Text and attribute-based messaging is a useful technique for sharing information and services between '''diverse''' application languages. The technique was made popular in Unix/Linux but is not exclusive to Unix/Linux. Characteristics include:

* The text message can be viewed as a discrete unit of text. It can be thought of as a file.

* The text is human-readable without special viewers or browsers other than a plane-jane text editor. (Although the meaning of the structure may be context-sensitive).

* Any data-structures used are relatively easy to parse.

* It is mostly declarative information rather than commands.

Examples include:

* Markup text or files, such as HTML or XML

* Comma-separated value files

* An HTTP message