Cloud learning is a concept being promoted by Judy Breck and plays a critical role in our concept of mobile learning. From one of her blog posts:
It is worth using this analogy in understanding that educational resources online have not yet managed to move much toward the cloud phase. Most online education stuff is embedded in structures like curricula and courses, which are patterns and not miscellaneous. GoldenSwamp.com is dedicated to writing about the cloud of learning resources that surely will be crucial in the learning enlightenment that lies ahead. Patterns are not inherently bad, but nodes that are free to participate in more than one pattern offer a richer learning environment. Cloud learning would be such an environment.
I like the concept, but wonder why she choose a ’swamp’ as her metaphor. Seems rather dark and unappealing, especially compared to a Simpsons-like cloud. More on the topic from her recent presentation at Microlearning Conference 2008:
Judy Breck - Cloud Education (Microlearning Conference 2008) from Teemu Arina on Vimeo.
One comment that particularly resonated was near 27:00 where she recommended un-bundling courses into micro-content to improve their findability. Or, one could just design modular lessons in the first place as we do with our LanguagePod’s.
Further reading:
- Interwingle (free PDF book download)
With the iPhone 3G release, mobile learning has been top-of-mind at Praxis Language. Internally, we have been using Woodill and Cunningham-Reid’s definition as a basis:
True mobile learning is personalized learning that unites the learner’s context with cloud computing using a mobile device.
As we evaluate how to improve our personalized learning system we have been comparing our approach against the above definition and contrasting our approach with other methods to learn foreign languages.
To support this analysis we further fleshed out a few definitions. First, in order to be sensitive to the learner’s context one must take into account their current proficiency level, learning goals, personal interests, and other personal/environmental contextual variables. A learning service needs to be able to be personalized to accomplish this. Specifically, it must have the ability to (i) identify student learning goals, (ii) remix lessons into customized courses that can set students on the pathway to these goals, (iii) provide one-to-one practice opportunities and (iv) facilitate personally-tailored, reinforcement opportunities. Second, ‘cloud computing using a mobile device’ not only represents the ability to access the Internet through an iPod, iPhone, PDA, laptop, etc, but also the social aspect, the community, that goes along with network connectivity.
Textbooks and CD-ROM’s are neither personalized, nor available on a networked device.
Offline classes can offer a degree of personalization, but are not available on a networked mobile device.
Software applications (e.g. iPhone 2.0 app’s), target language media (e.g. Youtube, Youku, etc.), and most language podcasts can be made available on networked mobile devices, but the learning is not personalized.
No existing language learning product currently fits the above description for mobile learning, that is, both personalized and accessible on a networked mobile device.
Customizable courses are the key to breaking down these barriers. Students, or their teachers, should be able to re-mix modular lessons into a course that is specifically designed to solve the student’s individual needs. SafariU is doing this with IT textbooks so why couldn’t the same model - for other subjects - be put online and made available for mobile devices?
The nature of podcasts makes them the ideal ‘plumbing’ to facilitate mobile learning. Being modular by design, podcast lessons could be re-mixed and put into personalized courses. Furthermore, podcasts rely on RSS distribution, which is channel agnostic and can equally distribute learning media to any networked device. Many language podcast publishers have forsaken this opportunity by adopting backward-looking, textbook-emulating linear curriculums.
Imagine a scenario where a student gets a 20-minute needs analysis with a counselor over a phone. The counselor identifies the student’s needs and then re-mixes an appropriate course from an archive of lessons. A family wanting to adopt a Chinese child - no problem. An expat who wants to be able to better work in his Chinese office environment - no problem. The student then starts a daily routine of (i) listening to podcasts on the way to work, (ii) taking 10 minutes in the morning to go onto the website and review the lesson they heard earlier, (iii) practising their new language with a co-worker or a teacher via Skype for 15 minutes at lunch and then (iv) reviewing the vocab they previously saved to the personal accounts by using flashcards on their iPhone on the trip home.
