iPaper
Left Border
Prof. Dr. Beat Signer
Vrije Universiteit Brussel
Department of Computer Science
Pleinlaan 2, 1050 Brussels
(Belgium)
+32 2 629 1239, bsigner@vub.be
Office: PL9.3.60 (Pleinlaan 9)
VUB
View Beat Signer's profile on LinkedIn twitter View Beat Signer's profile on Facebook View Beat Signer's profile on YouTube Instagram View Beat Signer's profile on academia.edu View Beat Signer's profile on Google Scholar View Beat Signer's profile on ResearchGate View Beat Signer's profile in the ACM Digital Library View Beat Signer's ORCID profile Slideshare View Beat Signer's profile on Speaker Deck View Beat Signer's profile on 500px View Beat Signer's profile on SmugMug

iPaper

The base unit for linking paper documents is a single page. Therefore, in the case of the interactive paper plug-in (iPaper) developed in the Global Information Systems (GlobIS) research group at ETH Zurich, a resource is represented by a single page and a selector is an active area defined by a shape within that page. The iPaper plug-in supports simple links between printed and digital materials as well as highly interactive applications where users can easily move back and forth between the printed and digital worlds.

Since we are not interested in capturing and digitising physical objects, but rather in augmenting them with supplementary information, we do not have to store a digital version of a document’s content within the iServer framework. We only need to store some metadata about a printed document, such as its dimension and the number of associated pages, in the form of a digital document model. The digital document model allows us to define selectors on documents which can then be used to link from physical documents to digital content, from any digital information to physical content or to build links between physical documents. For that purpose, we need the concepts of documents and pages providing us with meta information about specific documents. The document type has been modelled as a container which is able to contain zero or more pages. Each document has a unique document identifier which is used for the mapping from a physical document to its digital document model instance. The only information provided by a specific page is its page number and dimension. By modelling documents and pages as two separate concepts, in a first step, we can use either a whole document or a specific document page as a source or target anchor of a specific link.
iPaper architecture
Fig. 1: iPaper architecture
If one thinks of how we work with paper documents and organise them, often we do not annotate or link entire documents or pages but rather mark specific parts of a page with a highlighter pen or write marginal notes. Furthermore, section headings and document indices are often used as the basis for providing links within paper documents. Therefore, it is evident that we need to support a means of addressing specific parts of a page. The concept of the selectors provided by the iServer link model is exactly what is required to address specific parts of a resource.
The introduction of geometrical shapes as a specific form of selector for interactive paper enables us to define active areas on a page. Each time a user points to a position within an active region, its associated links will be activated. In the case of interactive paper, the page type becomes a specific implementation (subtype) of the more general resource type whereas the shape and its subtypes are resource-specific implementations of the selector concept. The definition of a page selector implies that an input device for the interactive paper framework provides a document identifier and a page number as well as the absolute (x,y) position within a page. This input data is then used to send a request to the iPaper plug-in which has to resolve the selected shape.

After a request has been sent to the iPaper plug-in, we first fetch the relevant document based on the unique document identifier. Using the page number information delivered by the input device, we retrieve the corresponding page object. The only thing to be done now is to check which page selectors (shapes) are selected by the input device’s current position. This means that, for each selector defined on the processed page, we have to check whether it contains the current (x,y) position. In the case of multi-layered active areas, the shape positioned on the topmost layer has to be selected.

Detailed information can be found on the Interactive Paper project page.

Related Publications

  • 2024

  • thumb Pen-based Interaction, , Handbook of Human Computer Interaction, Major Reference Work, Springer Nature, 2024 (to appear)
    Available:  document
  • 2021

  • 2019

  • 2017

  • 2010

  • 2005