educ391x blog

Last week was the final meeting of Educ 391, during which we had an expo for our projects. Lots of people with various backgrounds came from off campus, and I received great feedback from many perspectives. Two of the people I talked to even had extensive experience working with dyslexic children, one as a psychologist and the other as a special-ed teacher. One of the most rewarding experiences of the night was seeing the special-ed teacher’s eyes completely light up when she heard about my project. About thirty seconds into my explanation, she interrupted me to tell me how she currently uses flashcards to teach her dyslexic children, how cumbersome the system she currently uses is, and how SmartCards fixes a lot of the problems she had. It felt pretty good to know that what I had developed might actually be applicable.

Of course, not all feedback I received was praise. Most everyone I talked to, including the special-ed teacher, offered ideas on how SmartCards could be improved. Here is a bit of what they said:

-Allow parents and teachers to add their own cards so that kids can practice words they have difficulty with that are not in the system.

-In general, make more functionality for teachers in addition to parents, as SmartCards’ automatic tracking features would actually make their job drastically easier when teaching dyslexic children. When developing SmartCards, I didn’t realize this.

-Offer some sort of incentive for children to work on cards (high score list, game-like functionality, some sort visual prize, etc.).

I also had a pretty lengthy conversation with an HCI guy (whose name I am completely forgetting) about ways to make the flashcard system useable by a child without parental supervision. He had lots of really cool ideas about low cost ways, most centering around the Wiimote, to allow the computer to determine whether a child correctly spelled a letter out in the air with they hand. While I think, for now, this lies outside the scope of the project, it was certainly cool to know that it was possible.

Overall, Educ 391 was an incredibly rewarding experience on multiple fronts. First, I learned a lot about design, something that I had little previous experience with, and will definitely use principles like ABCD in the future. Because of this knowledge, I was able to work on a project that was both personally satisfying, as well as potentially empowering to a group of people whom I think desperately deserve it. However, even more important than my own project was learning about what everyone else was working on, and their plans for empowering the communities that they care about. I was throughly impressed with the dedication and open-heartedness of everyone in the class, and wish them the best in the future. Thanks Dr. Kim and everyone in Educ 391 for a great quarter!

During our last class section, Dr. Kim was out of town, so instead we had a lecture delivered electronically from Curtis Bonk of the University of Indiana. Dr. Bonk is an entertaining guy and somehow managed to fill up all three hours of class with interesting material. His talk centered around ten different ideas or technologies which he thinks are “opening” the learning world. The ideas make a cute acronym, and are as follows:

Web Searching in the World of e-Books

E-Learning and Blended Learning

Availability of Open Source and Free Software

Leveraged Resources and OpenCourseWare

Learning Object Repositories and Portals

Learner Participation in Open Info Communities

Electronic Collaboration and Interaction

Alternate Reality Learning

Real-Time Mobility and Portability

Networks of Personalized Learning

I don’t have nearly enough time or space here to go over everything he talked about, but you can check out Dr. Bonk’s website and find his slides for yourself.

Although I learned a lot for Dr. Bonk, I do have one serious criticism of his presentation. While many of the technologies that he described are indeed making information widely available and changing the way we think about education, they are not being used at all for the sake of the most underprivileged. Rather, much of the information presupposes some sort of relatively high level of education. While it is great that MIT is putting so much course material online, what good does a free college course really do for a kid in the developing world who lacks even a basic education?

This appears to be a more widespread problem than one might initially think. Take another website Dr. Bonk talked about: Scribd. Scribd could be a great resource for sharing learning materials for use amongst children in the developing world. Instead, an overwhelming majority of the content is scholarly articles. I was able to find some basic arithmetic books on scribd, but they were illegal, pirated copies, which are hardly a long term answer to educating the underprivileged.

