I was reading about this on Friday. If you want to avoid copyright infringement you need to do the following:

1) Get a team to look at the API and write a specification document

2) Get lawyers to look over document for any patent/copyright infringements

3) Get a completely different team to implement the specification knowing nothing about the original implementation

I'd like to know where you read that on friday :)..

I've developed software (and music--also great fodder for copyright discussions) for decades, ranging from companies that everyone on the planet has heard of to tiny companies no one ever knew existed.

We never did anything remotely like that "to avoid copyright infringement".

Step 2) there in particular is massively expensive, and I've never known that to be done as a matter of course.

What that does look like is how you might go about reverse-engineering something and be (semi) safe, from a legal perspective (i.e a 'clean room' implementation of an existing piece of software). I don't get the idea that's at all what the original question was about.

Frob (as always) gives great answer.

To avoid copyright infringement, write your own code from scratch.

That said, you can use common sense. If you look up a bubble-sort algorithm in a common textbook and implement it in 'C, you aren't infringing on any copyright.

if you look up something on StackExchange, etc, use it is a reference. If there's not much to the code (aside from simply an example of proper coding or API usage), then copying and altering it is almost certainly fine. That said, don't cut and paste willy-nilly from the internet, unless you are very aware of (and confident in!) the licensees associated with the content.

If discussing the ethical aspect of plagiarism (independent of copyright issues), one answer is to put a link in the source code to where you found the solution. This is useful for yourself (if looking at it years later, and you wonder why on earth you put that code there, or what it does), and is also indicating to anyone else reading the source that you're crediting your sources.

Just as a note, copying things from stackoverlflow.com is perfectly safe, since everything you share there get Creative Commons Attribution Share Alike license

I'm sure most sites like it works similarly

Fully understanding any code you include in your project is of course a good idea in general

Careful - CC BY SA would require giving attribution. The "share alike" would also require the source code that references the copied source code (and presumably any compiled binaries) to also be under CC BY SA. This would be a pain for commercial users (who presumably don't want their programs to be freely distributable), and also potentially Open Source (since CC BY SA isn't commonly used for source code, so you'd have to worry about compatibility with licences like GPL, BSD etc).

If you are using code from elsewhere, in my opinion you should be very careful about what the licence is.