. Getting rid of stuff that doesn't work on SO makes it a better resource. Trying to shove that stuff back in would be terrible.
By all means, go set up that 'SO for newb questions', and when it's full of ananswered questions (because all the answerers get fed up trying to decode incomprehensible questions and feeding the help vampires), and content no one wants, maybe you'll get it.
Beyond all that, it's better for those asking too. Getting a short answer out of context isn't what those people need. They need proper, in-depth tutorials.
All the basic questions have pretty much been answered quite thoroughly already. Any beginner is extremely unlikely to ask something that someone else hasn't already thought of.
https://stackoverflow.com/search?q=%22NoneType+object+has+no... - ~900 cases of people spending time to upload their code to SO and waiting for answers with explanations, rather than trying to understand what the error means. (or even reading the first link that comes up in "Questions that may already have your answer")
As for copy-pasting error-messages into google, the results may not apply, in the asker's mind, to what they are doing. The results might have information which is too technical to understand. If the error comes from a turgid framework, it might very well provide no useful clue at all because its just detritus from leaky framework abstractions.
http://stackoverflow.com/questions/19671962/uncaught-error-i...
Luckily, I was met with a kind and thoughtful answer, and it turns out that I'm probably not the only one with the problem: the question has had some 42K views. But I've been coding for 20+ years, and I still screwed up my question. Yes, even on a Q&A site, folks should research before posting. But some grace to beginners who don't even know the proper etiquette would go a long ways.
Though Google does seem a little worse at programming queries now than it was circa 2009. :/
Seymour Cray once said he liked to hire inexperienced engineers, because they don't know what's supposed to be impossible.
Or they're new to programming and lack the vocabulary to research what they're interested in or phrase a question well.
Unfortunately the assholes are far more promiscuous with downvotes, closing and toxic-responses than the good people are with _any_ response. To make it worse assholes see it as their duty to "police" SO to maintain the quality of the forum and will be laughably smug about justifying their actions.
The sad thing is it doesn't have to be that way. In fact, the person with the most high score, Jon Skeet, is remarkably gentle and kind in all his interactions. It is a pity more folks don't see him as a role model to emulate.
The extra up-votes help with other people taking the question seriously and it actually getting good answers.
If you or anyone else wants to be included, shoot me an e-mail (its in my profile)
Definitely email if you are interested. And to everyone else reading this: its an open invitation, we'd love to meet some more hackers!
There is a middle ground. There is a need to get rid of low quality content or content that doesn't fit on SO - if you remember the time before SO, you remember the experience of wading through the piles of rubbish and nothing that was forums and resources for programmers of the past. The reason SO is different to those and better is because they know what type of content works on the site, and they intentionally limit the site.
The right answer is to try and help those people who post the wrong stuff in a nice way (point them at the right places, explain the issue), move on quickly, and answer good questions.
Indeed, I wish that behavior was more common on SO. Unfortunately, smug remarks, closure and downvotes are the norm. This might be understandable for questions which are clearly ill-conceived, but it happens to ernest, sincere questions as well.
It's worth noting closure isn't always a negative thing - questions closed as a duplicate give the asker an answer and work as pointers for others.
- 5th question: https://stackoverflow.com/questions/36416969/error-with-pip-...
- google: pip install "Operation not permitted"
- first link: https://stackoverflow.com/questions/33004708/osx-el-capitan-...
- second link: https://apple.stackexchange.com/questions/209572/how-to-use-...
(downvoted, marked to close as dupe)
Yes. People get downvoted into oblivion by asking questions which they (demonstrably) did not spend a moment researching. The front page of SO contains a steady stream of them anyway. The downvote arrow means (tooltip): "This question does not show any research effort; it is unclear or not useful".
But! If someone actually posted something that's not googlable and that they actually tried to understand, they do get good answers in my experience. If a good question in closed, you can always report it for review. I'm not saying it never happens - but I helped to reopen a few.
This is sort of the unspoken problem in computer science. Most people aren't doing it for the first time. It's not uncommon for many people to have started programming on their own in middle school or even earlier. First year computer science is more of a refresher.
Then there's everyone else starting an entry-level class as beginners and getting completely demolished. "How does everyone get this so much better and faster than me?" The answer is because they're not beginners.