Slashdot Mirror
How St. Louis Is Bootstrapping Hundreds of Programmers
itwbennett writes "The MOOC (massive open online course) failure rate is notoriously high — only 1% of people who take the beginning computer science programming class, CS50, that Harvard offers over the EdX online platform complete it. A new effort in St. Louis called LaunchCode is changing that — and solving the city's programmer shortage. For the past several weeks, about 300 hardy souls have been gathering in a downtown St. Louis library to listen to the CS50 lectures and work together on the various programming problem sets. But the support offered by the all-volunteer run LaunchCode doesn't end with meet space. They're also doing an end-around on the traditional coder hiring process by pairing the students who complete the course with experienced programmers in one of more than a 100 tech companies who are looking for talent."
If you're going to co-op cyberpunk terms, at least get them right. It's Meat Space.
I've wondered why more online educational institutions don't try something this, real groups that meet somewhere public to work through a course together.
The aspect of being paired with a working programmer eventually is also a great advantage, but just having a group to work with would lead lot more people to have enough motivation to complete a class.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
David Malan, who went to Harvard himself and is a rockstar teacher, teaches the course. I watched a couple of his lectures and found them interesting and engaging, even when he covers some basic concepts that I have long known. If I had him teaching me programming back in the day, I might have stuck with it and become a coder myself.
i'm sure its just me, but isn't this possibly the dumbest excuse for not becoming a programmer around?
almost all programmers i know who really add value to projects learned the stuff mostly on their own...teachers don't teach this stuff, the computer does. for the first six months almost everyone who is trying to write a program is going to be pounding their head on the desk.
only through that struggle will you begin to grok it.
i still thank my first Comp-Sci undergraduate teacher (FORTRAN for those interested) for issuing this offer to his students...
"anyone interested in getting an A and skipping having to come to class, if you write a bowling league manager that does this, this, and that and have it done in 10 weeks, talk to me after class"
I believe i was the only one who took him up on his offer, and to this day i'm thankful for him for the things i "learned" about PROFESSIONAL programming.
never bring a twinkie to a food fight.
I'm one of those people who dropped it. Namely, because my IT classes (I was getting college credit for) picked up. I wouldn't discount a 1% completion rate as a sign of failure, or even one of difficulty. Hell, I'd go so far as to say that every person who signs up for it for any sort of personal growth is a success, even if most drop it later on.
The plea began to hammer progressively louder upon the desk of the Unification Board, from all parts of a country ravaged by unemployment, and neither the pleaders nor the Board dared to add the dangerous words which the cry was implying: "Give us men of ability!"
Why is it so hard to find talented people?
I want to delete my account but Slashdot doesn't allow it.
The point of MOOCs is that since they're free, those who enroll in them can pick and choose from what's there that interests them. Plenty of people enroll in a MOOC because they want a refresher on something, or to learn about just one aspect of what's covered, or just to see what it looks like. It's not failure when those people don't go through everything in the course.
Space game using normal deck of cards: http://BattleCards.org
aka cease fire with the endless decepteyecon hypenosys of pretend dogooder foibles all in beta forever
Where is this shortage or programmers problem coming from? Last I check there are lots and lots of them. If they are looking for good programmers, they wont solve it by offering one course...
did you forget to take your meds?
I did not complete edx cs 50.
The problem was time. The course is broken into weekly segments. If all segments were open at the beginning i would have sailed through.
When I signed up I had plenty of time to complete the assignments. 3 weeks later none at all. The problem was that the course didn't have the option for me to get ahead far enough so I could continue the lectures in sync.
I still did the course work but did not bother with doing the tests out of order to complete the course.
There has always been and will always be a shortage of good programmers. It's the way the art is .
I took Jennifer Widom's SQL course out of Stanford a couple years ago, just as a refresher (and to see if I could "hang" in a world class instutition). I found the class rewarding.
At its peak we had 120k students. Now consider 1% of 120,000 is still 1200 students; far more than she could teach in a year at a school like Stanford.
Yeah with MOOCs, like everything else accedemic, you get out of it what you put in. At least in these cases, they let us, the prospective student decide if we should be there, instead of weeding out students through the admissions process or with heavy prerequisites and other selective measures.
Just like real college, many will fail and few will succeed. At least this way, my outcome is all up to me.
These efforts aren't solving the programmer shortage, they are simply mills churning out unqualified candidates (only ~1% of which will get a job and %1 of those becoming a solid developer) in order to deflate wages for everyone else.
There is another program that is ramping up called CodeRed, which helps high-schools introduce a series of courses that will supposedly get high-school graduates entry level jobs from $45-60K.
I'm not too worried as ITT / Pheonix / have tried to do this for years with little success (and several lawsuits for promising things they cannot deliver). You'll get the same result out of these programs.
As an aside, I just wish the developer community had the political awareness to see these things for what they really are. Maybe it's industry maturity or the aggregate political / sociological leanings, but you don't see this kind of crap from Doctors, Lawyers, etc.