My point is this: while ideas like open source software and repositories of free online content are indeed opening and revolutionizing more advanced education for the already educated, they do very little for those who lack a basic education not because they are fundamentally incapable of doing so, but because no one is creating content aimed at this audience. While it is great to flaunt the merits of a website like Peer2Peer University, for the most underprivileged to take part there needs to be an equivalent Peer2Peer Elementary, Middle, and High School first. Unfortunately, organizations like Peer2Peer University, whose motto is “learning for everyone, by everyone, about almost anything”, seem not to realize this. In order for educational web technology to be truly revolutionary and “opening”, it needs to first address the needs of the underprivileged and uneducated in the developing world. Otherwise, we can’t really say that WE-ALL-LEARN.

Last week in class we all presented our second digital artifacts. There was a wide range of topics and technologies discussed, but before I get to them, I will talk about my own project. My project took the form of a spaced repetition system for teaching dyslexic children the sounds associated with different letters of the alphabet. I received a ton of great feedback from the class, as well as from a visitor to the class who happened to be dyslexic. Here are some of the suggestions that I got:

-While advertising may be a good way to keep the project economically viable, given the high concurrence of ADD and dyslexia, it is probably a bad thing to have advertising on any page containing a flash card. I should note that the advertising I was thinking of would be for other educational products only, and would not be anything garish that was likely to drive users away.

-It might be useful to have an audio file attached to each flashcard, as the pronunciation of some letters might not be instantly obvious to parents. I hope to add this feature before presenting during the final class session.

-The system could be more effective if it eventually began to incorporate flashcards with more than one letter. I agree completely with this, and, if I were to fully develop my project, I would love it to be a more long-term, comprehensive system. However, for the purposes of presenting for this class, I wanted to keep the artifact both focused and simple, and thus decided only to focus on single letters.

I was excited about pretty much every other project presented. However, it was the two ideas for games that really caught my eye. Catherine’s Zoo Escape game had beautiful concept art, and I wanted to play it myself. Furthermore, Jacob’s idea for a biology video game seemed incredibly viable. Many science teachers I had tried to make science “real” for us through the use of metaphor. Different systems in nature were compared to things we were already familiar with. This both excited us as students and made the concepts easier to learn. The power of Jacob’s idea is that it relies on such comparisons, which are plentiful. The example he brought up was comparing a red blood cell to a spaceship picking up different cargos. This is an easy concept to grasp for any child who has seen a science fiction movie, as well as a pretty apt metaphor. Through the use of such metaphor in a game, I feel as though his game could not only garner an audience, but actually be successful at imparting knowledge. I hope that he develops his idea further and that I see more like it in the future.

Last week we spent class listening to two presentations: one on a start-up for-profit company, and the other on a non-profit. Both organizations were in the very early stages, although the for-profit was much further along, as they actually had something concrete to show us. It is this for profit that I will discuss here.

Their technology basically allows one to treat a streaming flash video like google maps in that one can zoom into certain parts and pan side to side by dragging the mouse. While the technology was not invented with education in mind, the creators realized that the videotaping of lectures was one specific area in which they could apply it. Currently, they claim, the videotaping and online archiving of academic lectures is incredibly costly due to the specialized technology and large amount of manpower it requires. For example, at Stanford, there are numerous cameras mounted on rotating turrets used to record, which are operated by a human, who later manually uploads the video to a central server. With the company’s product, one would simply set up a single camera in the back of a room, which would turn on and off automatically on a time schedule, then upload the unedited video to a central server. It would then be up to the user to pan and zoom into specific parts of the video. The creators of the technology argue that this is a superior experience from the users point of view, as it allows them more freedom when watching a lecture online – they can view the parts of the room they think are relevant, as opposed to those that the camera operator thinks are relevant.

Obviously, this technology appears superior to the current setup from a school’s point of view, as it is drastically cheaper and simpler than the current way things are done. While there are obvious problems (lecturers do not always start class on time, etc.), the product is in its early stages, and I assume that these will be ironed out. However, I question whether or not the ability to control where the camera is pointing at all times during a video taped lecture is necessarily better than having predefined camera angles. When I watch a lecture online, I want my mind to be completely focused on watching the lecture and taking notes. However, requiring me to drag the camera around to see different content is distracting, and may cause unwanted cognitive overload.

