5 ms·
I imagine this varies, but for me most of college was very much about the pure bits of computer science - how things tick, so to speak. But very little of your
by bpicolo 9y ago
I imagine this varies, but for me most of college was very much about the pure bits of computer science - how things tick, so to speak. But very little of your day to day at most enterprises is about writing new versions of data structures, or academic-level operating systems/database work (obviously there are some roles in the industry where this is the task, but it's not the majority). That's not to say learning it wasn't important - I'd argue that it's a crucial foundational aspect of being a very capable software engineer, it's just not the whole picture.
What learning those things does do is drastically increase your future flexibility as a developer - new databases, new languages, new jobs entirely, whatever. It's all built on the same primitives and if you have that fundamental understanding it makes it easy to ramp up on new technologies given you have the willpower and motivation. There's still a learning curve for specialized fields (of course) but that's fine.
Colleges may well be adapting since I left, but the main issue is that people aren't really holding you to the standards of software that exist at capable software firms. Correctness is about all that matters in university. Students don't know how to optimize for testability, maintainability, deployability, monitorability, etc etc. And learning and developing those skills makes you far better at the 'correctness' bit too.
There are some courses that are collaborative, but in industry the code you write can affect hundreds or thousands of other engineers and there can be real economic consequences of issues in your work (see: plenty of interns/new hires that have had the opportunity to kill $100,000-$1,000,000 or more in revenue by taking down a site - not blaming them, it's just an issue that actually exists in the real world). The order of magnitude is just so different.
This isn't a problem per se, I don't think universities should be expected to perfectly prepare you for this (this is why internships are crucial, and are one of the strongest interview signals for new grads). But somebody does have to - the onus is really on employers of new grads to raise functional engineers if they want to have top notch engineering teams.
I'll be honest - I didn't really grok CS until my first internship had passed, but that one summer really changed both my existing knowledge and my desire to build those skills further. I'm really grateful to have worked with some people that sparked that interest in me. I was at a 2-fulltime-dev startup with a ton of opportunity to work on different pieces of the stack, and it was just tremendously fun.
Side note: an interesting note on taking down applications is that as software enterprises get more mature and taking down a site is that much harder, it feels like (to me) that new engineers in your organization actually have a less opportunity to learn-by-doing for the foundational pieces. This is a very bizarre catch-22 that I suspect has real consequences for the growth of new engineers in software organizations. Very hard to calculate that effect though.
- ptero 9y agoI agree with what you say for both the college (goal, value) and much of the real work on actual systems. To me, this splits pretty neatly into "prototyping" vs "operations". Prototyping mindset allows you to try many things due to low cost of failure; operations does not. Most software work and in general engineering work outside of academia has at least a touch of ops flavor (i.e., high cost of errors), so to be successful (e.g., not to be labeled a loose cannon) one must be able to impose self discipline required for this. But most organizations have systems and environments for prototyping (or will gladly set up one if you clearly express your wishes and articulate some benefits). Engineer that wants to work on new ideas then must learn to wear multiple hats (prototyping vs ops) and switch them as needed: the moment one glues one on he is limiting himself to either a rigid ops work (no, we cannot try new things) or a junior level dev (he cannot touch the real systems; his code does not work well enough).
- bpicolo 9y agoI agree that that's one really interesting way to consider the split, good call on that! Being able to experience those multiple hats is definitely important.
- dhimes 9y agoThank you for your thorough reply. Colleges are under fire, even from their own ranks, by people whom I believe confuse training with education, but it's clear that's not your issue. It sounds to me like there needs to be some sort of deep-dive "onboarding" program where new hires can work on a curriculum of projects and learn the SOPs of the organization. Colleges could take some of it on, of course (testability and maintainability for example), but one of the complaints I heard even ten years ago is that they can't keep up with the changes in the field. No true fundamental best-practice principles have evolved, it's largely company-dependant.
- bpicolo 9y agoIt would be interesting to see a course than spanned multiple years building upon the same project with the same team, emphasizing different aspects year-by-year, but I imagine that would be pretty nightmarish on the scheduling front.