This is the mobile learning vision - that is both personalized and available on a networked mobile device - that our Guided products are trying to achieve.
I am a big fan of Umair Hague and his thinking on new media economics have played an important role in our strategic thinking.
That said, I think it is important to not universally apply Hague’s thinking to all types of media as there are some assumptions in his thinking.
For example, he states in his recent User Generated Context paper:
Most user generated content, is, in fact, context. The bulk of what connected consumers create isn’t content: its context - information about the value of goods and services. Context in turn, lets connected consumers search and navigate the exploding universe of media more effectively, and massively amplifies incentives for quality.
I agree, but this assumes search costs and a degree of modularity in the media content. While an individual article/post makes sense in isolation and is often consumed as a unique entity, this does not happen with podcast lessons where students consume lessons in sequence - even if of their own selection.
As a result, the largest search cost for a student is finding an appropriate service provider and not an individual lesson. Correspondingly therefore, user generated context is more valuable in identifying learning services, rather than the lessons themselves. This can be seen by the dominant preference for third-party bloggers to link to our sites (e.g. http://ChinesePod.com) themselves, rather than individual lessons.
Harold Jarche commenting on Umair’s paper states:
Creating good content on a platform that lets users (teachers & learners) add context may be the the real killer application in education. Content developers and institutions have been so concerned with protecting their content that they don’t see where the real value lies. Letting others add more context will only increase the value of their content.
But who needs this context? I think there is a clear case for teachers who are trying to put together an appropriate course by remixing a variety of learning materials. But for the students themselves? The compelling need is much harder to identify.
The fact is that there is an associated cost to produce high-quality learning content. If there are no ‘amplified-attention incentives’ for individual lessons themselves, do attention-based business models (e.g. advertising) make sense? Just how does letting students add ‘context’ to lessons actually increase the value of the lessons themselves? Therefore, shouldn’t students just be charged for access to high-quality content (subscriptions), thereby justifying future investments in more diverse and better content?
The situation for teachers is much more clear. Teachers should have free access to allow them to add context to the lessons (through comments, rating, etc), with the benefit of helping other teachers discover appropriate lesson content for their students.
There is a good quote from a Mobile Learning report just out from Brandon Hall Research:
True mobile learning is personalized learning that unites the learner’s context with cloud computing using a mobile device.
This is very much in line with our thinking on Personalized Learning Systems here at Praxis Language and offers a framework on a simple acid test for true mobile learning services: 1) it must be personalized and 2) it must be context sensitive (location, age, learning preferences, interests, search history, gender, language and learner achievements) and 3) it is consumed on a network-enabled mobile device.
Over the weekend, I was browsing through the education app’s released in the iPhone app store. They seem to fall into 3 categories: dictionaries, phrasebooks, and flashcards/quizzes.
Dictionary App. - Ultralingua ($29.99)
Phrasebook - Lonely Planet (free)
Capitalizing on Olympics fever, but free.
Phrasebook - iLingo ($9.99)
Simple service, but broad selection of languages.
Phrasebook - Lingolook ($4.99)
Nice graphics, but no audio?
Correction: audio is there, but would an audio icon or some visual response when a phrase is selected be too much to ask for?
Phrasebook - Gorilla (free)
A carry-over from their previous browser-based effort.
Flashcards/Quizes - AccelaStudy ($14.99)
Actually trying to be a learning tool as opposed to a travel tool. Simple functionality, but at least starting to incorporate some of the iPhone input methods (e.g. switching flashcards, turning cards).
All of these are clearly first generation applications. Look for next generations to have the three characteristics that Brandon Hall Research emphasized: personalized, context-sensitive and connected to the network via a mobile device.
I finished two books over the weekend with polarized views on customer involvement in product design.
