Purpose

• compile ideas in a common place
• create a living document that will grow and further extend our knowledge base as we better define the problem
• contains both practicle and possible solutions
• get some feedback
• fully define the problem to create a viable solid solution and the steps needed to obtain it
• get started so we can create some kind of roadmap for development milestones

The Problem(this list should continue to grow)

A device is needed that:

• is low power
• is cost effective
• functions as a remote content repository
• distributes content to the talking book device
• is capable of content play back
• has content authoring capability
• can accept custom user created content
• can sync to files from another repository (remote sync or multimedia card)
• is usable by individuals with little to no literacy level
• is usable by blind or visually impaired individuals possibly with literacy levels similar to the above
• has some sort of visualization - Graphical User Interface
• retains the functionality of the talking book device for operation
• reports usage metrics
• has more stable power source than retail devices so that it can always provide reliable time/date when synchronizing with devices

"The 10,000ft view wish list"

Hardware

• Low power or solar power capable
• 1GB ram
• Embedded device(pico itx or normal itx motherboard)
• LCD or LCD Touch screen (touch screen would be completely inaccessible for the blind)
• Keyboard
• Talking Book UI Controller
• Media card reader
• Modem
• Ethernet
• Wireless
• Cellular
• Barcode reader/scanner

Kiosk Storage

• Internal Harddrive
• Internal Flashdrive

Software

• Linux Ram Disk
• XServer
• Java

Support

• USB
• Wireless
• Ethernet
• PPP(Dial up)
• Cellular
• Audio
• Barcode reading/scanning
• Single device connection per session

Feature List

• “Picture/Graphical” based Swing UI designed to accommodate little to no literacy level(should be able to function in a similar manner as the talking book)
• Accessibility for visually impaired must be based on synthesized speech rather than exclusively depending on pre-recorded mesages such as with retail device
• A paired down controller should possibly accompany the kiosk/software that allows for operation similar to that of the talking book device
• Talking book device should appear as such or as a specified storage device when connected
• Allow barcoding of published content created with the Talking Book Authoring System(ISBN)
• Device Uploads
• Device Downloads
• Determine device capacity
• Repository Synchronization or some sort or available media insertion
• Authoring system capability
• Net/repository upload
• Net/repository download
• Usage metrics recording:
  • • user created vs published content
    • uploads/downloads
      • • average time
        • file size
        • number

Component Breakdown of the wish list

We are going to breakdown the components here at a fairly highlevel still.

Operating System:

Topics to note

• If the kiosk will use a GUI interface then the only realistic option we have is Gnome with Orca and eSpeak with Firefox 3 for the blind and visually impaired. If it will use text-based console, then we could use YASR or Emacspeak. A text-based interface is not realistic given the target audience.
• Braille devices are prohibitively expensive. Those sold in the US start at over US$2K, and even a Chinese-made version starts at over US$1K.. We need to stick to synthesized speech via Gnome + orca + eSpeak to keep costs down.

Distros to consider

These are not set in stone and should continue to grow as distributions are found that support what is needed:

• Suse Linux
• Puppy Linux
• Solaris or Solaris derivative

The distro of choice would basically be stripped of all non-essential software and left with the following:

Third Party Packages

OS level packages that are utilized

Third Party and Open Source Libraries

Libraries utilized in the software development

Hardware

What we should be looking at for hardware and why