Here is a paper on “Relative Focus based Simulation: Re-using Software Screen-casts for Mobile Learning“.
You can download it from https://samirshomepage.files.wordpress.com/2010/01/relative-focus-based-simulation.pdf
Below is the details of the Paper:
M-Learning/Mobile Learning is a vast domain of training methodology using mobiles and cell phones similar to the e-learning where the medium is normally a desktop or web access over a PC. If we notice ground reality of the usage of M-Learning, it is limited to some state of the art course ware with aim to provide mini-chunks of learning materials while the user is on the go. Mostly it is limited to mini m-book, glossary, mini test/assessment modules with MCQ type questions. Where as the e-learning is widely used in the software training industry and the screencasts /software screen simulations are nearly synonymous with the e-learning term itself it fails utterly in the M-learning domain due to the small screen size pf the mobile devices where the bigger resolution screens casts of the desktop/web applications are nearly impossible to display with the usable tag attached to it. Even with rise up of the bigger mobile resolutions through the handsets like Nokia N95, N96, N97, 5800 Xpress Music, iPhone and some HTC handsets (with some typical screen resolutions of approx. 240 x320 px )the issue still remains as the minimum screen resolution for any typical e-learning screen cast is approx 800x 600 px which is easily usable over a player that runs on a standard PC having 800×600 px (in full screen mode of the player) or the 1024 x 768 px resolutions.
Another factor is even if it would have been possible to deliver the typical 800×600 px screencasts in the over the 240 x 320 px mobile screens, what about the screen casts already developed and deployed in various e-learning courses across different WBTs/CBTs ? Would that need creation of two different compatible video files for these two different screen resolutions? Let’s assume that a screen cast is meant for web viewing, then this would mean that more space would be required for storage of the video files in web server, from where the PC and the mobile can access the different video based on their compatibility for the same screencast. Looking at this situation, the ideal solution for the above said issue must implement re-usability of the existing screencasts in some way.
Scope of the Proposed Solution
The scope of the proposed solution consists of the three major elements:
- mechanism to make the screencasts of 800 x 600 px resolutions usable in small device screen resolution , typically 240 px X 320 px
- Provide mechanism to reuse existing screen casts if possible
- Provide a way to use the same screencast video file for both the devices (PC and mobile) so that it will make deployer’s life easy for not having to store 2 versions of the same screencast.
The proposed solution is based on a very simple theory that we see only what we want to see. This is a very common idea that even drives many screencasts of web /desktop based applications created for many CBT/WBTs. The idea is that we need to highlight the area that is being referred in a screencast. And this is what is called Relative Focus based Simulation.
This technique involves
- Showing a selective region of the screencast that is of higher importance at any point of time and is more meaningful in reference to what voice is being played at that time.
- Zooming in/out provided for the selected region for better viewing making the overall appearance more meaningful.
For example in a screencast when the voiceover says “Using Adobe Photoshop you can edit images and export them in different formats” , typically we focus on the whole application rather than on any particular button, menu item, link or any specific text.In this case smaller minute details loss their importance in the over all appearance of the screencast.
When the voice over says “Go to File > Open in the top menu to open an existing Photoshop document” – here we focus on the region that has the top menu bar and we follow the mouse cursor movement. Here we actually do not notice what is the label text written in the far bottom corner in the right, or if the whole window is visible or not. In this case the micro details of the region of higher interest gets higher priority than the rest.
Now in above two scenario we can see that the region of relative higher importance will decide what the user needs to see in a screen cast rather than showing the whole window of the application. And this is the key to the solution being proposed.
This solution can be implemented on existing screen cast videos of 800 x 600 px resolutions via an editor, that will help in determining the region of higher relative importance in the overall appearance of the screencast at any point of time. The tool kit required to implement the solution must have two components –
- A screencast editor component, which will help in editing the existing screen cast video file of 800 x 600 px and export the video with the compatible meta tags in a form of either embedded metadata or an external XML or any meta data formatted file.
- A screencast player for mobiles which will take the edited/published screencast video with either the embedded metadata or along with the xml /metadata file to play as per the relative importance of the different screen regions of the original screencast video.
As all the editing information will not change the video binary and will only change the meta-tags only, it will take much less time to convert a normal screen cast video into a compatible one during its publishing phase. Also the same video can be used as usual in normal screencast video player in the PC based e-learning, without impacting its performance or any thing related. Also the same file can be streamed from the server for both PC based e-learning and the mobile learning where different type of views will be served depending on the device screen resolution.
Proof of Concept
Based on the above proposed solution, I have created a proof of concept tool kit codenamed “MobiCast Tool Kit” (Mobi for “Mobile” and “Cast” for “Screencast”). This Tool Kit has two main elements:
- MobiCast Editor-cum-Publisher
- MobiCast Player for Mobile handsets
MobiCast Editor-cum-Publisher is a desktop tool that allows the user to to edit existing screencast videos in .flv format (800 x 600 px screen resolution) and publish these as compatible for 320 x 240 px resolution.
The current version of this tool is developed in Adobe Flash. This tool allows developer to import an existing screencast video in .flv file format and creates a timeline for easy editing . It provides Normal, Preview and Edit Mode to carry out different task . This currently supports 800×600 px input video resolution and publishes for 320 x 240 px screen resolution which is normally can be played in Nokia N96, N97 handsets. The publishing of the video screencast exports the required mobile player component MobiCast Player. The out put can also be implemented as XML file.
This is developed with Adobe Flash Lite 3, and currently supports the resolution of screensize 320 x 240 px . This requires the handsets to have the Flash Lite 3x to be installed.
You can see a resultant output video demo in action on N96 at the following YouTube link: http://www.youtube.com/watch?v=VsYpNe2u8To
© 2009 , Samir K. Dash, www.mobilewish.com