December 17, 2017

Learning frameworks


It's funny how we all associate Google with learning, rather than just searching. For specific information, I can see why: a single result for Oxfordian theory, for example, satisfies my needs pretty well. Not bad for a time investment of a few seconds.

But searching and learning are not one and the same. Consider what happens when you search for a broader topic. The word Shakespeare yields 52 million random and mainly unhelpful results. After hours of related search I'm not learning much, relative to the time I put in. I discover that Project Gutenberg has the plays for download, but its messy and I find a book or DVD much more convenient for the actual plays. Meanwhile, the Merlot  collection has similar offerings.  Clearly, we are still some ways off manageable OER resources, and I find no systematic way to tackle my subject. For sure there are some good articles on Google, but it's all random (and all text) and I get the feeling, again, that a decent book might serve me better.

Why haven't more people tried to create learning frameworks for the disciplines online? Google search results do not coalesce in any sense. By contrast, a book will pull the information into context, even if it is arbitrary or limited. Online, I can store stuff in a PLE, but that is still just a loose collection, rather than a framework per se. And while text results were at least plentiful on Google, the audio and video results were abysmal (essentially nothing for 'Richard 2nd', for example and even the most popular plays.)

Try another search

With my 2nd search - learn English - things really went downhill: 50 million results, but all random stuff, of mostly poor quality. Much web-based language learning content is out of date, compiled by hobbyists who patched sites together over time with no notion of networked learning, instructional design, or even the tools of social media, and still less any notion of how to present it. The visitor pays the price, in the form of time, particularly since it is difficult for a non-expert to distinguish the good content from bad. Learning English on the basis of of serendipity is not a learning strategy.

Google simply takes us to whatever is out there - bits of information. Very little of that was even designed for the web,  but  simply migrated there from elsewhere. Learning a discipline, however, takes more than just data, which is why people still go to night classes and buy textbooks. Efficiency, time, and focus are all hugely important issues. A framework ties this  together and offers as sense of direction.

For some, the framework might involve human guidance, while for others it could simply mean a book, or a schedule at a night class. For us, the solution has been a platform that brings together the people, content, system necessary to enable a sustained learning endeavor. We try to give learners enough guidance to set them on their way but enough freedom to choose their own lessons and actively create their own context.  (Learners really should make their own decisions.)

More specifically:

  • At the people level: this includes advice/guidance from practitioners and other learners. Aggregating the experience of practitioners and successful learners can motivate, save vast amounts of time, and accelerate learning. In a sense the people are part of the framework. Pretty soon you have a community of practice  on your hands.
  • On a systems level: the framework offers searchable, accessible learning objects that are tagged and organized along a number of possible learning objectives.
  • At the content level: lessons are designed for relevance, engagement, etc. These learning objects form a repository of lessons that can be tied together at will.

Our objective was to enable the learner to hit the ground running and create real value. With a clear social object, the community that grows around it adds to the context of learning, and is part of the framework.  I discussed this issue with the brilliant Vincent Wade from Trinity College, Dublin last week, when I met him at Elliot Masie's Learning 2008 Conference, in Orlando.  There was much fruitful discussion and I believe there will be more.

Ken Carroll

If you liked this post...
sign up here for free email updates and special offers


  1. Martin says:

    The evolution of language learning in the internet is really fascinating and I’m very impressed by what PraxisLanguage has achieved in this area. However, the longer I learn with SpanishPod, the more I’m missing Wiki-like collaborative features that would allow users (at least the more experienced users) to modify the content or to add new content. Typos could be fixed much more quickly, transcripts of all-Spanish shows could be edited collaboratively, additional exercises could be created, etc.

    Is there a need for these collaborative features? Would enough users actually use them? For the answers, just look at what’s happening at SpanishPod. I think the question is not whether language learning in the internet in general – and PraxisLanguage in particular – will integrate these collaborative features but when.

  2. Ken, I have to say I don’t associate Google with learning. You’re right: it’s about searching. I used to teach basic computer use to adults, back when the IBM PC was new; I’d explain that the computer was fast as hell, but dumb as a rock.

    Google’s not quite that dumb, but as you point out, it’s not too helpful if you can’t construct a search that meets your needs and harnesses Google’s fast but relatively unthinking indexing.

    A textbook, or any type of book, is as you suggest a kind of framework. In fact, the advantage of a book is almost the converse of a Google search: a good book is highly focused, reflecting a framework of the author or editor.

    In thinking about communities of practice, and the larger communities into which they fit, I realize that all along we make decisions about what’s useful and not, who’s helpful and not, based on our own experience.

    Fred in engineering may be a bore, but he’s a well-informed bore, and if you want to know about the client database, he’s the guy to consult. Lucille in the Wichita office is wildly eager to help but rarely goes below the surface — so you’ll check with her to get someone’s name or a location for data, but not for anything in depth. And Carl, who posts often in the C of P forum — Carl just seems not to have much to do.

    Someone said once that data is like hay: it’s very dry, and it comes in stacks. Information is like a needle: it’s got a point, but it can be hard to find in a stack of data.

  3. Dave Superman says:

    Ken, love the blog and as a self-confessed “exploratory learner” I related to this article. I feel that technology trends are moving towards a solution to this search problem very quickly.

    The atomisation of online content (breaking up content into pieces), combined with advances in tagging conventions/systems, is set to primed to be the basis of “learning-centric” search engines.

    There is the so-called semantic web which can pull these atomised content chunks into (learning) insights. Take for example, the learning potential of a semantic search app like

    Then, there is the search engine itself. Take human powered search engines (e.g. Even tags and even links are search tools too.

