for document annotation is the successor to which has been offering web-based document annotation since 2008. It enables collaboration based around document annotation and review. was launched in April 2007 to facilitate access to scientific research papers. It allows researchers to easily keep complete and reliable records of their research output on the web.


We're always interested to meet with enthusiastic developers with a passion for clean and maintainable software design who might be a good fit for the team. Just get in touch, preferably with links to some examples of your work.

Internships - Summer

If you are a student looking for work over the summer in a small friendly software company in central Edinburgh, we may have just the thing for you. These are some of the projects we're interested in pursuing.

Annotation on tablets and mobile devices by converting our custom rendering code for Android and IOS and adapting the user interface for touch input (Java, Objective C).
Realtime collaboration using Node.js. We already do this for special applications. The aim is to deliver the same capabilities as a light-weight add-on for any Annotate workspace (JavaScript, Node.js).
PHP memory caching. Experiment with PHP memory caching (APC) to see what kind of performance improvement it can give us for serving selections from very large document sets (PHP).
Cluster architecture. Exploring alternative technologies to synchronize data across the cluster and provide warm fail-over servers (various).


Drupal Module to connect content in Drupal, including documents and users, into Annotate workspaces so that documents could be opened for annotation with a single click (PHP, Drupal).
Drag and drop search interface for composing complex search queries incorporating fuzziness, proximity, exact phrase searches and boolean relations (HTML5, JavaScript, CSS).
HCI and usability testing. Observe and record how people interact with various products to identify where operations could be improved.
Supervised document review to help in manual processing of large volumes of documents by suggesting matches and classifications based on exiting user input (Maths, Machine learning).
Client side test automation frameworks. Exploring what is available and developing browser based tests for the user interface.
Moving some logic to Google Dart. Taking existing client side javascript code and systematically, possibly automatically, generating equivalent Dart versions (JavaScript, Google Dart).


Since March 2016 Textensor Limited has been part of the Opus2 International. We work on software development for Opus2 products and develop and run the service.

A little history

Before starting Textensor, the founders had been working on problems of management and reuse of biomedical data. The key problem was with the large number of studies that yield heterogeneous information including papers and other textual material that does not fit well in conventional database models. Efforts to shoehorn information expressed in documents into structured or semi-structured databases have had mixed results and there has been very little progress in replacing text as the preferred way to represent complex information.

Realizing that databases could add to but were very unlikely to supplant written text, the developers of A.nnotate started from the other end. They began working on ways to let authors enhance texts with a structured annotation layer that would enable some of the results to be processed automatically while keeping references back to locations in the original text for each assertion. Sidestepping efforts towards automated natural language processing they focussed instead on how the author interacts with the system, with the aim of enabling authors to add structure and value, rather than just mechanically re-expressing what was already there. As their work shifted from pure research to more practical prototypes, they founded Textensor (Text + (ext)ensor) with an Research and Development grant from the Scottish government to pursue these ideas. The first prototype is described in a white paper: Enhancing documents with annotations and machine-readable structured information using Notate.

The origins of Annotate

The most enthusiastic user response to the early software was not for the sophisticated structured content features, but for the core capabilities of tagging, annotating and collaborating on text documents. The focus therefore shifted to rendering and annotating a wider range of document types including PDF and Word as smoothly and easily as possible. This led directly to the hosted solutions at launched in January 2008 and was soon followed by standalone server products for Linux and Windows. A major upgrade at the end of 2008 saw the addition of image annotation capabilities, comprehensive document storage and management, and an extensive API for the growing number of customers who combine A.nnotate with their own web aplications.

The company's origins in university research are reflected in the substantial discounts and flexible licensing conditions offered for servers to be used in education or publicly funded research.

Get in touch

Textensor Limited is registered in Scotland, company number SC294230.