I think it's a good thing. We still have regular programmers, and the word "engineer" actually means something.
Still, the level of competence is far from that of surgeons. It's possible to do a bachelor's with fairly little programming: at least where I studied, most courses include a large portion of teamwork, and I'm sure quite a few teammates passed their course thanks to me (or other people) doing all the work. When they graduate or enter internships, they just do the less technical work, like evaluating software packages, testing, or reading bids. I don't think you could find surgeons who can't operate like you can find engineers who have trouble with fizzbuzz.
Thanks for the confirmation.
Plus, it's always possible that you got your CS degree and aren't actually all that smart. I'm not a genius, but when I'd been programming for maybe a year, my ex-Googler friend asked me a Google interview question and said I did better than a PhD in CS that he'd recently interviewed.
So education and credentials don't always prove that you can solve problems or do everyday work in this field. Which is why I'm GLAD when an interviewer wants me to solve real problems in the interview. If I don't have to, chances are my coworkers didn't have to, either...
If my own background is anything to go by, CS is almost a textbook university subject: it presents a lot of theory and academic analysis, much of which certainly does underpin practical software development, but the emphasis is not on teaching specific applications, nor on teaching the supporting practical skills required to implement those applications. These things are experienced to some extent via examples that accompany the theory, and to a greater extent via reading around the subject and experimentation where the course has provided enough context to direct the student's personal study.
IMHO, one of the major differences between good and bad courses is that those that are taught well will also encourage the student to explore useful applications for what they are learning in this way, and thus to experience not only the practical implications of what they are learning but also the surrounding tools, processes and techniques that go with them in the real world.
Someone with a broad foundation in the theory and experience of varied practical applications should have no difficult adapting to other applications the same theory, nor to newer theories that have evolved through industrial R&D and not yet made it back into academia. Whether a recruitment process is any good at recognising this important ability to adapt and learn independently is an entirely different question, but any remotely useful process should be able to weed out the people who managed to spend 3-4 years studying theory without experimenting practically and reading around their subject, and avoiding hiring those is one of the most important basic goals of any software developer recruitment exercise IME.
Not unimaginable at all. In a world where surgery was run like software "engineering" [1], we would interview surgeons this way.
Why don't we interview surgeons this way?
(a) Because surgical training is via apprenticeship. For any given surgeon, you can ask "which group of other surgeons taught you everything you know? Which ones signed sworn statements that you are minimally competent?" Then you can find those senior surgeons and ask them candid questions about the candidate. This is always possible. If a surgeon turns up without such references they don't get as far as an interview. ("I taught myself surgery from blog posts, and I practiced on my cats?" Eeeek. NO HIRE.)
(b) Thanks to the long history of medicine and its very serious life-and-death implications, surgery is a very thoroughly attested event with a very definite chain of responsibility. Look at a hospital's records for a given surgery, and you will get a complete list of everyone in the room. If the surgeon was the only surgeon in the room, you can be fairly confident that they handled the operation themselves. And though I'm not sure there's an actual law that forbids a surgeon from handing the scalpel to a nurse, or an intern, and having them do everything, that would (a) not be something they could do in secret; everyone in the room would know and (b) the surgeon-of-record would still be the responsible party: If something went wrong in surgery guess who would be sued?
What I'm saying is: It is hard to bluff being a surgeon. Which is not to say it doesn't happen: For various all-too-human reasons, doctors and nurses do cover for each other, and scandals occur. But it's not like software, where the level of bluff is really, really large. Software "engineers" are often self-taught by necessity, especially in the tasks they actually do all day; their work has no legally defined standards or practices; there are so many ways to solve a given problem that two engineers with the same job title and official responsibility may have completely different skillsets; they work largely inside their own heads where nobody can see what is going on; they can very easily copy others' work, or sit in the corner while the rest of the team does the heavy lifting and then claim full credit.
---
[1] For this post, I'm going to use these scare quotes to distinguish "software engineers" like myself from actual Professional Engineers, who are like surgeons and are not, in fact, asked to build little bridges out of Lego bricks during their job interviews.