First, Groundswell (p.194):
We’re not suggesting that a company like Apple turn its development over to its customers - that would be a tragic waste of talent. No, the companies that win by embracing their customers incorporate those suggestions into their own development and process strengths. The customers don’t tell these companies what to do - they just make suggestions. The difference is, these companies are listening to and acting on many of those suggestions. That’s what accelerates innovation - starting a conversation with your customers and using your skills to understand and exploit their knowledge.
I agree with most of this, but frankly the first line seems somewhat like fence sitting. Especially when compared to Inside Steve’s Brain (p. 63):
A lot of companies like to say they’re customer-centric. They approach their users and ask them what they want. This so-called user-centric innovation is driven by feedback and focus groups. But Jobs shuns laborious studies of users locked in a conference room. He plays with the new technology himself, noting his own reactions to it, which is given as feedback to his engineers. If something is too hard to use, Jobs gives instructions for it to be simplified. Anything that is unnecessary or confusing is to be removed. If it works for him, it’ll work for Apple’s customers.
From our experience at Praxis, users have definitely been a powerful force in improving our LanguagePod platform, but bigger, step changes - for good and bad - tend to be more the result of a vision for how we see the service evolving.
An Open Source conversation on the media pioneer, Tony Schwartz, between Christopher Lydon and documentarian David Hoffman.
Listen:
http://www.brown.edu/Departments/Watson_…
In a similar spirit as Marshall McLuhan, the conversation uncovers Tony Schwartz’s opinions on the power of radio - many of which can be equally applied to the application of audio podcasts. There are a lot of tips here for the creation of podcast lessons, as well as the differences in using video.
Notes on the difference between radio and TV:
- People are born without ‘earlids’…what determines what people hear?…they hear what interests them, what concerns them….If we ‘pre-search’ peoples’ concerns and media channels - we will know that our message will be heard.
- There is a fundamental difference between seeing and hearing; with sight anyone can blink and look away, but with hearing and no earlids - listening depends on interest.
- Televisions is not a medium of information, it is a medium of effects - emotional effects, personal effects.
- The ear carries the heaviest freight, the eye is a distraction (almost need to neutralize image to get message across). The eye distracts, the ear is what we are really about.
- (levels of engagement) Compare your frequency of changing TV channels as compared to radio stations?
Notes on the ‘responsive chord’:
- Communication is not what I say, it is what you hear. It is not the words I use, it is how you interpret those words. For me to communicate with you I need to know how you feel.
- Don’t tell me what you think, talk to me about what concerns me.
- Sometimes you speak to people who don’t agree with you…start first with their position and gradually move them to your position.
- When speaking to people that already agree with you, just tap stored memory and remind.
- When you make yourself human (e.g. self-depreciation), you increase your credibility with the audience you are trying to reach.
And a gem of a quote by Lydon:
The human voice - that integrates so much chaos into a palpable human stream - is the answer to information overload.
These lessons are just as applicable to the creation of learning media as they are for mass media.
All of the MP3 lessons on our Praxis LanguagePod’s are licensed under a Creative Commons Attribution 3.0 license. We originally did this to make it as easy as possible for students and teachers to use the product; instead of having to clear copyrights before use, teachers only need to attribute the lesson to us and then are free to use the lesson in their classes.
Creative Commons has just launched a database of case studies and we have added our LanguagePod’s to the list, e.g. ChinesePod.
A while back silicon.com asked me: “Do you get annoyed that people rip your material off? “
I don’t really see it as ‘ripping us off’ and my answer now is the same as it was back in 2006:
No. We encourage people to use our Creative Commons-licensed podcasts as it assists us with our product development and helps push our brand into the community.
Frankly, we haven’t done a particularly good job of promoting this aspect of our service, but there still have been a few unique applications.
Back in 2006, Sam Brown tried to mashup ChinesePod with hypnosis.
