Just curious, why the "millennial" descriptor? Is having trouble understanding the OS something unique to only "millennial" self-taught programmers?
Re:The Imposter's Handbook
by Ensorceled on Wednesday August 24, 2016 @01:31PM
(#12353680)
In my experience, Gen X and Boomer self taught programmers spent a lot of time hacking at the application or system level and have less holes in that area. They have other holes in their knowledge base, just not those ones.
That's exactly the reason I chose that descriptor. In fact, the field "Computer Science" may not have existed before.
Everyone has gaps in their knowledge. There are two devs where I work, and we are both better than average in my opinion. I know Django inside out and deal with server deployment, while the other guy knows more front end and HTTP / web stuff. We both know how to use a database properly and not to over engineer stuff. We compliment each other pretty well.
Right, and I agree everyone has gaps in their knowledge. People obviously specialize and I wouldn't ever work on compilers, for example.
I'm just describing what I often see the self-taught "millennial" programmer is missing with a minimal number of data points. Doesn't mean all of them do, but hopefully by me listing it here a book such as the above can tackle these topics. I hope somebody reading this doesn't take offense of my laundry list; it's just intended to show what I think would be useful to cover in such a book given their importance to app development.
I'm no way knocking self-taught programmers. They learn a lot faster than those who went the formally trained route. One of my interns who was trying to get a master degree in CS I showed how to link to a library 3-times, and he still couldn't get it.
Pipelining is another concept that I feel self-taught programmers have weaker foundations--many of whom I have worked with write code that waits for all results to become available, while an operation that's blocked by I/O doesn't necessarily mean we can't do stuff with the CPU while we're waiting for the next batch of I/O to come through.
I have also seen self-taught programmers accidentally write O(n!) or O(2* * n) functions and not realize it. I think data structures is definitely a good chapter to have. Especially when writing queries to a data store.
I think explaining how a hash table works would be excellent since it is such a useful and fast data structure. A lot of set-taught programmers sort of treat them like magical black boxes when it's not a very complex data structure yet it's practically O(1) for most insertion/reads/deletions.
Memory management is fortunately something we don't really need to worry as much about. With languages and interpreters that do a very good job of cleaning up after our code and now that memory is relatively cheap, we can afford to ignore it until we need to scale.
Of course, if you have a good product, you can get away with inefficiency and hire CS guys when you have built a unicorn. ;)