Category Archives: Online Learning

Google’s Bard on Language Acquisition and Large Language Models

Prompt

Based on your understanding of the ideas and theories of Stephen Krashen write a short presentation that Krashen might give on the possibilities offered by large language models.

Bard’s Response

Good morning. My name is Stephen Krashen, and I am a professor emeritus of linguistics at the University of Southern California. I am here today to talk to you about the possibilities offered by large language models.

Large language models are a type of artificial intelligence that are trained on massive datasets of text. This training allows them to learn the statistical relationships between words and phrases, and to generate text that is similar to the text they were trained on.

Large language models have a number of potential benefits for language learning. First, they can be used to provide learners with comprehensible input. Comprehensible input is input that is just a little bit beyond the learner’s current level of proficiency. It is input that the learner can understand with a little bit of effort.

Large language models can provide comprehensible input in a number of ways. They can be used to generate personalized stories, articles, and other texts that are tailored to the learner’s individual needs. They can also be used to translate texts from one language to another, which can help learners to access a wider range of materials.

Second, large language models can be used to provide feedback on learners’ output. Output is the production of language by the learner. It is the process of speaking, writing, and reading.

Feedback on output can help learners to improve their accuracy, fluency, and grammar. Large language models can provide feedback on output in a number of ways. They can be used to identify errors in grammar and usage. They can also be used to provide suggestions for how to improve the accuracy, fluency, and style of the learner’s output.

Third, large language models can be used to create new learning experiences. For example, they can be used to create virtual worlds where learners can interact with native speakers and practice their language skills. They can also be used to create games and other interactive activities that make learning fun and engaging.

In conclusion, large language models have a number of potential benefits for language learning. They can be used to provide comprehensible input, feedback on output, and new learning experiences. I believe that large language models have the potential to revolutionize language learning.


Thank you for your time.thumb_upthumb_downuploadGoogle itmore_vert

Blackboard vs Canvas vs Moodle

A colleague at another university, which is contemplating a move to one of these four learning management systems, wanted to know the preferences from among this list of systems:

Here is my reply:

BYU was a huge Blackboard user for several years, before deciding to develop their own LMS

BYU was a huge Blackboard user for several years, before deciding to develop their own LMS (BYU Learning Suite). The decision to build one’s own system was not one that I supported, but given Blackboard’s exorbitant cost, I totally understood the motivation.

That said, in the lab I directed before retirement we implemented Moodle on a couple of projects. We have also worked with Canvas and even received an award a couple of years ago from Instructure for our implementation of LTI (Learning Tools Interoperability) with Ayamel.

Canvas and Moodle each have advantages. There is a large community of Moodle users, but we found certain functionality lacking. The fact that it is free is also a bit misleading, given that a particular shop is on its own for the system’s implementation. Yes, there is a community of users “out there” to help, but the system still requires an in-house support capability when technical issues arise.

You have probably done your searching, but I just came across this piece that compares Blackboard, Canvas, and Moodle. Blackboard comes in third place, and Canvas edges out Moodle, but only very slightly.

To me the key factors lie with a concept a colleague and I wrote about here and labeled “tool and content malleability.” A key aspect of this is LTI a capability that both Canvas and Moodle have to some degree, Thus it all comes down to the level of technical support you can expect at your institution.

Learning Analytics

My two most recent columns as a Contributing Editor for Educational Technology explain the beginnings of my exploration of learning analytics, a topic I feel is crucial for education today. Learning about learning has the potential to help us understand not only more than we know about learning itself but also about how to (and probably how NOT to) insert technology into the learning process. The first is entitled  “Learning Analytics and Educational Research” and the second is “Importance and Implementation of Learning Analytics.” To say the topic is crucial would be a serious understatement.

Webinar on Ayamel

I had a great time yesterday doing a webinar on Ayamel. I have given many presentations in many different settings, but this was perhaps the most challenging. Sitting there looking at the computer screen with no visual feedback from the audience was interesting, to say the least.

As I was speaking during the presentation, I noticed a couple of typos in some slides due to some late additions. Also, listening to it afterwards, I noticed way too many “uhs,”which I considered editing out. Now, however, I have decided to throw vanity to the wind and make the link available immediately, simply because I believe that the brilliance of Ayamel will come through, despite the imperfections in the presenter’s delivery!

With that explanation, here is the link on YouTube for the video of the webinar. I have posted the PowerPoint deck here (corrected of course! 🙂  ). All I ask is that anyone who downloads the deck give attribution, should any slides be used in another presentation.

Finally, here is a link to a short (10-minute) Qualtrics survey for anyone who would like to have additional permissions for using Ayamel, such as “Create Course.”

Moving Forward with Online Learning?

