So the solution to increased cost of education is apprenticeship? Is this part of the current trend at bashing higher education to make sure that we become obedient but efficient drones?
There is a lot of things broken in higher education, but saying that hiring a CS/SE graduate for a developer position is like hiring a theoretical physicist to repair a car is disingenuous at best.
Guess what, graduate students who write a compiler as part of their course, who are working on a vm for matlab, who are improving IDE auto-complete based on all sorts of algorithms, who are devising a new distributed merge algorithm and evaluating its performance through hardcore network simulation, well, they know how to program! As a bonus, they know how to apply the scientific method and be rigorous when they report a result or an improvement. They have been exposed to all sorts of things that an undergrad don't even suspect their existence.
Sure, some grad students aren't good. Sure, people who don't go to college/grad studies can end up being way better and knowing more than grad students, but don't discredit a degree because you believe that it's too theoretical. Just ask about the homeworks, the projects, and the thesis the grad student worked on.
"Is this part of the current trend at bashing higher education to make sure that we become obedient but efficient drones?"
No, the goal is to remove vocational training from higher education and return it to its supposed goal.
The graduate student that you describe is rare. A masters degree in CS typically means that the person took more random computer courses and knows no more about software development. What you fail to realize is that nearly anyone who wants a masters in CS can shop around and find a school that is willing to accept them.
First, I believe there is a cultural difference between Canada and the US. In Canada, a Master's degree is typically not a professional degree and you usually cannot buy a M.Sc. Typically, half of the credits come from courses and the other half (often even more) comes from your thesis. I don't know about the situation in the UK, the author's country of origin.
Second, I TAed and taught programming, algorithms, and software architecture courses for undergrads, Master's and Ph.D. students so I'm well aware of the advantages and limitations of higher education. I saw students in an advanced architecture course who did not know what a thread was or who had never written a single SQL query. Well, they learned it in my course.
The graduate students that I described aren't rare. At least 75% of the Master's and Ph.D. students in the SE lab and PL lab at my university match that description. Maybe they would have a hard time building a web application in a day, but I think they have demonstrated that they can learn pretty complicated things and that they will learn how to solve your particular technical problems.
I felt the article was really about "degree snobbery", meaning that the author promoted snubbing people with degrees. I understand the frustration of people without degree who need to prove every time that they don't need one. But I don't believe that having a systematic negative bias on candidates with a degree is wise either. Honestly, does it make sense as the author says that someone with a bachelor in C.S. don't know how to implement a binary search (see [1] for a possible explanation)?
Regarding the vocational training that needs to be removed from higher education, I believe a compromise is needed. I agree that you cannot efficiently learn development process and all the latest languages and frameworks at school. But you need to learn some good programming skills and software engineering practices, otherwise, it is a lot harder to understand and play with more complex concepts and it is also more difficult to bring a significant contribution if you do a Master or Ph.D. later.
There is a lot of things broken in higher education, but saying that hiring a CS/SE graduate for a developer position is like hiring a theoretical physicist to repair a car is disingenuous at best.
Guess what, graduate students who write a compiler as part of their course, who are working on a vm for matlab, who are improving IDE auto-complete based on all sorts of algorithms, who are devising a new distributed merge algorithm and evaluating its performance through hardcore network simulation, well, they know how to program! As a bonus, they know how to apply the scientific method and be rigorous when they report a result or an improvement. They have been exposed to all sorts of things that an undergrad don't even suspect their existence.
Sure, some grad students aren't good. Sure, people who don't go to college/grad studies can end up being way better and knowing more than grad students, but don't discredit a degree because you believe that it's too theoretical. Just ask about the homeworks, the projects, and the thesis the grad student worked on.