I TA'd for a prof that felt that even in exams students should learn something. So he'd often have a question that could be solved by applying concepts learned in a way that hadn't been covered in class. Or by explaining some new concept that when combined with material in class - should allow the student to solve the exam question.

It was really interesting to see students' reactions. Some people loved it. Some people bitched and moaned about how incredibly unfair it was.

I was sort of in the middle, but felt it was completely acceptable if it was worth something like 10-20% of the exam mark.

The thing is once you're working in software for a living, doing the routine stuff is easy. But you get paid to figure out the problems that you haven't encountered before. In a class or a bootcamp it's probably a good idea to have those types of issues pop up from time to time. If a student can't handle that kind of thing in a class, they are going to struggle with the unknowns in the workplace.

I think I've worked with some of those who would have been in the "***** and moan" category...And they're still like that in the workplace. I was in the "loved it" category. It is absolutely legitimate to throw this type of stuff at students, especially in a programming class, for the reasons stated in the post above this one.

So apparently the homework was mislabeled and was supposed to be a "challenge" that doesn't require completion, just an attempt.

I've been going over my solution with instructors and asking questions and clarifying things, they seem pumped I was able to get it done! Said it never gets completed, are kinda surprised I actually did it.

Probably going to start doing code wars, etc. because that was the most fun I have had in a while.

The flip side of this is that the exam is an artificial thing that is meant to try and judge how well you learned the material in the course. It's a different beast than even an assignment (different time constraints, resources available, etc.).

You can argue an exam shouldn't be used at all - and I see both sides of that too.

This is awesome to hear, it's a great thing for both student/teacher if you have that sort of experience together

thx.

Good stuff, man. It sounded like a possible NP problem from your description, but not sure.

I have some sympathy for the students here. Figuring out problems you haven't encountered before isn't something you solve with your superior brainpower, it's something you've learned how to do over time by applying strategies and processes you've learned and employed in the past. Someone who's claiming they'll teach you how to be a programmer isn't holding up their end of the bargain if they aren't doing some work to teach you how these processes work; being like "here's 'Hello, World!', now let's throw you in the deep end" isn't going to help anyone.

In this case, the student mentioned having to use indexOf() which they hadn't seen before; has the professor at least introduced them to the MDN website and suggested there are other array methods they may need to discover themselves to solve the problems they'll be assigned? This **** is obvious to us because we've done it before, but it's probably not so much to students on day 3 of a coding bootcamp.

And to be a little results-oriented, given that it wasn't supposed to be a homework assignment and was mislabeled, sounds even more like you guys should have cut them some slack

When I did my 2-week training session of our Infosys team in India, I threw in a couple questions on the final exam, designed to make them think, like "Which of these is the best example of asynchronous communication? (an email, a phone call, an in-person conversation, an IM conversation)".

Totally blew their minds. They were like "When did we cover that?" I get the feeling their entire schooling has been nothing but rote memorization. Which is why they do pretty well with angular (lots of memorizing things). But I have a feeling they would struggle with react - which requires more deep JS understanding.

Of course it's not a black&white thing. These unknown solutions shouldn't be impossible to figure out when they are presented. Also googling stackoverflow should probably be presented on day 1. Spending time going over debug techniques/strategies would be good too.

I'm not necessarily talking about this specific problem in this particular boot camp, but in general I think part of programming education should involve a couple of curve balls thrown in to the problem sets. A big part of the job is figuring stuff out that hasn't been encountered before.

I'm old... Having access to stackoverflow when I was going through my programming classes would have been amazing. Even in my first job, I remember the first really tough issue I encountered would have been something that was easily figured out from a 10 minute search of stackoverflow. I think I spent about a week trying various things going through mountains of books on Programming Windows. That website creates a ton of productivity gains.

I bet its interesting to see people's perspectives based on age.

All of my serious programming education took place when I had access to the internet but before StackOverflow. You could search for problems but it was hard to find useful stuff. Lots of stale answers. Bad answers. Impossible-to-use UI. etc.

I remember when StackOverflow came out. It was just so much better than everything else.

Same here. Also, there was no real expectation of homogenous environments - different unix systems had widly different versions of compilers, shells, libraries etc and often there was little you could do about it. So many times you'd find something that looked like it would solve your problem but it just wouldn't work in your specific environment. Owning your own unix-y machine was still relatively rare so you just got whatever you got.

Finding a response directly related to your search was rare. Now I can just paste error messages into google and 9/10 times I find 20 links with the exact same problem.

