Learning Tech To-Do List

A lot of us are working on projects that support the idea that learning is everywhere and want to empower learners to access high-quality experiences that match their interests. We work because we know our lives as learners are bigger than any single institution, or location, or subject, or age. We work because we know that learning is the formation of a life, not merely the acquisition of knowledge. Long before me, Henry David Thoreau talked about this idea in better terms:

“We are all schoolmasters and our schoolhouse is the universe.”

Some exciting visions for the future of learning have been cast, but there is still so much foundational work to do. In order for connected learning to connect in technical practice as well as in theory — in order to build the universal schoolhouse and empower all learners as their own schoolmasters — there are 10 technological challenges we need to work on.

We are committed to working on this list for the foreseeable future and would love to work with anyone else who is doing the same. It is time to move beyond our aspirational rhetoric and inspiring videos and do the hard work to build a technical foundation that actually connects learning.

Below are the 10 things we need to do, with accompanying visuals and examples of those projects that have made some progress.

#1: Distributed Knowledge Creation

At the beginning of the learning cycle, we need tools that enable us to conduct distributed but networked knowledge creation. Generally speaking, we currently either have large silos of internally connected learning resources (e.g. Coursera, Lynda.com) or individual resources or topical platforms disconnected from any system.

Ideally, a new sort of platform would bring together ideation, publishing, sharing, remixing, and collecting into a single interoperable framework, distributed but connected across the web. This type of technology would enable knowledge creators — creators of ideas and learning resources — to work outside the confines of a monolithic LMS (Learning Management System), but still distribute their work at scale and connect with a larger community of practice.

We need something like the Federated Wiki, except prettier.

In a Wired article from 2012, the inventor of the original Wiki, Ward Cunningham, notes that he “feels bad” about his creation and its inefficient, anti-creative centralization. The traditional Wiki, the article notes, “forces you to give up your own perspective” … “There are issues that no one will agree on, but with the federated wiki model, everyone can have their own version of controversial pages. ‘And they’re all linked together, so you can still explore them like a wiki.’ ”

With this new infrastructure spearheaded by Ward and others, ideas can be developed independently and then forked, shared, merged, and built upon by others. This platform for distributed knowledge creation combines the scalability and impact of a central platform with the freeing creativity of independent publishing.

The FedWiki project is still in its nascency and, as the Wired author questions is maybe “too nerdy to catch on?” There’s a lot of work to be done to push the idea further, improve the visual design, and make it easier to create your own instance. If FedWiki married Medium.com and was then improved and employed for the creation of learning resources, we’d be on our way to something really amazing.

#2: Learning Resource Clearinghouse

While networked creation tools will help to connect some of millions of great digital resources, we must build tools that help aggregate all resources across both local and digital environments. Of first importance is the creation of regional data-stores that operate local learning opportunity clearinghouses. The entire “socialstructure” of learning is difficult to capture and make machine-readable, but offers us a wealth of resources that are not currently organized.

These local collections could publish APIs that allow any publisher to re-share a subsection of the collection based on their audience or area of interest. For example, instead of working alone a project like the Stemisphere here in Pittsburgh could aggregate STEM-related resources, add them to the clearinghouse, and then pull out an even broader collection to showcase to their constituents. Stemisphere benefits from having a more comprehensive and consistent collection (due to the other parties collaborating on the collection) and the Clearinghouse benefits from Stemisphere’s topical expertise and regional relationships.

If we have one open data store to rule them all, learning providers would save a tremendous amount of resources just by the reduction of the administrative time currently required to post their events across numerous disparate platforms.

As part of our current learning laboratory portfolio at the Sprout Fund, we are prototyping one open-source tool that tries to solve these data collection problems with a specific focus on lifelong learning resources for adults.

#3: Resource Curation Tools

Once we aggregate the masses of resources, we will need increasingly powerful tools for the sorting and highlighting of learning resources. Most importantly, this curation must engage not just with digital resources, but the complicated wealth of all local, in-person opportunities. In order to make the scope of resources manageable, the best resources need to rise above the rest to eliminate the Craigslist problem. To do this we need to raise up the ancient role of librarians by developing curative tools that help us build lighthouses amidst the flood of information.

One current project in this vein is Bento.io, a volunteer part-time endeavor by Stack Overflow developer Jon Chan. Built on a custom layout by developed by Jon, Bento represents quite possibly the best learning resource curation for web development resources on the entire web. Remember, this is one person volunteering in his spare time. Imagine if we invested in great tools and more comprehensive curation across all subjects!

#4 Learning Resource Visualizations

Once the resources are all collected and curated, we need to share them with learners. Beyond the obvious formats for search and hierarchical browsing, we need a user-interface that promotes unintentional knowledge. One foundational problem of learning technologies is that you cannot search for what you do not know or have never considered. You do not have the words and do not even fully know what you do not know.

History is rife with examples of knowledge maps and our best video games have intricate skill trees to highlight progress and opportunity. It is time that we had a user-interface that enables an immersive visual exploration of learning. We need to see where we have been and chart a course for where we want to go. Let’s make something better than a list.

To begin this effort, we have started another project as part of our learning laboratory portfolio: an early prototype of dynamically generated learning maps that can showcase connections and topical affinities. It’s been a very incremental progress, but we are headed somewhere interesting.

