If you are pairing a program, do you still need an expert review?

I think that in general Peer reviews are a very good part of the development process, they often catch or ask questions that were not obvious when the code was originally written and made you more self-conscious, so that you better format, add comments, etc.

However, if you are pair programming, you have effective live reviewing, is it still worth considering it as part of the process? Can you get a couple of reviews?

I ask how para programming begins, where I work, and it is usually seen as a substitute for peer review. I'm not sure, but I think that the development time spent on pair programming and peer review can hurt performance.

There was a similar question a while ago, but with a different emphasis and without a clear consensus

+5
source share
5 answers

It depends.

The purpose of the expert assessment, in my opinion, is not only to find defects directly in the written code, but also to make sure that the code will also work well with the existing code base. Sometimes you may need an expert code that you write, and it cannot be a member of a pair.

For example, if you are writing some of the 3D graphics of an application, you might need it to be viewed by your OpenGL expert.

, , , . ( - ).

, , . , .

, , 100% .

+7

, ( , "" ). , , - , () - , - .

+3

, - , , , , , , invloses , , , , , ,

+3

. , , / . .

, .

+1

Paring - . , , - , . , , .

, , . , , , (, A.K.A Test Driven Development) , .

- . :

. , , .

: , , , . , -, ... , .

. . , , .

Transfer of skills: Rotating couples learn new skills (engineering and domain) to each other, working together. The team level will grow for everyone, and knowledge is disseminated through the team.

The team chooses independently: The team learns one of its skills and quickly eliminates those who do not.

0
source

All Articles