I have heard rumblings that BYU Provo that BYU Provo is showing an increased interest in online learning.

I will confess to mixed feelings, however, which I will admit are totally selfish…

Shortly after I arrived at BYU in 1992, I visited with the late Dwight Laws, the manager of BYU Independent Study at the time. The thrust of the interaction was a discussion of the work I had been doing with the technologies of the day at the Air Force Academy. His response was that they were totally open to doing creative stuff, as long as it could be delivered with paper and ink.

In all fairness to Dwight, it is only now, inLLC2 the 15th year of the new century that we can do on a wide scale the sort of things we were doing in 1992 at the Air Force Academy. Even then, what we did was on a very limited scale. A primary
limiting factor of course was the expensive nature of the technologies of the day: The price of each Sony View 3000 videodisc workstation was $8,000, which would be at least $12,000 today. Even then, the limited number of workstations required that we assign two students to each workstation at a time, which explains the two headsets and one mouse seen in the photo. So much for individualized learning.

The primary irritating factor today is my advanced age, which will limit my involvement in a very exciting time for learning: If the technology can do something useful for teaching and learning, don’t waste the teacher’s time doing those things. We know that there are many exciting, interesting, and motivating things the teacher can do that the technology will not, for the foreseeable future be able to accomplish (Star Wars’ C-3PO not withstanding!)

A second irritating factor is that online learning efforts are not yet fully exploiting the tremendous computing power people are carrying around in their pocket.

Perhaps that is about to change?

 

Educational Technology Bust

It seems that Apple and Pearson are about to shell out big bucks, $6.4 million to be exact for ipadwhat appears to be a massive failure in supporting a tablet initiative in a thousand schools in Los Angeles:

The Los Angeles Unified School District has reached a tentative $6.4-million settlement over curriculum from education software giant Pearson that the school system said its teachers barely used.

The pact is the latest fallout from an aborted $1.3-billion plan to provide an iPad to every student, teacher and campus administrator in the nation’s second-largest school district.

The Board of Education is expected to vote on the settlement in October. The bidding process that led to the original contract is the subject of an FBI investigation.

This is from a piece in the Los Angeles TimesLA Unified to get $6.4 million in settlement over iPad software, that says that an FBI investigation is underway into the manner in which the contract was let. The writer also quotes an attorney for the school district, “one goal is to make sure schools are ready to use technology before they receive it.” Well duh!

I often say that forecasting technology applications is really tough, especially when it involves the future. We also know that we will be wrong twice… We will overestimate the near term and underestimate the long term. Although Pearson claims that their materials are being used effectively in many other schools, it seems that in this case these folks are in the near term of effective uses of tablet devices.

Read more here and here.

Storytelling and Education

I have believed for some time that good educational practice is intimately related to the notion of good storytelling. As I was reading today about this massive investment that Facebook founder, Mark Zuckerberg, and other tech leaders are making in education through “Altschool,” I was surprised by the connection that I quickly made between that investment in those schools and storytelling. The connection came from a passage in that piece that described the schools as implementing individualized learning, a “Reggio Emilia-Approach Intertwined With Technology.” Knowing nothing about Reggio Emilia caused me to jump to this Wikipedia article on the schools in Reggia Emilia, Italy. Here is the quote from that piece that caught my attention:

 For example, teachers in Reggio Emilia assert the importance of being confused as a contributor to learning; thus a major teaching strategy is purposely to allow mistakes to happen, or to begin a project with no clear sense of where it might end.

It seems that the idea of being confused, allowing mistakes to happen, or having no clear sense of how a project is to end is tantamount to the essence of good story telling, no?

MOOCs?

Udacity is throwing in the towel on MOOCs with the company’s founder, Sebastian Thrun, declaring that they don’t work for higher education.

I described in a recent column in Educational Technology how the primary advantage of MOOCs is that they can bring educational opportunities to learners who have no other opportunities. Unfortunately, they fail to deal with individual needs for learning in ways possible with other means. Check out this article to learn more about Thrun’s reservations: What Will Happen to MOOCs Now that Udacity Is Leaving Higher Ed? — Campus Technology.

The Flipped Classroom: Getting Serious

For various reasons the flipped classroom is of particular interest to me. While accountability in any enterprise is one of the key elements of success. what students do outside of the classroom is one of the least accountable aspects of education today. Getting a handle on the impact of the flipped classroom could contribute to that particular gap in our knowledge of how students learn.

This research project seems to hold excellent promise in exploring the value of the flipped classroom: Assessing the Flipped Classroom’s Impact on Learning — Campus Technology.

This report on the flipped classroom at several institutions provides several interesting insights: How to Make the Most of the Flipped Classroom — Campus Technology.