    In general, integration of search with content has the potential to transform any online object into a learning object. Hyperlinks throughout wikipedia is a basic example, but for something more revolutionary check out photosynth (on on their website).

    The extension of exploratory learning would be that it starts anywhere, not just in Google. In fact, this infrastructure is harder to monetise so don’t necessary look to Google at all!

    From your perspective, how will the PLS leverage these new trends and technologies (e.g. sementic web and exploratory search objects)?

    Dave Superman

  4. standuke says:

    While ChinesePod creates a great framework for personalized learning, I’m with Martin in that it still seems to fall short in terms of providing a framework for community.

    In some ways the customizability of the ChinesePod site reproduces some of the sins of Google. You wind up with lists rather than a stable framework. My ‘personal’ page changes frequently and is unique to me–I can’t count on others seeing the same thing. The ‘Lessons’ page turns over weekly, and with it go the threads (usually I haven’t listed to the Podcast yet). The ‘Community’ page defaults to listing the most recent few posts (temporal context, as opposed to topics–a very odd choice considering how many conversations take place across time zones). Once you apply a filter to the conversations you again have a rather ‘personalized’ view of what is happening in the community. I find myself asking ‘Should I comment on this thread? Will anyone notice? Will other users apply the same search to find this thread and stumble on my comment?’

    The Forums portion of ChinesePod, on the other hand, suffers from hyper-organization and isn’t really integrated into the rest of the site. The subforums are too sparsely populated to encourage regular visitors, and their multi-tiered structure makes it hard to browse through the subforums where there has been recent activity.

    If I were King of ChinesePod I would try to extend the ‘Channels’ paradigm. I’d move it to the ‘Community’ page and each ‘channel’ would include virtual classrooms in which users at more or less the same level can interact in a ‘stable’ virtual environment. Basically there would be Newbie, Elementary, Intermediate, UI, Advanced forums. Maybe each ‘classroom’ would feature the current lesson, a 4-6 week old review lesson and a ‘rerun’, just to give people a little extra something to talk about. Users should be able to view most recent 10-20 threads (not posts). From what I’ve seen on other sites, a forum is the ‘right’ size if >5-10 people are posting on a daily basis, but not so busy that threads are pushed off the bottom of the thread list (and into cybernetic oblivion) in less than 24 hours with no responses. Obviously the comments thread for the current lesson would be accessible on this page, but it would be alongside other threads that may or may not fade away along with the current lesson. I think this format would promote more thoughtful, long term discussions on many lessons, while encouraging subsets of users to pursue other topics.

    Just my 2 cents :>)


  5. Henning says:

    Hi Ken,
    interesting thoughts.

    I know the term “framework” primarily from qualitative / explorative research where it comes in the flavour of a “conceptual reserach framework”.

    A “conceptual research framework” is a tool that broadly maps out the basic structure you regard as valid and relevant with respect to the research subject. It depicts what concepts and interrelations are considered to be both basal and worthwile to delve deeper into during research. Such a framework functions like a sketch of a world map guiding your journey of knowledge gathering. Exploratory research without a conceptual framework easily deviates from the intended path and fosters an arbitrary and chaotic gathering of information. It is noteworthy that conceptual reserach frameworks can be flexibily expanded or modified or even completely rewritten during the course of gaining deeper understanding.

    Transplanted into the domain of language learning, this would mean that the “learning framework” delineates your personal broad view on the language. In my understanding it would show that limited number of core aspects that you deem to be most relevant. I think it is furthermore imperative to keep the elements of the framework indeed on a structural “meta level” – the framework is different from the concrete content it actually is supposed to frame. “Grammar patterns” might be a part of a learning framework, “function words”, or “phonetic compontents”, or (more content driven) “Expressions commonly used in news items”. And of course over time the learning framework also undergoes fundamental transformations – it needs to with the level of understanding you built up and the changes of focus you apply.

    For me, the subjectivity of the framework forbids it to completely take it from a text book, a learning site, or even a teacher. It might be helpful to get a framework in the very beginning to take an initial starting framework from such sources and it is definately worth to look at templates for frameworks – e.g. from other users (goulniky, xiaohu, etc.), from other learning resources (the HSK follows an implicit framework and so does Chinesepod).

    But those are templates and not the real thing. You have to built that yourself.

    So, to come back to the question: I would understand the level-based framework you presented to be more a “systems framework” than the actual “learning framework”. Whether or not both match depends on the user. For me, CPod provides contents that I need to fill out some of the boxes of my framework – but not all yet, as CPod is still on the road to becoming a “one stop” Chinese provider. Also relevant for me is that currently some of the bigger boxes are filled with user generated content.

    This is highly relevant for me, as I am currently in a “reconstruction” effort for my framework (which I also need to efficiently structure my limited learning slot) – which comes parallel to a temporary reduction of my Chinese learning activities (work related) and the insight that my old approach got somewhat stale…

  6. Hi Ken,
    Love your blog.
    I met you on Olympic inauguration while Jennny was guiding at Praxis but you might not remember.
    I love learning with the mood at ChinesePod as well practice with native Chinese whenever I can.

    If you’re interested in self learning, I came across a book “Pragmatic Thinking and Learning” by Andy Hunt. One of the challenge with the technology is the constantly learning approach. Insightful contents :
    – the Dreyfus model: a model per skill.
    – the cognitives in our brain and how to effectively use them.
    – tips: mind map, morning pages technique which are now my source of ideas.

    I have my own mental framework for languages. Like programming, structure and pattern matching (grammar) are fundamental. Then vocabulary can be assimilated with time.

    As your new year resolution, you should post shorter.


  1. […] Ken Carroll » Blog Archive » Learning frameworks (tags: elpc3 network) […]

Speak Your Mind