The Twilight Minds Design Group announced the release of TMML, an extremely minimal subset of XML, along with a Java TMML parser and tree implementation.

Twilight Minds Markup Language (TMML) is similar to Minimal XML. It pares XML down to elements, text, and well-formedness, omitting attributes, mixed content, DTDs, entities, CDATA sections, PIs, comments, the XML declaration, validity, and just about anything else. Said Reason, a Twilight Minds co-founder:

XML is very useful, but also comparatively slow for simpler parsing operations. TMML is a functional subset of XML that can be parsed more rapidly with the specialized TwilightMinds.TMML package. In addition, TMML can easily be used with XML-compatible applications. In fact, a TMML document itself is a valid XML element.

The code, which includes a parser and indexed and non-indexed trees, fits into a .jar of just over 6KB and is available under the GNU Lesser General Public License. The tree implementation has a single kind of node (an element), which can contain text as well as children. It has most of the expected operations, although the ability to find a parent and to delete a node are conspicuously absent.

Like Minimal XML, TMML could very well find a place in data-centric applications where speed is critical and the sources of XML documents are trusted.

To download the code, see http://www.twilightminds.com/TMMLfull.html.