I also wish we didn't devalue education by stating this is all it takes, but, hey call that the Holiday Inn effect.
if there's a shortage of programmers in St. Louis, does that mean there is a surplus of businesses ready to hire programmers in St. Louis?
(I'm happy where I'm at now, but this would have been good to know two months ago.)
Coding is hard for people that are able to understand it. It's impossible for most perple to learn enough to be proficient with it. No class teaches it, it is learned by doing. If you're fortunate, you might have a mentor to guide you through the diffuicult to referance parts. The only education that is helpfull is a degree in -or- self-taught knowledge of Electronic Engineering, period! Coding is a language art, not a science or engineering skill. Some exceptions apply of course (read this sentence)! Basic math and logic understanding is plenty for 99% of jobs. The diffucult part is the complexity in keeping the scope of projects in check... Lack of documentation and the complexity of the tools; a mostly nessassary complexity! And the fact that there is NO "RIGHT" WAY TO DO IT! Basicly, if you want to be a programmer and can't teach youself, then it isn't for you. ; (
The shortage of good coders scales directly with the intelligence bell-curve, sadly. Nuff said!
Most of the tech companies in the area treat programmers/developers (and IT as a whole) as a fossil fuel, to be immediately burned for their energy and quickly forgotten. Attitudes are slowly changing and quality of life is improving at a glacial pace. Still, it's a hard market to thrive in-- long hours, pay that is commonly bottom 25% of national medians, and special types of business people that can only be the result of inbreeding. Expect to be worked like a rented mule, especially in the health care sector.
STL does have its gems (Enterprise RAC, Savvis, Panera, MasterCard etc.), but they're pretty difficult to get in to with all of the competition.
Essentially doubled the number of people who had ever passed that course- versus 50 years of the slow way.
Having a ton of fun trolling today, are we?
By that argument, the onus is on the companies, rather than the individuals. As has been discussed, there is a massive supply of talent willing to work, but the companies that refuse to compete in the market are having a hard time. There are ways for companies in areas with low costs of living to compete -- they just don't want to. You said it yourself: you want experienced talent, but you aren't willing to pay for it. People can complain about a "bad economy" or "labor shortages" until the cows come home, but cases like this are entirely self-inflicted. Hiring and retaining talent is a cost of doing business, and the talent itself is an asset, not a liability -- only by tackling those issues head-on can real progress be made. Sure, it's going to cost you a little more on a per-person basis, but you'll save money in the long run by reducing training costs with less-frequent turnover/hiring, increasing your overall production quality (and thereby reducing time spent debugging/refactoring/etc), and increasing internal morale (happy people are motivated to stay that way; dissatisfied people can get stuck in a rut). Even small companies can do this -- it's just a matter of prioritizing your goals (and budget) and keeping the big picture in mind.
IT / tech needs apprenticeships and CS is not = IT.
Both IT tech work and programming some kind of trades / apprenticeship system.
The older college system is to much of a one size fits all and at times can be theory loaded / has lot's of skill gaps.
Some of the theory is nice to have but others is only really useful for very low level OS stuff that most programmers witting code should have to deal with much less wire there own systems bypassing the build in os ones.
Also with IT / desktop / sever / networking is more hands on and the over load of theory is bad as well doing stuff out a book without being in real settings that can be quite a bit off of what the book says.
I took an edX course as part of a mandatory computer engineering course at SJSU and it was terrible. The entire thing was like a marketing ploy designed to boost the reputation of the instructor that spearheaded it. In the end, we had to have many extra sessions of traditional lecture to get the kind of real learning you can only get in a classroom, because the videos and online components were worthless.
I love technology, but the intersection of education and technology has always been forced with a very heavy hand, and it seldom, if ever, works. There's a reason the traditional form of classroom learning has persisted since ancient times -- it just works. Virtually all failures in modern education are more about misappropriation of funds or corruption of policy by administration rather than teaching or the teaching methods. And guess what, a lot of these poorly utilized funds get diverted into needless technology for the classroom. Video conferencing anyone?
edX doesn't work. Maybe the intentions behind it were legitimate, but all it is now is a platform to prop up the standing of professors who are more interested in seeing their name in the newspaper (literally, he spent an entire lecture talking about the newspapers he was featured in) than teaching anyone anything.
You could go further with this idea. Maybe have an expert in the topic present to help people. You could even gather a bunch of meetups for different courses under one organization. Provide equipment, develop new courses, etc. You could call it I don't know a college maybe?
I'm a contactor here in St. Louis.
There is no shortage....
especially when you consider the big companies that
take college business grads, give them 2 weeks of Java
and pawn them off as "programmers" paying them about 40k/yr
while billing 120+ / hr...
They're not good but the stoooopid (intentionally mispelled) companies don't want to know that.
It might interfere with kickbacks.
Manage your expectations. If the MOOC is truly massive, one percent is good.
If you have one class of 30 students and 2/3rds actually graduate, you get 20 graduates.
If you have 3000 students in a MOOC and "only" 1% completes, you get 30 graduates.
If you can't understand that difference in percentages, it's not worth talking about cost-effectiveness.
But having an experienced mentor is definitely an improvement for any kind of training.
