By Gordon Freedman, President, National Laboratory for Education Transformation.
The purpose of this post is to tease apart the Open Education Resources movement (OER, globally available knowledge in free form), for its own sake, from the intent to improve education and personalize learning (structured education and structured personal learning). While the former is assured by now, the latter has not really begun. New and better teaching and learning models are desperately needed in K12 and higher ed, and OER can be and should be playing a major role. But to do so, the field of learning design, content management and personal analytics needs to embrace better processes for OER creation, use, tracking and re-use. This transformation will need to start with the application of data science to learning content, personal use and education outcomes.
The Leap to OER, Sweet 16
When the OER movement started, it had all the promise of a makeover for the world of learning in an open and free way, like what open source code brought to the world of technology development. Since its inception, OER has taken on an almost global religiosity about its intent and purpose, and its opposition to the for-profit textbook and academic journal publishers. Due to that zeal, some very good and truly great things have happened by freeing knowledge and learning content, shared around the world, unmoored from its traditional anchors in academia and commercial publishing. In many ways, the promise of OER stopped at the doorstep of making a structural difference in education (although it did radically transform academic journal publishing in a quiet and profound way).
There is no question that generous seed-funding and the foresight of the William and Flora Hewlett Foundation started OER down the path of making knowledge free and ubiquitous globally. From the Foundation’s kick start in 2002, OER has not let up. Also, in 2002, MIT boldly launched its Open Courseware (OCW) initiative, turning loose its entire faculty’s instructional content to the world. The OCW historic first webpage was an act of higher education machismo. It set the academic bar and issued a clarion call to other institutions.
The other critical lynchpin for OER, also in 2002, was the issuance of the first Creative Commons (CC) open licenses. Like OER itself, these licenses followed the General Public License (GNU) of the Free Software Foundation, and led to a parallel licensing universe for open content. If MIT and Hewlett were the battering ram against the publishing establishment, CC licenses were the tips of the intellectual property (IP) spear.
Stanford faculty member Lawrence Lessig’s tireless efforts to establish a framework for the Creative Commons opened the way for governance in OER and license types for building, or not, on previous works. Again, Hewlett Foundation funding at a sufficient level lubricated the machinery of change.
The Hewlett Foundation, the Melon Foundation, MIT, and the Creative Commons provided a formidable intellectual property challenge to textbooks and academic publishing. Similarly, the robust efforts of the UK Open University’s offering, Open Learn, grew from its beginnings in partnership with the BBC. In 2002 as well, UNESCO held the 1st Global OER Forum, from whence the acronym “OER” was formally adopted.
Where Are We Today 16 Years Later?
What has taken place over the last 16 years is the establishment of a global movement for free learning materials that manifests itself in practically every country on Earth. Open knowledge is flowing, albeit in a self-organizing way. There are many open OER repositories, products, and services that utilize OER, and even the most commercial textbook and academic journal publishers incorporate OER materials. Similarly, a whole field of open textbooks, open learning modules, and OER resource sites abound. And the stepchild of OER, MOOCs, are relatively alive and well as intentionally designed and developed massive open online courses. Now a global standard for open learning, MOOCs have seen impressive commercial, university, and nonprofit investment.
For the small group of players who formed the movement and are still active, it is a rewarding accomplishment to see OER in use worldwide. However, for those devoted to the replacement of textbooks and lowering the costs of learning materials, the OER promise isn’t as bright. Also, there exists a tension between freely putting any type of materials, at varying quality levels, into the public domain—the religion, and those who look to the efficacy of these materials and the evolution of true high-quality learning goods and their effects—the science.
Here, the movement lags behind other OER categories and the publishers. Outside a few well-funded efforts, OER materials are generally not tagged, tracked, or part of an academic community effort to accelerate the quality of education and learning, or even establish a systematic ability to search and retrieve the best of OER globally. There is no global OER repository search, though there have been a number of attempts.
By contrast, in their closed world, publishers move forward with little transparency managing the market testing, development, editorial, marketing, and publishing very carefully. They are matching well developed content to very specific needs in the market. They are experts at this.
