purpose: to develop a structured, rich media digital paper authoring tool with an objectbased model that enables interactive, playable, and convertible functions.

design/methodology/approach: we propose dpaper to organize the content (text, data, rich media, etc.) of dissertation papers as xml and html5 files by means of digital objects and digital templates.

findings: dpaper provides a structured-paper editorial platform for the authors of phds to organize research materials and to generate various digital paper objects that are playable and reusable. the phd papers are represented as web pages and structured xml files, which are marked with semantic tags.

research limitations: the proposed tool only provides access to a limited number of digital objects. for instance, the tool cannot create equations and graphs, and typesetting is not yet flexible compared to ms word.

practical implications: the dpaper tool is designed to break through the patterns of unstructured content organization of traditional papers, and makes the paper accessible for not only reading but for exploitation as data, where the document can be extractable and reusable. as a result, dpaper can make the digital publishing of dissertation texts more flexible and efficient, and their data more assessable.

originality/value: the dpaper tool solves the challenge of making a paper structured and object-based in the stage of authoring, and has practical values for semantic publishing.

purpose: to develop a structured, rich media digital paper authoring tool with an objectbased model that enables interactive, playable, and convertible functions.

design/methodology/approach: we propose dpaper to organize the content (text, data, rich media, etc.) of dissertation papers as xml and html5 files by means of digital objects and digital templates.

findings: dpaper provides a structured-paper editorial platform for the authors of phds to organize research materials and to generate various digital paper objects that are playable and reusable. the phd papers are represented as web pages and structured xml files, which are marked with semantic tags.

research limitations: the proposed tool only provides access to a limited number of digital objects. for instance, the tool cannot create equations and graphs, and typesetting is not yet flexible compared to ms word.

practical implications: the dpaper tool is designed to break through the patterns of unstructured content organization of traditional papers, and makes the paper accessible for not only reading but for exploitation as data, where the document can be extractable and reusable. as a result, dpaper can make the digital publishing of dissertation texts more flexible and efficient, and their data more assessable.

originality/value: the dpaper tool solves the challenge of making a paper structured and object-based in the stage of authoring, and has practical values for semantic publishing.