OER 4 Low Bandwidth

From Bjoern Hassler's website
Jump to: navigation, search

Articles:


2007-11-20

Overall web design guidelines (as well as overall access guidelines) apply, see Web Design 4 Low Bandwidth.

But two paradigms are particularly relevant for OER delivery in general:

  • Alternatives ('video-audio-text cascade'): Provide audio and text alternatives to video assets
  • Layered delivery (as opposed to compound delivery): Users should be able to obtain the object in parts, not as an aggregated object. Another good term for this would be 'disaggregated delivery'.

These two ideas, while they are basic, are new paradigms in the delivery of Open Educational Resources. They were first published (as far as I can see), in the multimedia chapter Web Design 4 Low Bandwidth guidelines, and I thought it would be valuable to pull these ideas out from that chapter, to examine their relevant for OER delivery in general.

Video-audio-text cascade: Of course many video items are provided as a 'high bandwidth' and a 'low bandwidth' version. However, we argue here, that you should provide an audio only version. The audio only version can be very small indeed (say 5-12kbps), and can be reasonably high quality at moderate bandwidths.

Note that an audio or text preview of a video is not necessarily the version that is eventually downloaded by the user. But through a low quality audio preview or a text description, the user can make an informed decision as to whether they want to download a higher quality version or not.

Layered delivery is the idea that users should be able to obtain the object in parts, not as an aggregated object. In a sense, it is a 'progressive download' for a whole learning object. Rather than having to wait for a zip file to download entirely before opening it, you should be able to obtain sections of the course in part, but ideally without having to download them individually, or having to track with parts you already have. Not only should you be able to obtain parts individually, but you should be able to get different versions.

A surprizing ally for layered delivery is podcasting, and podcasting application, like iTunes. For instance, podcasting into iTunes does exactly the right thing by default: You can subscribe to a text only index of what's available (the podcast feed itself), and then choose what to download. Once you are downloading something, the connection can fail, you can quit and later resume iTunes, pause the download etc. Eventually, the file will reach you. The process could be fine tuned, by setting iTunes to not download any episodes automatically, and by refreshing the feed only manually.

An interesting corollary of this is the idea to provide 'layered course cartridges', where you might be able to download the cartridge information, but none of the contents, and you decide from within your virtual learning environment which elements to get.

For example, an OER package might then have the following components:

  • The index. A text file, e.g. rss, or another more customised format, that details the contents of the package.
    • Part 1: Background notes on the OER, teacher's notes or similar (say Chapter 1)
      • Layer 1: Plain text
      • Layer 2: The same document as optimised pdf
    • Part 2: Background notes on the OER, teacher's notes or similar (say Chapter 2)
      • Layer 1: Plain text
      • Layer 2: The same document as optimised pdf
    • Part 3: Video asset (e.g. a short documentary)
      • Layer 1: Video script as plain text
      • Layer 2: Some images from the video, to illustrate the contents
      • Layer 2: Audio only version
      • Layer 3: Low bandwidth video (e.g. mobile phone compatible)
      • Layer 4: Standard high bandwidth video
    • Part 4: Video asset (e.g. a lecture relevant to this)
      • Layer 1: Video script as plain text
      • Layer 2: The powerpoint presentation as optimised pdf
      • Layer 2: Audio only version
      • Layer 3: Low bandwidth video (e.g. mobile phone compatible)
      • Layer 4: Standard high bandwidth video

This is of course an ideal scenario, but aspects of it can be incorporated into OER production now. Additional advantages are that some components, e.g. the script, are archived with the resource, for later reuse in other contexts. Otherwise the script might have been lost.