How similar are the questions on exercsim.io and the stuff in Cracking the Coding book?

Any vets have advice on getting integrated/comfortable with new code bases, ie new jobs? How the hell do senior ppl jump in at the senior level anyways?!

I think part of it is that you get used to recognizing patterns, and you become familiar with the tools and what not, and you just recognize stuff faster.

I usually start by trying to fix a few bugs. Then maybe add a feature. Finding out a piece of the system that everyone agrees sucks and rewriting it. Writing tests isn't a bad place to start if something is untested.

I worked off and on as a contractor and being able to understand a code base quickly is a hugely important skill. I probably learned a lot from that.

Im writing tests right now, as well as looking at files where I have concept of what theyre supposed to accomplish and the tests related to those files. I also have been looking at random pull requests where they explain what its doing. So those all seem to help.

The major issues are:
1. I have no idea what most code is doing because I know zero about the domain/modeling and naming conventions
2. Theres a **** ton of integrations and I dont know how the integrations are supposed to work or what theyre doing.
3. Its in Ruby and Ive never done a non JVM language professionally :/

Is anyone on your team walking you through the code base? I'm pretty sure that was part of the first week for devs at Tendril.

Tests are good, but I think actually adding a feature / fixing a bug is better. I think it forces you into more parts of the codebase quickly.

But given the 3 things you describe above, its just going to take some time and work. Very few senior people when faced with those 3 issues are going to be up and running full speed right away.

Edit: For Kerowo's post, I'm not even sure a walk through of the code base is that useful. We don't do that with new developers. Instead we make sure they understand the general purpose of the application/service they're working on and the problem they need to solve and then assign a dedicated person to answer any questions that come up / review code.

Yeah, I thinks thats actually closer to what we did, explain the app more than get into the code and after it was all micro services it was walking through what the services did and how they communicated.

The best thing imo to get up to speed jumping in the middle of a project - is some menial search-and-replace-ish task that touches like 100 files in the codebase. It really helps speed up the process of learning where everything is, and after a while the patterns become clear. I always ask if they have something like that - which they might be hesitant to assign because they think I'd be annoyed or something.

If nothing else just go through and reformat the entire codebase the way you like it, then throw your changes away. Or do a PR with it and get off to a great start with your coworkers lol.

This is probably too glib of a description, but I think being a "senior" dev is less about how much you know about your specific given codebase and more about how you can strive to consistently make things better as you work rather than just adding **** as you work.

Re: bootcamps

It does not take Sherlock Holmes to find the profiles of people who have attended and their resumes are almost always the exact same thing. They have a "blog", cause someone told them to, which never has more than 3 entries and ends the moment they get a job.

All of their apps were clearly hand fed to them, on teams, so everyone has the same app, with a bunch of fancy talk about "what they hacked together" and it is always either a social network clone, something about recipes or uses some basic API to httparty up some json and do really really really interesting things with it... It does not take much imagination to wonder how everyone at every boot camp simultaneously decided to use the Starswars API for their "final project".. yes, I am sure it wasn't suggested by their half-asleep teacher for the 52th time.

Their resumes are filled with nauseating crap like "When I am not hacking things together I play with my cat", and "..built with Ruby on Rails, node.js and oolong tea." and "Ivy league graduate with a degree in Chemical Engineering who interned under the great Jackson Godson and helped him cure cancer while simultaneously working on a biodegradable alternative to Styrofoam, just finished the amazing Smash Code Ruby on Rails 3 month Blitz and ready to build something cool" ... they all have the most impressive education backgrounds, yet somehow ended up in a 3 month Ruby on Rails course "hacking together" a recipe app as their final project.

Yes, maybe there are few geniuses who want to program and enter a bootcamp but I doubt most of these clowns were doing that amazing in their past career if they decided to start over.

Well, I am a self taught programmer. Pretty cynical. Kind of a dick. Hopefully I can find a job cause I need to make $$ and really like this **** and hate human beings and want to be able to stare at the screen like I did with poker for all those 6 figure years.

cmon, tell us what you really think

Maybe programming salaries look pretty attractive to people who are doing average (or perhaps even relatively well) in other fields?

You should probably make something cool to show people that isn't as generic as all these bootcamp final projects you speak of, and then look for a job where you can work remote.

First day on Monday so I'll be running into the same issue but as a new grad.

Thankfully I don't have to jump into a codebase written in Ruby though. Should just be all Scala.