His introduction to the service:
http://www.sambrown.co.uk/chinesepod/dow…
One big potential application of this licensing regime is the possibility of reforming university language labs. It seems a little ridiculous to me that after more than a decade since the Internet was introduced, students are still forced to schedule a time in a computer lab and then have to physically travel there. They are not able to consume the learning materials at home or on-the-go because of licensing restrictions by the publishers of the learning materials. This is not ‘learning on your terms’ and I hope we can play a role in putting an end to this practice. The OpenCourseWare Consortium is leading the way in this regard and hopefully we can help them out with their language learning materials.
Technology has the ability to radically disrupt any industry’s value chain - including education. Harold Jarche (a fellow Canadian) on his Learning and Work on the Web blog comments:
.. the dominant education business model may suffer the same fate as the manufacturing industry - commoditization.
I would take a slightly different view and argue a better term would be ‘optimization’. Standardization in manufacturing has been removing inefficiencies in manufacturing value chains - namely sourcing search costs - and has resulted in Walmarts packed full of low-cost goods. Maybe bad for US workers, but great for US consumers (and people who want to offshore manufacturing pollution; my poor China-stained lungs!).
The same holds true for education. By breaking down the value chain in education into discreet value-adding activities, technology can be used to optimize the process resulting in more student convenience and inevitably lower prices as the inefficiencies get shaken out of the system and competition pushes prices down.
This is exactly what we have been trying to do with our LanguagePod’s. By breaking down the language learning value chain for adult students here in Shanghai, we believe we can smartly apply technology where it is appropriate providing a service that is not only more convenient and more personalized, but with a reduced cost structure as well. This is how we see the emerging value chain for language training breaking down:
More market-sensitive education products (e.g. adult learning) are likely to be the first to shift to student-centered learning. With a huge chunk of adult English training here in Shanghai served by private schools, one would expect a large amount of innovation to occur as entrepreneurs try to solve problems in innovative ways.
One other example. The Economist has an interesting article this week entitled ‘Financial Exams‘ which discusses the rising popularity of the CFA exam. It seems to me a good example of Jarche’s point in a previous article:
At a certain point in time (2008?) the cost-benefits of a university education will be put in question. How expensive does it have to be before the majority opt out or look for “good enough” options? Once a certification body gets recognized by enough employers, it could become the de facto as well as the de jure standard.
If I am an aspiring investment banker looking at making an investment in either an MBA or the CFA, why would I stop work for up to 2 years and pay significantly more in tuition for an MBA only to have future employers value a level III CFA certification more?
I have been having a conversation with Graham Attwell - the driving force behind the concept of the personal learning environment (PLE) - on how mobile devices will impact the PLE. Channel agnosticism is a big part of our ‘Learning on Your Terms’ philosophy, or in other words the student should always have the choice on how they access networked content, be it through a desktop browser, laptop browser, mobile phone, networked TV (e.g. Apple TV), etc. The technology is just ‘plumbing’ and should focus on facilitating study access to learning tools.
Beyond this ‘plumbing’ though, increased network access will put increasing importance, on not only personal contextual variables, but also environmental contextual variables.
There is an insightful report from 2006 entitled ‘Ambient Learning - Ambient, multimodal and context-sensitive lifelong learning (PDF link)‘ that explicitly lays out what these variables could be.
To date, personalization in modern web design has very much focused on the personal contextual variables, be they socio-demographic, marketing mix-related criteria, medical context, or business context.
But by bringing the network into the context of people’s lives through mobile technologies, a whole set of environment contextual variables are going to be thrown into the mix: local context, time context, physical context and infrastructure context.
It seems clear to me that there will be a trend in the design of web-based learning toolkits, to not only incorporate traditional variables for personalization, but also leverage these new environment variables enabled by the ‘plumbing’ of increasingly powerful mobile devices.
Recently, I have been finding myself defending why podcasts work better as lesson objects when compared to other types of e-learning, such as webcasts. So, a quick review:
What is a podcast?