The company appears to address this problem with their “lecturer follow” feature, which automatically tracks the lecturer through the frame. However, the current movement is very jumpy and distracting. Furthermore, the lecturer is often visually unimportant compared to the boards he or she is writing or has just written on. What I like about the current way video taped lectures are done (at least at Stanford), is that they zoom in on and single out the important content at any point in the lecture.

To this end, I believe this start up should have human “edit” a video of a class before it is uploaded by creating a predefined track for the virtual “camera” to follow. However, if for some reason the viewer wanted to look at something off the predefined track, they could always drag the camera off of it, then hit a button to but the camera back on track. This would allow them to effectively view the information they wanted to without having to rewind or pause, while remaining undistracted during normal lecture viewing.

On a whole, I think that the company has a very marketable piece of technology. However, I suggest that they perform significant user testing before committing too strongly to a single implementation.

At the beginning of this quarter, I said that I was interested in developing technology to help kids with LD. Since then, I have narrowed my focus to elementary school children with dyslexia, specifically those with parents who may not have significant time to help them outside of school. As I mentioned briefly in class last week, multisensory learning is critical to dyslexic children learning which sounds correspond to which letters of the alphabet. One common technique is to show a picture of a letter and have the child say the corresponding sound (not the name of the letter) while drawing the letter in the air with their hand. As I understand it, such multisensory methods are usually not used in regular classrooms. As such, I have been spending time trying to come up with ideas for a sort of game that dyslexic children could play on their own where they would respond in a multi-modal manner to a letter displayed on the screen. However, I’ve run into the problem of this being difficult to do simply. Drawing a letter out with the mouse seems a poor way to learn kinesthetically when compared to doing it in the air with your hand, and analyzing the sounds of single letters not only requires a microphone, but is also a computer speech processing challenge beyond my skills.

However, I have begun to consider making a set of digital flashcards of the alphabet with accompanying instructions for parents, perhaps in the form of a short instructional video, on how to use them to teach their dyslexic child the sounds of the letters. There are numerous websites that outline how to employ multisensory methods to teach children, but these websites do not offer materials in an easy to use form to parents. For a busy working parent, this lack of easily accessible materials may be frustrating, and the parent thus may not give their child the help he or she needs. However, a centralized website with both advice and materials would not only draw more parents with dyslexic children, but also be more effective in motivating them to give their children extra help. I feel as though this idea encompasses the following theories of learning discussed in class on last Wednesday.

Modality Theory: Dyslexic children learn through through speaking (aurally), seeing (visually), and moving (psychomotor). Also, parents will perhaps learn how to instruct their children more readily through a video (aurally and visually) than simply by reading (visually) how.

Personalization Principle: Materials shown to children will be designed in ways that make them appealing. In videos for parents, a conversational and non-scientific tone will be kept.

Coherence Principle: Despite a design that appeals to children and a conversational tone in training videos, neither will be visually or aurally distracting.

Total Physical Response: Children both move and speak while learning.

Interactive Principle: Children will learn better through an interactive flash card game played with an instructor than simply by repeating what an instructor tells them.

This week, representatives from the e-learning division of Edison Learning visited the class and gave an overview the proprietary system that runs their online charter high schools. I have always been skeptical of online schools’ ability to successfully engage students and allow them to grow socially. However, on a whole, I was very impressed with Edison’s system. There are many measures in place to keep students engaged and on track, including easy access to tutors, an intervention system when a student fails to perform, and counsellors who helped plan a personalized learning tracks for students. On the social side of things, I was excited about the soon-to-be-launched idea of students being able to host discussion or tutoring sessions with other students and receiving a sort of “merit badge” for doing so. One concern I have about online schools is that the students will not be able to interact with other students on a meaningful level. These discussion sessions not only allow students a forum for interacting with and learn from each other, but also a motivation (in the form of social currency) for doing so.

