Steps involved in drawing a context/data flow diagram? (1 Viewer)

[fakermaker]

I'm pretty shit at drawing these.

Context is pretty easy, but DFDs can be so complex, I almost always miss important shit.

What process do you guys do to draw one?

I've started making a preparation guide for every DFD I draw. I write down the processes, external entities and data stores and draw them first.

From there, I link each external entity to a process and write its input/output. I do this until every external entity and process is linked, making sure I link it to a data store when necessary.

However, I always find that something is forgotten. Will the makers be strict if there are a few omitted details? How can I be sure I've covered everything in the exam?

 

[Uzername]

I think your 'preparation guide' is good, but I wouldn't think drawing them all up first & linking them later would be a good idea, you would probably have arrows going everywhere. I would just write them down first, then reading the information draw it up as I go, crossing off the entities/data stores/processes as I go.

If you're finding you are omitting details then it's a matter of re-reading the question. Each time you read a sentence, double-check and follow the path on your DFD and see if it makes sense. Should there be something in between? Perhaps this data should go to the process first, then to the data store? Ask yourself questions.

Consider the example of sending a tweet off a laptop to Twitter. One person's response might be:
[User]-->tweet-->(Upload Tweet)-->tweet-->[Twitter Server.
But, adding in more minute details, you could say: [User]-->tweet-->(Upload Tweet)-->tweet-->[ISP]-->tweet-->[Twitter Server.
And you could probably even add in a data flow from (Upload Tweet) back to the [User] giving a confirmation message. Lol, I know it's hard to convey this properly but it's about picking up the smaller dets and re-reading the information to make sure everything makes sense. Hope it helped