I thought the point of it being an online class was so no one would be prohibited due to location or time. If I have to meet at a set location doesn't this defeat the purpose of holding it online?
Why are these folks bothering. They will get hired by some company and that company will outsource the development work to India or some other country that they can pay the folks a dime on the dollar to do the same work. Why is there a shortage? So many companies sucking up the H1-B visas and outsourcing. Get a grip....
The Truth is a Virus!!!
Given the low entry barrier as compared to traditional higher education systems, the surprise isn't the failure rate, but the success rate. Given the low cost per student of providing the course, even at a 1% success rate I expect that the cost per successful student is much better than the traditional systems, though I don't actually have numbers to back that up.
I am gratified to hear you are willing to hire midlife people who are tired of the rat race. There is something to be said for programmers who understand how to understand your problem, figure out a solution in the language of your choice (and learn it if necessary), then explain what they are going to do and how they are going to do it. You will seldom get programmer/analysts from a quickie course in CS, and generally people need about 10 years in practice to have any idea what I am talking about. You should not be trying to compete with Silicon Valley for the cream of the young programmers. Even if you could afford them, and you can't, they would not be happy with you. The country is full of unemployed middle-aged and older programmers. You have to be willing to pay them a bit more than entry level, but of course there is value in these people.
I looked for programming jobs in St. Louis exclusively, originally. I work elsewhere but still want to move back. I don't get this hatred of the midwest--what's wrong with having a good job, a big backyard, cheap attractions and the amenities of any big city? I mean, what are you giving up, weekly operas at the Met? Man, I want to move back.
Something around $200k/yr for entry level programmers sounds about right
Are you insane or merely trying to stir the pot? An entry-level computer programmer hired as an employee should be earning 45K per year. By the time the programmer reaches 5 years experience their salary should be 50K-75K+ depending upon their skills, aptitude, and work history. After 5 years their salary is only limited by their specialisation, their preferred geographic region, and market rates. These figures are for normal cost of living areas which excludes Boston, San Francisco, Silicon Valley, New York City. St. Louis, Missouri, qualifies as a city with normal cost of living. The most useless job advertisements post a laundry list of products and languages yet want to pay barely above minimum wage.
Coders code because there is something about it that they love. I am not sure if it is passion/curiosity/interest or what. But when we have a free moment after school or after work or after the kids go to bed, we think of something to code or we work on something we want to code.
I believe that anyone can learn to write code in 6 months. But they can't become a quality Software Developer in that time. They aren't going to become experts in 5 years like those who have passion for it.
Why won't they become experts very fast? Remember the 10,000 hour rule. You are expert when you have put 10,000 hours on something. That is 5 years right?
Wrong!
Sure a job gives you 2000 hours a year, but since many coders only code for only part of their job due to meetings, interruptions, etc.. coders usually only get 1200 to 1500 hours of coding time a year at work. So it is going to take between six and nine years for them to become experts.
However, many of us with passion code at night and on weekends. We work longer hours at work. I know people who code for 16 hours straight some days. They are coding more like 2500 to 3000 hours a year. So they reach their 1000 hours in 3-1/3 to 4 years.
I really wonder if programmer shortages really exist or if it just a ploy by employers to undercut the worth of people who are already writing code? This is quite distinct from the facile discussion about "coder" vs "computer scientist" or "designer" and all the complexity of skills needed. Clearly there is a big difference from writing some static language with few abstractions, even coding HTML, CSS, and Javascript, and Haskal or Python. It may be that maintaining legacy code such as FORTRAN and COBOL is real demand that is in fact hard to satisfy, and that there are too many people trained in newer more powerful environments. I just don't know, but I am somewhat skeptical of claims of shortages where the range of need is so complex. It sounds like the real problem has always been matching skills a certain person has with a need out there and that the average recruiter, even the average technical recruiter out there, is not very good at making the matches.
Even more problematic is this tendency to believe that in economically disadvantaged places like Oakland Ca, or St. Louis Mo. that teaching inner-city kids how to "program" is going to help but a very few of them. You may find people who are able to thrive as developers at random in any population, but the number will be small in any given collection of people. Teaching large numbers of people the basics, and especially if the language chosen is strongly typed, like Java, is just not going to get very far for most. Just because software development is glamorous doesn't mean everyone should do it, or even try. In my experience it requires a special set of skills and attitudes that in fact few people have.
I think that basic language literacy skills, very possibly using a computer, are more important for disadvantaged youth than programming skills, or that programming should be used as a tool in pursuit of another interest. So that if people can find self-expression in imagery, or graphic arts, or writing, they these come first and that programming be viewed as a tool that might aid that pursuit.
Finally, it must be said again that opportunity is based not on the needs people have in a Capitalist economy, but in the recognition by investors that funding a need of people is worthwhile. Since investment has run askew because of financialization and international investment, there is no one to one mapping of need with resources. There is some mapping but it hingers on the wisdom of investors, which is something that reasonable people can question.
Springfield is ok. St. Louis and KC should be avoided if possible. Cheap Google Fiber wouldn't even make me want to go there.