Case study: Softeco Sismat (TELL ME Project, #2)

The large companies that use Dojo are widely known. This series features lesser known users of Dojo, and their stories. Eight months ago, we conducted a case study about the TELL ME project with Stefano Bianchi from Softeco Sismat, an ICT Italian company. Here we have followed with up Stefano to get an update on their progression from desktop web app to mobile with Dojo.

TELL ME Login
TELL ME Mobile UI – login

Q: How did you first learn about Dojo?

A: We were looking for a solid JavaScript framework to develop a prototype for a complex, desktop-like web user interface (UI) to support learning activities for blue collar workers at the workplace. We evaluated several different renown frameworks and we found references to Dojo in several JavaScript framework evaluation tables – we then decided to give it a chance considering its features and the positive feedback. The experience made during the technical activities (development, testing, debugging etc.) and the feedback gathered from both developers and end users confirmed the general good impression we initially got.

Q: Why did you choose Dojo?

A: At a very first glance, Dojo seemed like a robust JavaScript framework for client-based desktop-like applications, shipped with several interesting built-in features (plenty of ready to use components and widgets, such as e.g. the Store-based Tree widget and the rich chart & graphing library). Also, the documentation (and tutorials, in particular!) was rather rich and complete, providing an incremental guide for differentiated (novice, intermediate and expert) users. The Asynchronous Module Definition (AMD) and Object Oriented (OO) approach was also a plus. As for our previous expertise in Rich Internet Application (RIA) development, what really convinced us to adopt Dojo was the possibility to replicate on client-side the same desktop-like functionality that we previously deployed with a Java-based RIA frameworks on server-side. As the project we are participating in also addresses mobile devices, the Dojo Mobile toolkit represented a valuable asset for future development. This case study focuses in particular on the development of a Mobile UI that we designed and implemented to port on smartphones some RIA functionalities we developed for plain web desktop UI, in order to provide a pervasive training experience and allow launching native multimodal apps on Android devices.

Q: Were you previously using another toolkit?

A: In previous research projects, we extensively used several JavaScript libraries and frameworks – Prototype, Scriptaculous, jQuery, ExtJS to name a few. Working with Dojo represented for our Research & Innovation division a good opportunity to extend our expertise in JavaScript frameworks. All technical developments for the TELL ME learning content consumption UI (desktop and mobile) have been carried out with Dojo and Dojo Mobile from the very beginning.

Q: What does your application or service do?

A: The prototype we developed for the front-end user interface of the TELL ME (Technology-Enhanced Learning Livinglab for Manufacturing Environments) system includes several different widgets (“chunks” of HTML5/JavaScript/CSS code that can be used independently or assembled in a portal/desktop-like layout) that provide different learning-oriented functionality. The main aim is to allow a blue collar worker to express a learning need (e.g. “what do I need to learn/be trained/recap about?”) by e.g. selecting a set of predefined tags from several complementary domain taxonomies (i.e. trees of concepts describing activities, tools etc. in a given workplace) and to receive back a learning path (i.e. a list of learning-oriented activities) including different learning contents to be consumed in order to acquire or improve specific working skills. Each type of learning content (documents, videos, pictures, lessons etc.) should be then consumed/launched in a specific widget, with the possibility to also provide social feedback such as ratings and comments. After developing the TELL ME web UI, we started designing and implementing a porting of the main functionalities on mobile devices, to allow using the same REST WS back-end services and launch native apps on Android devices.

Q: How does your application use Dojo?

A: The whole mobile TELL ME UI has been developed from scratch with Dojo Mobile. The application (with Dojo core “has” module) “sniffes” the features of the device and forward to the desktop or mobile UI. The trainee/blue collar worker can then login and access the main page where several options are provided as a rounded list (“Search”, “To do”, “Expertise”) and a useful bottom navigation menu is included. The “Search” functionality in particular allows to browse a domain-related taxonomy where tags can be selected to launch searches on a remote Apache Lucene indexing server and retrieve a list of learning contents to consume, by launching native applications on the device. If the returned learning content has a specific proprietary MIME-TYPE, a dedicated Android app is automatically launched on selection, allowing the trainee to follow a step-by-step job card in a multimodal way (Text-to-Speech plus Speech Recognition features).

Q: Overall what is your user experience with Dojo?

A: Also Dojo Mobile generally confirmed the initial expectations: the good impression it gave at a first glance was confirmed by the technical results achieved by the prototype, which was developed without particular problems. Dojo Mobile required a short period of training but, after the experience with the base framework, the learning curve was not steep at all and the examples provided in the tutorials helped a lot in setting up working prototypes in a few days. We implemented a navigation tree to allow selecting the tags to launch the search and we will work in the next period on optimizing both usability and layout.

Q: What’s your favorite thing about Dojo?

A: Dojo architecture is truly inspiring when approaching Object Oriented (OO) JavaScript, and the rich set of examples provided help understanding how to make JavaScript code modular and reusable throughout the application – the whole approach is clear, clean and scalable. Reference documentation and tutorials are also a valuable resource. The design we followed for the implementation of the desktop UI was useful also for the mobile version as several JavaScript modules were easily reused allowing to link immediately the new presentation layer with the existing REST WS business layer.

Q: What are your future plans with Dojo?

A: After a first technical evaluation phase, where end-users tested the developed applications, we collected useful feedback to work on in the next months. In particular, we would like to make the desktop UI more responsive, and we are likely to start from a fully responsive framework (e.g. Bootstrap) to revamp the presentation layer while keeping Dojo modules and back-end services unaltered.

The templating mechanism still have to be completely understood and evaluated, and surely represent a direction to follow for further developments.

The prototype will be extended to include the complete logging of all learning experiences by mean of a dedicated Dojo-based logic connecting the UI to a Learning Record Store (LRS) as specified by the ADL specifications for the Experience API.

Furthermore, as the TELL ME project addresses also the use of advanced Human-Computer Interfaces (HCI), the Dojo-based UI will be ported as the HTML layer for an Augmented Reality (AR) application developed on the Unity 3D engine.

We moved to Dojo 1.10, but we are definitively eager to start playing with Dojo 2.0 as soon as possible!

TELL ME Mobile UI – main
TELL ME Mobile UI – main
TELL ME Mobile UI – user info
TELL ME Mobile UI – user info
TELL ME Mobile UI – taxonomy tree tag selection
TELL ME Mobile UI – taxonomy tree tag selection
TELL ME Mobile UI – search results
TELL ME Mobile UI – search results
TELL ME Mobile UI – seamless launch of Android app from web UI
TELL ME Mobile UI – seamless launch of Android app from web UI

TELLME Technical Webinar

About TELL ME

TELL ME (Technology-Enhanced Learning Livinglab for Manufacturing Environments) is a research project co-funded by the European Commission to improve training in small and medium-sized manufacturing environments by using the latest technologies and insights. The aim is to bring innovative learning methods to manufacturers so they can better supply the needs of their markets, which include larger industries. TELL ME is a three-year project that started in November 2012 with a partnership of 14 leading organisations in academic research, technology and industry from Italy, the UK, Finland, Sweden, France, Germany, Spain and Portugal. It is a significant €8.3m R&D project, funded by the Consortium Partners and a €5.9m research grant from the European Commission under its FP7 Framework Programme.

Thanks!

Thanks Stefano for telling us about your experience with Dojo. If you would like to share your experience with Dojo, please contact us.