JSON is not a document markup language; it is a data structure interchange format. KnitML actually needs to be a document markup language. Therefore your exampl

JSON is not a document markup language; it is a data structure interchange format. KnitML actually needs to be a document markup language. Therefore your example implying that XML was the wrong choice for KnitML is based on a lack of understanding about what the language's intent. Also, you seem to imply that an "extra 30Mb in software" has something to do with the fact that it's processing XML. That is a pretty wild assumption and is simply not true. First of all, the 30Mb is in reality too large because of a Maven bug which duplicates JAR file dependencies in a ZIP archive. The true size of the app is about 15Mb (and the upcoming release will in fact be much smaller). Secondly, JiBX is the only third-party code which does XML processing. The rest is built into the JDK. So you're really only talking 7 Mb. But it's not relevant, because you are not making a clear argument.