Early Decisions on Reuse of OER: Copy or Link?

Mar 23, 2009 at 9:13 pm, Jared Stein

In David Wiley‘s IPT 692r – Intro to Open Ed course students have fragmented into two small groups, each of which has chosen to research and catalog appropriate open resources that may be used to fulfill learning objectives for one of the secondary education core curricula for the state of Utah. As I have begun searching for, tagging, and sharing resources, I’ve begun to consider the long-enduring web question: link or copy?

I mean, of course, with respect to appropriately licensed (Creative Commons, Gnu Free Document License, etc) open educational resources specifically.

And though the question is not staggering, it may be taken for granted, even at the cost of the long-term success of the web project.

Linking

The link approach typically uses hyperlinks to the target source document, but may use iframes to embed the element within a locally-hosted web page.

Linking…

  1. preserves integrity of the original source by maintaining all original qualities
  2. respects original source by trajecting traffic to the host site
  3. saves local hosting resources (storage & bandwith)
  4. ensures that source updates are reflected in the current version
  5. is, therefore, particularly well-suited for frequently updated or improved sources, like wikis
  6. is much easier, particularly when numerous multimedia files are embedded, or multiple files are referenced
  7. may provide learners with context and hyperlinks that lead to further, relevant exploration of the source site and the web
  8. avoids problems with licenses or terms of use that restrict copying

Many of these arguments for linking presume that there is more to the information than the information itself, and that the source has some inherent value that may be passed on to the learners or should be maintained for its own sake.

Copying

The copy approach is similarly self-evident: a digital copy of the source file(s) is downloaded, then hosted on the local server.

Copying…

  1. provides for adaptation or modification (if the license allows) of:
    • content (cut, insert, remix, extend)
    • presentation (e.g. surface design)
    • interactions
  2. supports localization
  3. captures and preserves a version that may be discarded or replaced in the future
  4. allows designers to produce seamless learning experiences that support learner focus
  5. respects original source host’s resources (storage & bandwith)
  6. ensures technical availability of the resource is within local control (no dead links)
  7. allows contextual indexing for site (or public) search engines
  8. may improve reach and increase circulation of source information
  9. may thereby enlarge original author’s prominence and visibility
  10. avoids problems with licenses or terms of use that restrict leech-linking

A couple notable obstacles to copying:

  • Server-generated content, markup, interactions, or hyperlinks may be difficult to acquire or reuse (e.g.
  • While CC By-ND allows reproduction of works, it may restrict modification of presentation or interactions in addition to the more clear prohibition on modification of content

Dynamic Scraping and Importing

There are other approaches that fall somewhere in between. For instance, web scraping of the source file(s) on the fly, followed by parsing and processing of the data on the local host. This sounds complex, but it’s not too bad; Google Docs & Spreadsheets has implemented this functionality into it’s data importing spreadsheet formulae:

  • =importHTML grabs the content of a TABLE or list (OL / UL [/DL?])
  • =importXML uses xPath expressions to target XML/XHTML elements
  • =importData takes structured data files, such as comma separated values (CSV)
  • =GoogleReader intakes the RSS or Atom of a target URL, such as a blog post

Often used for mash-ups, this approach can also be useful for replicating and formatting data. And, though Tony Hirst has found numerous exemplary applications for this feature using Google Spreadsheets, a Google Spreadsheet is not required; anyone with some significant Javascript experience could tackle this task, and there are a number of web scraping software apps that deliver varying results.

5 Responses to “Early Decisions on Reuse of OER: Copy or Link?”

  1. Tony Hirst Says:

    I’ve often thought about the link or embed issue too, and I think there is a third option – to use a lightbox (shadowbox) to pop up the linked to resource in an overlay on the current page.

    e.g. http://ouseful.open.ac.uk/blogarchive/007632.html and http://ouseful.open.ac.uk/blogarchive/014014.html

    The rationale behind this approach is that it:
    1) allows the reader to see the content in the current context;
    2) whilst retaining the original context in the background.

    If you just embed/transclude content you are forcing the reader to engage with it inline/in the flow; if you use a lightbox, it’s a bit like a box in a print document -it takes you out of the original narrative flow but reminds you why you are viewing that document by retaining the original context (the reason for visiting the popped up/lightbox site) in the lower layer. It also retains a linear reading line. The link lets you open in the same window, open in another tab, open in another window or bookmark for later, which changes the dynamic, I think? Ie there are now multiple open documents that you have to read.

    As far as the publishers of the linked to content go, if you pull the content from their site (as you do with a link, transclude, or lightbox), they get traffic stats and maybe page referrer information too.

  2. Tony Hirst Says:

    For an example of how to embed a complete oer course in a page using a Grazr widget see eg http://openlearnigg.corank.com/tech/story/an-introduction-to-information-securit which shows an OpenLearn course from the OU embedded in a page (hundreds of other examples of openlearn widgets are available on the same site).

    If the content you want to refer to/copy/link to is on a wordpress blog, you can easily get the single page rss and then render or embed that – eg see the PS to http://ouseful.wordpress.com/2009/02/02/single-item-rss-feeds-from-wordpress-blogs/

    Here’s a bookmarklet that will pop a single WordPress blog page that you are currently viewing into a portable grazr widget:

    javascript:window.location=”http://grazr.com/gzpanel.html?pl=ou&exp=1&file=”+window.location+”?feed=rss2&withoutcomments=1″

    (The pl=ou includes demo OU branding; remove it for your own bookmarklet, or chat to the Grazr folks about getting your own Grazr skin…)

  3. Mr. Jared Stein Says:

    Very cool, Tony; thanks for posting.

    You’ve given a lot to think about in terms of alternate approaches. I’m going to take some time today to read through these posts and reply or follow-up later.

    A couple issues off the top of my head:
    * concern over mobile device performance
    * concern over bad browser performance (aka JS-disabled)

    Also, I had thought last night about an XSL for custom-rendering of XML-written full feeds taken from external sources, mixed in with “staple” XML from local hosts for local branding and navigation. I suspect this will require a special plug-in on the server side.

    Regardless, it’s another approach that may be worth looking into in instances where the XML is predictable (semantically correct XHTML is a viable target, but may not be consistent enough.

  4. Tony Hirst Says:

    I haven’t actually given much thought to accessibility/JS; i’m still trying to identify the “wiring patterns” that are possible and seeing what they support; eg the grazr widget could equally well be a server side feed2html service?

    Re the xml – interesting; have you been following @patrickgmj’s stuff – eg http://semantic.umwblogs.org/2009/03/10/whose-link-is-it-anyway/

    My dabblings with XML in OpenLearn context are partially recorded here, but i think everything has rotted since i posted?!

    http://ouseful.open.ac.uk/blogarchive/010124.html
    http://ouseful.open.ac.uk/blogarchive/013035.html

  5. Mr. Jared Stein Says:

    I have a penchant for testing accessibility particularly for mobile devices, though that might be putting the cart before the horse (that is, presuming that the demand for OER content on mobile devices may not outpace the development of mobile browser support for these technologies).

    More good links for me to read through! I can’t keep up with you, but this is definitely appreciated nonetheless.

Post a comment

Anti-Spam Protection by WP-SpamFree