Slashdot Mirror
Can a New Type of School Churn Out Developers Faster? (dice.com)
Nerval's Lobster writes: Demand for software engineering talent has become so acute, some denizens of Silicon Valley have contributed to a venture fund that promises to turn out qualified software engineers in two years rather than the typical four-year university program. Based in San Francisco, Holberton School was founded by tech-industry veterans from Apple, Docker and LinkedIn, making use of $2 million in seed funding provided by Trinity Ventures to create a hands-on alternative to training software engineers that relies on a project-oriented and peer-learning model originally developed in Europe. But for every person who argues that developers don't need a formal degree from an established institution in order to embark on a successful career, just as many people seem to insist that a lack of a degree is an impediment not only to learning the fundamentals, but locking down enough decent jobs over time to form a career. (People in the latter category like to point out that many companies insist on a four-year degree.) Still others argue that lack of a degree is less of an issue when the economy is good, but that those without one find themselves at a disadvantage when the aforementioned economy is in a downturn. Is any one group right, or, like so many things in life, is the answer somewhere in-between?
Or is that community college?
Given who is starting up the school, I bet this is being driven by companies who want to pay their employees less because they only have a 2 year degree.
It will not churn out developers faster (i.e. people who develop things from ground up, Starting with basic application and adding features). And it will not churn out Software Engineers (i.e. people who engineer the solution from top-down using abstraction). It may churn out copy-pasterino-code-monkeys who copy paste from stackoverflow, and complain if it doesn't work.
You can't churn out developers like automobiles.
I began programming casually in elementary school on Commodore Pets. I started programming on my own computer in fifth grade on a Commodore 64. Afterwards, I had plenty of short work stints during junior high school, high school, and my 7 years at the university, but I didn't begin programming full time for more than an 8 month period until I was 24. Even then, I was still very green.
The best developers have been at it for 10-20 years at a minimum, and I'd even go as far as to say I prefer programmers who've been at it for 30 years.
What I don't care about is your physical age. If you started programming at five years old, and you kept at it continuously until age 25 then you'd meet my criteria.
Developers are created over many years, they've worked on many generations of technology, and they've proved flexible with time. Many of the good ones have been at it since childhood, but I don't think that should disqualify anyone.
That's why developers need to get paid so much. Training over a decade to achieve basic competence at something is expensive. Many have a very expensive university education they have to repay. For me, I had to forgo my social life pretty significantly from age 15-25, and I'll never get that time back. The only way I can be repaid for that is with money.
If you're trying to shortcut the process somehow by picking up someone who knows nothing about creating software, hope to train him or her in a few years, and expect to pay him or her poorly then you're going to produce some pretty awful software.
Correction. Anything you can learn about software engineering you can learn without going to school in the first place and the theory is best learned and reviewed and re-learned organically alongside practical experience.
Take two software engineers and set them side-by-side. One with a four year degree and one with four years of self-study/work experience. Ask them to devise, implement, deploy, and test a solution to a real problem you are having and don't yet know the answer to. That four year student will be lost. They've never learned that in the real world nobody else knows how to do their job, nobody provides you all the information or the tools necessary to solve the problem like in a lab or even knows what that would be. In the classroom your problems are presented in a progression that implies what you've studied recently is what will be required to answer them. In fact, in the classroom solving a problem without using what was just taught (and thereby demonstrating you've learned it) will often penalized. There are no such hints or clues in the real world. The self-study engineer will immediately set out figuring out what he's going to need and how to go about finding and getting it just like he has done with every challenge for the last four years.
That said I think going to a university AFTER 4-8 years of self-study and experience would be a very valuable experience. By that point you have a context and mental framework to put all that organized and spoon fed material into and you'd get a lot more out of it.
You're not comparing apples with apples - or more directly, you're not comparing people with the same level of experience. The self-study programmer already has up to 4 years real/world experience, depending how they did their self-study; the graduate has no commercial experience so of course they're going to flounder initially (in comparison at least). Give them both six months from that starting date and you might be surprised at how quickly things have evened up, and given another six months it'll be all the other way.
Although it's likely that if you ask both of these developers to develop an efficient algorithm/data structure to do something novel, the one with the traditional four year degree is more likely to come up with a better solution -- and that will likely remain true for the remainder of their careers. The four year degree developer will likely be "caught up" with the self-taught one (given the same base intellectual capabilities of course) within two years and then always be ahead.
There are, of course, exceptions.
Why is there an "insightful" mod and why isn't it "-1"? If I wanted insight, I wouldn't be reading
But not for that reason. Profits for some, fuck everyone else. That is the current mindset with too many people holding power. Nepotism, cronyism, and quid pro quot is the overwhelming number of rich people today. Oh I know, there has always been some of that but we used to teach morality. Morality is one of those things omitted in current schools, and you'll have to give less than that to try and expedite programmers. Here is the test: Ask a person today "If you are rich, how much money is too much money?" 30 years ago most would put the number in the couple million mark. Today, most people will laugh and tell you know such thing. So we have gone very far backwards in morality as a society, in a very short amount of time.
Could a school turn out "programmers" in 2 years? Sure, they will know enough to do some "programming" but not how to solve problems, and won't be able to communicate with people. Further, they will be ignorant to history so not know what to look out for in actions by the powerful which makes a large group of people fodder.
I heard something similar the other day, where 100 years ago people from Universities were well versed in every subject. They studied Math, Music, Chemistry, Languages, Art, Philosophy, and History. A person with a degree was very high valued. That was supposed to be the goal of Public Education and Government funding and control in Universities. And look where we have gone. Specialized degrees like "Sports Marketing" with little to no other knowledge to fall back on.
-The wise argue that there are few absolutes, the fool argues that there are no probabilities.
That's a good thing. We don't need more useless code monkeys who think making shiny webapps in CSS+JS+HTML is computer science. You don't need calculus to be a programmer, but you'll probably be a shitty programmer if it was too hard for you. Is being a well rounded human being really too much to ask? If you want to learn a trade, be a plumber. Anything having to do with science or math is not served by people who think learning anything is a waste of time.
Side note: this article presupposes that pumping out more developers is a good thing. I'm not convinced it is. Quality over quantity.