ReqIF’s XML overhead

XML formats can have a tendency to be a little bit verbose (the AUTOSAR exchange format can be an example for this). To get an impression on the overhead of the ReqIF standard, I fed a document from Project Gutenberg to our Requirements Modeling Framework (see Eclipse Proposal). Breaking down “Antarctic Penguins A Study of Their Social Habits” by George Murray Levick into one requirement per paragraph (what text would be better suited for an open source project than a text about penguins), results in a document with 636 requirements.

The original file size is 198936 bytes, the .reqif is 560221 bytes. That is an overhead of (560221-198936)/636 = 568 bytes per requirement.

The paragraph

is in XML:

 

In addition, to build an hierarchical structure, each requirement/paragaph needs to have a spec hierarchy.

Bildquellenangabe: tokamuwi  / pixelio.de

2 thoughts on “ReqIF’s XML overhead

  1. Yes that is very true in my opinion. One more example is the element in every hierarchy.. The added value is less than 0.

    would contain the same information. When implementing an import or export this causes additional time consumption an also headaches.

    What is really strange is that there are two critical errors left in the reqif-model and reqif.xsd which where konw before the acceptance as a standard by OMG. Do you think they will ever be fixed?

    Eckhard Jokisch

Leave a Reply

Your email address will not be published.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