A podcast is simply an audio or video file that is automatically delivered to you via an RSS subscription. You can think of it as subscribing to a magazine that automatically gets delivered to you home every month as opposed to the inconvenience of having to visit the local newsstand. Podcasts & RSS makes staying up-to-date easier, as well as in our case acting as an ultra-efficient method to distribute customized, learning materials.
Why are podcast lessons different?
1. Rapid Publication Cycle
Unlike traditional textbook publishing, which takes years and then is infrequently updated, our podcast lessons usually take only a few weeks from ideation to publishing. This means that we can publish extremely relevant and contemporary topics. Take today’s ChinesePod lesson on the Sichuan earthquake published only 2+ weeks after that tragic event.
2. Set Schedule Creates ‘Learning Events’
As a new lesson is published every day, a sense of anticipation is created amongst users. This surprise or ‘breaking news’ effect helps create a sense of continuous, on-going learning. Or in Ken’s words:
The balance between the familiar (structures, formats, hosts, etc) and the unexpected (topics, themes, moods) keeps interest high. Through user response, the daily learning events are negotiated conversations rather than lectures.
Contrast this approach with a traditional archive. If we simply created an archive, of say 100 lessons, and then released it to the public where would students focus their attention & comments? By releasing one lesson a day, we can focus a mass of attention around one lesson thereby adding fuel to the conversation. We tend to view our lessons as almost ‘pre-baked’ and are only finished after students can comment, question and add their own insights or stories. This is the difference we see between a ‘lesson’ and a ‘lesson event’.
3. Style
A common mistake I often see with the use of podcasting in education is ‘lecturecasting’ - simply recording a lecture and publishing it online in the form of an RSS podcast. Content needs to be designed for the medium. Could you listen to a radio show in a movie studio? Yes, but radio is clearly not optimized for that medium. Similarly, lectures may work when they have a captive - if not always awake - audience, but they need to be optimized for the podcast medium. Put yourself into the shoes of an iPod-bearing student. First, the lesson needs to be engaging enough for them to listen to it over their favorite song - which is only a click away. By using two teachers and a radio-style, conversational approach the audio is more engaging. The student must want to listen to this lesson input (edutainment?). Second, the student will likely listen to the lessons while doing something else (in transit, shopping, at the gym, etc.) so short, 10-15 minute, manageable chunks of content are more easily adaptable to their physical world activities.
Then for language learning specifically you have additional benefits. Again, in Ken’s words:
The high-frequency language takes the listener into real-life situations. Hosts offer expert insight into the language and the learning in short, manageable lessons that include, stories, anecdotes, humor, mnemonic devices, and more.
4. Re-mixable, Modular Lessons
Maybe the most unique feature of our approach is that each lesson is modular, but graded by difficultly level. For example, a new student to ChinesePod could start with any Newbie lesson on the site and be able to start learning Chinese. Even though this modularity is native to podcasting technology, many of our fellow language podcasters have tried to force course architectures onto their lessons, e.g. one lesson should follow another. A podcast is not a physical textbook, so why limit oneself to the constraints of the previous media?
By making lessons modular, it enables lesson re-mixing allowing for fully customizable study regimes. The study goals of learners can be identified and learning pathways can be charted by freely selecting and combining lessons according to their needs as determined by themselves, software tools or their language teacher.
5. Channel Agnostic / Mobile
Part of ‘Learning on You Terms’ is to be able to consume learning materials on one’s own preferred media. A personal RSS feed is used for the ultra-efficient delivery of learning materials. Once subscribed to this feed in a program such as iTunes, the student can subscribe to future lessons at various difficultly levels, bookmark lessons from the archive or be assigned lessons from their teacher and have them automatically delivered to their iPod ready for consumption. Because the lessons are delivered via RSS, they could be delivered through a web browser, to an iPod, on a mobile phone, in a customized workbook or CD (more on this later), on TV or through a learning service/API.
This is learning adapting to your lifestyle, rather than you adapting to the learning.
Hank.