#5 Learning Pathway Technology

While visuals will help us begin to express the structure of learning, both what we have done and what we hope to do, we still lack the basic foundations of that structure. In order to construct pathways — and relationships between resources in general — we need an interoperable, machine-and-human-readable specification. What do connections between learning resources mean? What do they look like? What are the lower and upper bounds of this expression?

While learning is non-linear and dynamic, there are useful non-obligatory pre-requisite structures that answer the question “What should I do next?” These kinds of relational models will help to build a network of resources that can express both predefined and emergent learning pathways. Pathway technology will empower learners and educators alike by providing powerful way-finding amongst the wealth of opportunities.

A small group of us (in one remote circle of the learning-tech world) are just beginning to define a linked data structure that can provide this definition and usability for learning pathways. While the “learning pathways” idea has been in full force over the last two years, it still lacks any technical reality. We still have difficult questions to answer. For instance, what are the nodes or steps along the pathway? Are they competencies? Credentials? Experiences? Resources? And how do we account for difficulty variabilities?

#6 Digital Credentials

To show our accomplishments along these pathways, we need robust digital credentials. Given the amount of recent attention projects like Open Badges have received, one might think this “to-do” was done. Far from it. We are still on version 1.1. Open Badges exist, but the standard is still very much in its nascency and has yet to fully support the visions that are often discussed in relation to their use.

The anatomy of badges (and the work of digital credentials more generally) has a lot of evolution left to do in order to: make them truly stackable; enable remixing, sharing, and endorsement; map them to standards and competencies; embed them with evidence for use in a portfolio; provide version control; etc.

A powerful digital credential standard is inevitable, but will we build one that empowers learners and distributed learning networks or one that only meets the needs of institutions and existing Learning Management Systems?

#7 Learner Identities

Credentials, and learning data in general, are only valuable if they are verified and organized around your identity, not scattered about the internet. Right now I have details about my progress scattered about the world: college transcripts, high school grades and diplomas, portfolios of work, awards, workshop certificates, DuoLingo stats, Khan Academy achievements, Coursera work, professional development experience, a Degreed profile, etc.

In the Open Badges world we have the “Badge Backpack”, but the reality of learning technologies is that everyone has 100 “backpacks” strewn about in isolation. In the past, we talked about ideas like Federation in which all the backpacks would connect. The reality is that we need a single service that empowers learners to have a common identity and data-store across learning environments.

We need an infrastructure that can represent a single learner identity across 100s of local and digital environments. A single identity controlled by the learner that they take with them to each and every learning experience.

Though not directly related to learning, projects like OneName and OpenID are working on things in this vein. As was the now de-prioritized Mozilla Persona initiative. Local endeavors, like the the Municipal ID project being undertaken here in Pittsburgh have relevance to this work as well and will have an important role in how we bridge the local and digital contexts.

#8 Seamless Interoperability

Once my identity is sorted out and we can be ourselves across learning places, we must work to ensure that all these places can share information. By asking for properly scoped permissions when you sign in, platforms should be able to read out and translate your learning data to customize your experience.

If we are pursuing a distributed learning program (maybe part of a pathway), we need to go from Khan Academy to Coursera to a Mobile App to a local workshop and back again and have each place understand our progress.

The folks behind the TinCanAPI or xAPI projects have been working on a simple “Noun, Verb, Object” data structure that could work to capture any type of learning experience. While still maturing, it represents a beginning towards a very difficult problem.

#9 Lifelong Learning Archives

The end result of all this should be a comprehensive portfolio. An archive of everything you have ever learned. This archive would provide permanent storage for your learning data and an API service that allows you to control and use it across the web.

User settings on your dashboard could include a default option to “share data for research” to allow for important meta-analysis of learning so that future learners can see the successful pathways taken before them. Your archive, or parts of it, could be shared or displayed at your discretion on other platforms.

As a good example of a powerful experience archive, Github does an amazing job of archiving your accomplishments. In their profile view, you can see all of your recent accomplishments at a high-level and drill down to see the exact detail of each contribution.

#10 Personal Learning Analytics

Utilizing the data from these extensive archives, we can then work on providing powerful personal analytics for learners. Things like goal-setting, progress metrics, and personal dashboards will help us be intentional about the past, present, and future of our learning.

This kinds of learning dashboards would help us be intentional about our “super hero lists” (read/listen to Act 2: Wonder Woman). This everyday “super hero” created a list when she was young of everything she wanted to learn:

“The list included martial arts, electronics, chemistry, metaphysics, hang gliding, helicopter and airplane flying, parachuting, mountain climbing, survival … weaponry, rafting, scuba diving, herbology — yes, I, studied that — CPR, first aid and mountain emergency kind of medicine…”

Due to the 9 other things that are prerequisites for the full realization of this, little exemplary work exists. The Degreed platform is the best example of this so far, but suffers from its manual nature, it’s lack of accurate accounting, and exclusion of local resources.

In a way you would be able to see the development of your brain. We can see how far we have progressed towards our masteries in video games, why not in real life?

Does any of this matter?

Will any of this empower the most struggling students? Will it improve access to quality programs? Is it, in Dewey’s words, serving the primary criterion for a successful student: “a desire for continued growth” and a “capacity for further education?”

I believe that, if done right, learner-centered execution of the above list would result in a set of tools that would empower learners, foster a culture of “yearning for the sea”, reduce costs, and help us to once again see the pursuit of learning as something paired with the pursuit of life.