Is it necessary to have some basic knowledge of algorithms and datastructures in order to be a good programmer?
The short answer, yes, absolutely.
This is a follow up to my previous blog post where I claimed that you don't need to know advanced algorithms by heart in order to be a good programmer. That is still totally true. However, you should have at least some basic understanding of algorithms and data structures.
Having this knowledge will help you select the correct tools for the job, for example knowing when to use a stack over a queue, when a linked list is a better selection than an array, when to use a hashset, or a dictionary, and so on. This will help you write easier and more robust code.
It is also a really good idea to have some feeling for how your code will perform when input data grows. How will memory consumption and running time be effected if the input data grows by a factor 100, or 1000? This knowledge will help you identify performance bottlenecks in your application and help you come up with ideas on how to attack them.
So, if you haven't done it already, take an algorithm course or two. But don't put too much effort into remembering every little implementation detail of quicksort, quickfind, priority queues, red black binary search trees, binary indexed trees, etc. Instead, put some effort into learning what they can be used for and look them up when needed.
This is a follow up to my previous blog post where I claimed that you don't need to know advanced algorithms by heart in order to be a good programmer. That is still totally true. However, you should have at least some basic understanding of algorithms and data structures.
Having this knowledge will help you select the correct tools for the job, for example knowing when to use a stack over a queue, when a linked list is a better selection than an array, when to use a hashset, or a dictionary, and so on. This will help you write easier and more robust code.
It is also a really good idea to have some feeling for how your code will perform when input data grows. How will memory consumption and running time be effected if the input data grows by a factor 100, or 1000? This knowledge will help you identify performance bottlenecks in your application and help you come up with ideas on how to attack them.
So, if you haven't done it already, take an algorithm course or two. But don't put too much effort into remembering every little implementation detail of quicksort, quickfind, priority queues, red black binary search trees, binary indexed trees, etc. Instead, put some effort into learning what they can be used for and look them up when needed.
Var ett tag sedan jag var inne och läste, men så glad jag blev över att få läsa flera bra oc h intressanta inlägg! Jag har fått börja vara med i rekryteringsprocessen på mitt jobb, och håller helt med dig i att man inte bara kan basera sig på ett onlinetest... Ska ta en titt på algoritmkursen också :)
SvaraRaderaHej!
RaderaJag började undra vart du hade tagit vägen, saknade dina kommentarer :).
Algoritmkursen kan jag rekommendera, kräver dock att man lägger ner flera timmar i veckan för att man ska hinna med.
Ang rekrytering, kolla in den här länken, tycker det är vettigt skrivet: https://codeclan.com/blog/hire-developers-online-tests
Kul att du saknat kommentarerna :) Jag har varit pappaledig i sommar då vi fick nummer två i juli, så det låter inte som att det passar så bra för mig med algoritmkursen nu då. Inte så många sekunder jag inte vet vad jag ska göra... :) Tack för tipset, ska ta och läsa blogginlägget!
Radera