UML use case diagrams define a bunch of relationships between use cases. The two best known are include and extend. There seem to be more questions on these two relationships than on any other part of use cases, perhaps even anything in the UML.

You won't find an explanation here, rather my advice is to ignore extend. Just pretend it doesn't exist. Using extend properly isn't going to make any significant difference to your project - you almost certainly have much more important issues to worry about.

Include, on the other hand, is useful. You use it when you have a bunch of steps in a use case that's either duplicated between use cases, or makes sense as its own chunk. However don't take include too far - one level of included use cases will probably suffice for most cases.

In any case remember that use case diagrams are very close to useless. The real value of use cases lies in the content - the text that describes them. The diagram makes a visual table of contents, but nothing more.

If you want to learn about use cases, you should get hold of Cockburn. It is by far the best book on the subject, and is preferable to my advice.

if you found this article useful, please share it. I appreciate the feedback and encouragement