Find similar code with NLP using software documentation

find similar code using NLP on software documentation

the value in good software documentation

Writing software documentation can be frustrating. It’s a post-hoc, abstract task (at least for the programmer). Nobody might read it, and even if they do, they might not get it. The partner post describes how an individual software engineer could overcome not being understood, this post is meant to show the added value that good documentation could bring, if you would put it to good use.

The pain: scattered pieces of code

Imagine that your documentation could actually help your company with future software development. I’m not talking about someone trying to figure out what you did with a piece of code, I’m talking about what piece of code the company thinks would be useful to develop. The problem is overview. Software gets written throughout large corporations, across different departments, that sometimes happen to need very similar things. Unfortunately it is very difficult, if not impossible to know what other piece of similar code already exists. 

The solution: let documentation lead you

But what if you would write documentation about what it is your code solves, and then magically, you would be directed to the most similar pieces of code? 

It could lead you to find previous expertise that would speed up your development by pointing you towards pitfalls he encountered when implementing the thing. 

But it could also lead you to finding out that what you wanted to make, already exists. So you can make use of that other person’s code and put your efforts into making something brand new.

Is this interesting for you?

Would you like to know more about how NLP can help you when writing and managing technical documentation? We are working on it! Please let us know and we will keep you up to date about our solutions. How?

  • Write to us in the comments below
  • Send us an email to information@avantopy.com
Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn

Leave a Reply

Your email address will not be published. Required fields are marked *