OER, to date, is not. There are either a very small number of well-developed implementations or there is open content randomly produced. For education purposes, the publisher developed content may have restrictions on its use and the more random OER content is hit or miss in finding it and determining its quality, utility and past history. So, on one hand you have the lockbox of publisher content and use data, and on the other, you have an array of millions of bits and pieces in the OER ocean.
In both cases, education loses. Education cannot get access to the data locked away in publisher content and assessment use-data. And it can’t make wide use of OER because it is not tied into a system like the publishers’, where quality, authorship, editorial, marketing, and publishing processes know how to find what is most appropriate, apply it for the best effect, store the results, and make it easy to re-use. Even the staunchest proponents of education OER, must hunt and peck to get materials assembled for their own classes. Maybe this is okay, but it does not scale where it is needed most.
There is a battle to produce more and better educated people. This battle needs all the help it can get. If it is not coming from traditional publishing or from the OER movement, where can we can that scalable help come from?
What About the Rest of the Content in the World?
In the 16 years since OER sprang into action, the theory and science of content deployment on the open web, in closed sites and in apps, has advanced as fast as any existing technology-data-processing field. Astoundingly, the five largest companies in the world by market value are Apple, Amazon, Alphabet, Microsoft, and Facebook. These are largely companies that have services that are tied together using content managed by data science. The same is true of YouTube, Twitter, Instagram, Snapchat, and the list goes on.
How is it conceivable that these advances haven’t been visited on the OER community?
The Hewlett Foundation and Stanford, literally a stone’s throw from each other, are the beneficiaries and the generators of a great deal of Silicon Valley success. MIT is arguably at the pinnacle of engineering institutions in the world and is committed deeply to education. Yet, the keys to content kingdom – natural language processing (NLP), machine learning (ML), and artificial intelligence (AI) – which are the brain children of the best universities or university educations, haven’t picked up the reigns of sorting through what learning resources belong with what education operations for what students toward what careers.
Expedia can knit together pathways for travel from multiple databases in seconds or Zillow can simplify locating a house to buy as it tracks all the houses for sale in the country and their prices and their likely sale prices. This type of data and content work is no longer rocket science.
OER, Where Are You?
OER is not alone in not having moved ahead with precision. The education publishers are not in this brave new world yet either. They don’t have to be because their education publishing models are still opaque and secure from change, largely because OER is a huge, disorganized cottage industry, unlikely to displace the central publishers. The caution in this is that the academic journals business did upset the applecart. And the reigning success of general OER as ubiquitous knowledge distribution globally is now evenly spread across developed and developing nations.
In education OER, there are two shining examples of OER at its best in the education world. OpenStax from Rice University in higher education and the Khan Academy in K12. Each is a gold standard within its domain. OpenStax produces open textbooks that rely on the same diligence of commercial publishers. They can do so because of the ample support from the Hewlett Foundation, the Arnold Foundation, the Gates Foundation, and others. The business model is not yet here for open textbooks. It may never be, and that might be okay. This is subsidized publishing.
The Khan Academy evolved from Sal Khan working out of a closet in his apartment (I took my son to visit him there in the early days). His global operation and K12-focused US efforts are thanks to generous funding from the Gates Foundation and gifts from Silicon Valley icons John Doer and Reed Hastings, who also gives to OpenStax.
These models are great, but they are a long way from being replacements for publishers. They are not the open web of knowledge automatically offering up a way for the bits and pieces to find each other for specific applied purposes or specific learners or jobseekers, at scale.
What Next, OER?
All this is possible, but it requires a second wave of OER research and development. If the creators and funders of OER want to change education at scale in an equitable and effective way, not just spawn learning resources generally and park them repositories not known to most educators, it will require a marrying of what Silicon Valley has wrought with what educators and learners need most.
Gordon Freedman is founder and president of the National Laboratory for Education Transformation, a research and development nonprofit dedicated to creating “processing parity” between modern social, consumer, commercial, and financial solutions and apps, and those that are used in education, training, and learning. Freedman was part of the beginning of online learning, education technology, and open content, and in preparing for this post went back and talked to many of pioneers in OER, education technology and the traditional publishers.
NLET is working with data scientists, advance search firms and higher education institutions in experiments aimed at turning learning content into data that can match up learners, content and labor opportunities.