Discovering OER re-use and derivative works using DOB codes
Introducing “DOB codes” (date of Birth codes) to help link similar or related OER works, or to see the ‘family tree’ of resources.
It would be most useful to potential users of an Open Educational Resource to know about any similar works, or to see the ‘family tree’ of the resource, just by clicking a link. Re-users or creators of derivative works would benefit themselves, new users and ancestral authors by continuing this linkage as they evolve the material.
This can be achieved by adding a simple code to the copyright citation (See below). In addition to the benefit of helping connect released and re-released works to ancestor works and subsequent derivative offspring, this approach could be extended to provide a publicly available family tree of all related works and help checking the application of open licensing is historically consistent. While a little different from more conventional methods employed by librairians etc. it is a pragmatic method based on the affordances of the Web.
Each Resource should be allocated a code on the following format, (like a triple code but simplified):
man: An (optional) three letter name, no vocabulary, could be “oer”, “Joe”, “man” (Manchester), oxf (Oxford) etc. If not used remove the colon that comes after it (a-z, 0-9, A-Z)
240265: the date of the creation or release of the OER in the form DD/MM/YY
rh4xs: a unique 5-letter code, allocated by the project. (a-z, 0-9) details on how to generate a new unique code below.
This code would be added to the copyright citation,e.g;
(c) Poppleton University (ppl:240265:rh4xs)
Though not guaranteed, propagation of the code becomes a simple part of the legal obligation of the user or re-user of the material under all the Creative Commons (CC) licenses, thus linking all works incorporating parts or all of the original work together.
A lot of coding protocols such as this incorporate a url, e.g. http;//notveryprettyandlongandtechie.com/resolver and when the link is clicked details come up. Due to the penetration and ubiquity of Google’s search tool this is dispensed with. (Particularly as Open Resources should be discoverable by normal means to be properly “open”) . By adding a Google search URL into the original OER where possible, perhaps on the legal disclaimer page or the introductory text in the case of documents e.g
a click on that link will bring up all visible online linked and derivative works as well as metadata if the resource is cited in a open repository, for instance, JorumOPEN.
Similar vs. new derivative works – how the code evolves
Its not terribly important the code evolves at all. It is sufficient that derived materials just cite any codes from other materials used because of the legal obligation of the re-user to do so. Linking all works incorporating other works is then straightforward. The judgement on whether a derivative work is a significant departure from the original work(s) to be a new treatment cannot be policed once “out in the wild”, and any attempt to do so would soon “choke” this approach. However if this is decided the following happens:
Any major new work, whether it contains other original works, (and thus still incorporates the codes) or is a new treatment or merely related to other works by subject, can be linked to another existing work (and thus its ancestors) by deriving a new code from one of the existing ones. To get a new code, make a new 3-letter name and date part of the code, keeping the unique code at the end and adding the whole new code to the copyright citation of the new work.
Adding an explanation to a document:
To support this idea the “legal page” of each resource could include:
“It might help other users to find your work if you reference others. If the copyright of related material contains a code, e.g. man:240205:rh4xs keep the last bit of the code, change the date part to your release date (dd/mm/yy), then choose your own 3-letter code at the start. Then add it to your copyright declaration.”
Or a more complete example that I have used:
To refer to or reuse parts of this work please include the following copyright notice. (The only exception is if you intend to only reuse a part of the work with its own specific copyright notice, in which case cite that.)
If you create a new piece of work based in part on this one, it will help other users to find your work if you modify and reuse this copyright notice. Keep the last bit of the code after the last colon, change the date part to your creation date (ddmmyy), then choose your own 3-letter code (e.g your initials or an organisation code) at the start. Make the new code your copyright declaration or add it to an existing one.
If you create a new piece of work or do not wish to link a new work with any existing materials contained within, a new code should be created. Choose your own 3-letter code, add the creation date and search as below on Google with a plus sign prepended: +tom:030504
If nothing comes back citing this code then add a new 5-letter code of your choice to the end, e.g; :01Lex Do a final check search for the whole code. If the search returns a positive result, make up a new 5-letter code and try again. Make the new code your copyright declaration or add it to an existing one.
How to generate a new unique code for new OERs:
If an OER is new or the author doesnt wish to link a new work with any existing materials contained within, a new code should be created. Choose your own 3-letter code add the anticipated release date and search as below on Google with a plus sign prepended:
if nothing comes back citing this code then create a new 5-letter code of your choice, e.g; “a0000” and append it to your resourse. If the search returns a positive result, either make up a new 5-letter code and check for that appended to the previous search, or start again.
It is just possible that someone else may choose the same three letter name, on the same date and make up the same code. There is also a chance that as Google cannot record everything currently on the web immediately an identical code may be missed by a search. The chances of this are very small and two resources having the same code would not cause a critical failure of the system.
Although a straightforward Google search will allow the history of a resource to be derived with a bit of work from the user, further development could include a customised Google search to provide an easy checking service to show the entire propagation and online usage of a resource – in other words, the resources family tree – check cascaded CC licensing is historically consistent, inform the provenance of a work including a copyright recording index or even a “Facebook” style linking service for OERs.
Using Google as an ad-hoc resolver service makes sense and simplifies the service considerably. Google also has the biggest reach into the web so it’s the right tool for surfacing linked works. However a more organised search result could be provided by using a Google sponsored link to highlight an “offical” source of metadata for any search that incorporates a code.
keywords: ukoer, oer, engscoer, dobcodes, dob codes, cc , creative commons