A History of Capacity Challenges in Computer Science [pdf] | Hacker News Reader

A History of Capacity Challenges in Computer Science [pdf]

cs.stanford.edu

66 pointsluu5 comments

I hope this faculty shortage is still in effect when I finish my Ph.D. :)

Anecdotally, many of my peers are going to industry instead of academia, because 1) better pay, 2) no begging for funding, 3) they don't want to teach. Academia is seen as signing up for a difficult life. Many industry research labs are allowing their staff to publish papers, so you can even participate in the academic research community. Look at how many NIPS papers come from Google-owned companies.

Personally, I'm interested in academia mainly because I'd rather not be part of an institution whose main goal is making money - but that attitude is not common in CS.

How sure are you that a university's main goal isn't to make money these days?

Based on the evidence it seems that is very much the point of universities these days, and research only exists to secure funding and prestige, not to push the boundaries of knowledge.

This is not a happy situation, but it seems to be where we are. And it probably won't change for a long time.

It's not impossible to disintermediate teaching and accreditation. But I can't think of any workable way to disintermediate high quality research, especially when significant resources are required.

The best approach might be to herd postdocs into research-only facilities - like Bell Labs and Parc, but independent - and throw money at them to see what happens.

That's not very imaginative, but research-only labs seem to work well when they're managed correctly, and a lot of smart people would like the chance to do pure research without having to teach, mark, and justify their existence with constant grant chasing and committee meetings.

Having watched a few rounds of CS faculty searches, I'm certain the shortage will still be there. (At least at my university.) The reasons are more or less what you mention, but objectively, it really isn't a particularly difficult life and teaching can be fun - especially when you hire a student to grade.

We haven't really had anything as collapsing as the "dot-com" crash occur in the past decade (knock on wood). As the article points out we are at all time highs, Congressional Research Service [1] reports that computer science jobs are projected to continue for the next decade.

I also like to mention [2] because it seems like this is where female enrollment's number dropped and never recovered. For a short while, I was researching about how these numbers came to be. Unfortunately, the majority of articles only talk about how computers were marketed to males and "Revenge of the Nerds" stereotypes pushed females away.

Instead, Dr. Roberts' article looks at a simple realism - everyone got sick and tired of CS. Sure, it was lack of resources in the 80s, but it shows a different perspective than marketing did it. I cannot find any research on CS perceptions stating it, but there was/is a severe "elitist" attitude among CS students. I think this is where a lot of the unattractiveness of the subject comes as well. In the 80s, you had to be the best of the best to get into these limited programs. I think it inflated a lot of student egos, which is where that personality quirk was created.

Maybe all STEM people have it, I don't know, but I do see it in CS. It's getting better as other programs are starting to teach things like coding, but if we continue to have limited resources, something needs to help relieve the strain. Better MOOC/Cyberlearning retention, grade automation, novel teaching approaches, etc. Otherwise, we'll just be repeating the 80s drop soon.

[1] - https://fas.org/sgp/crs/misc/R43061.pdf

[2] - https://www.npr.org/sections/money/2014/10/21/357629765/when...

Reply on news.ycombinator.com