The other aspect of the system I was impressed with was the modular course design. Basically, Edison has a huge set of roughly forty minute lessons that they assemble all their courses from This prevents them from having to rewrite course material every time they enter a new market. Instead, they simply assemble the modules differently so as to be compliant with local education standards.

I am curious if this modular course design is something that only Edison is doing, or if their competitors are as well. From the perspective of a for profit company it certainly seems like a good idea, as it keeps the cost of producing course material and curriculum low. However, such a strategy seems like a good one to keep in mind for anybody working to empower undeserved communities due to the fact that it facilitates the wide dispersal of educational tools. Take, for example, the strategy employed by Naledi3d, the South African company I made my first artifact about. By making modular VR simulations, they hope to allow educators to pick and choose lessons to construct a curriculum tailored to their community. Especially when dealing with topics pertaining to the developing world (health, sanitation, etc.), such a modular approach seems almost necessary, as the needs of each individual community vary so much.

Last week in class each student gave a presentation on a piece of technology, website, etc. aimed at educating the underserved. All of the presentations were interesting and I learned a ton about what people are currently trying to do and how. However, I have been thinking a lot about a project called English Seekho, which Ashley talked about in her presentation.

English Seekho is a product of Tata Telecom (the largest cell phone service provider in India). Basically, for about twenty rupees a month, not including calling fees, one can call a phone number and receive recorded English lessons. What is cool about the project is that it utilizes voice recognition software in order to test the learner.

I actually have a bit of a background in speech recognition, having worked developing dialog control systems for robots at Stanford’s Center for the Study of Language and Information. Coincidentally, about two hours before class, I started wondering whether speech recognition over cell phones was being used to teach English anywhere in the developing world. When I found out what it was, I was curious to see specifically what was happening. In Ashley’s presentation, she raised the question of how effective the actual speech recognizer would be. I can actually speak a bit to this.

The presentation made it sound as though the cell phone simply asks the speaker to repeat a known string of words in order to check the learner’s pronunciation. If this is the case, then the software is likely effective. Programs like Rosetta Stone do a similar thing, and, in my experience, they do it very well. However, learning to pronounce English words correctly is only marginally useful if the learner is never given a chance to speak English with someone more advanced. As English teachers are hard to come by in rural India, I think speech recognition over a cell phone could be used effectively to give learners practice generating very simple English language utterances on the fly. However, the system would have to be implemented very carefully in order to work.

When no context is given for a conversation, speech recognizers are generally unable to parse what the speaker says, as they have no idea what classes of words to expect. However, given a specific context or type of conversation, speech recognizers are incredibly accurate, because they know what words are likely to appear, and can “listen” for them. This is why the computerized operator you get when you call your bank is so effective: it knows you are probably going to say one of maybe ten different things, all pertaining to your bank account. I think that a similar system could be used to teach very basic English to the undeserved. A computer could ask a series of basic questions (What is your name? Where are you from? Etc.) with answers that would come in a predictable format (My name is Mike. I am from California. Etc.). Because the speech recognition software would know what to listen for, it would easily be able to understand the answers given by the learner and offer feedback. Obviously, this would only work in the most early stages of learning English, as sentence structure would rapidly become too complex for the speech recognizer to effectively understand the entire utterance. However, I feel as though it could be a useful substitute for a real language partner when one is first starting with English.

This week I want to dedicate my blog to explaining two important concepts that have been brought up in class since the beginning of the quarter: ABCD and Bloom’s taxonomy. For anyone in the class, this will be nothing new. However, I find that I learn best when I write down and explain concepts on my own terms, so at least this will be beneficial to me.

ABCD

We spent the second half of the last class doing a role play where we had to analyze a situation where a client wanted to implement an e-learning system and propose a solution using ABCD. This exercise made me aware that my grasp of ABCD was not perfect, and that my classmates’ definitions of the four terms sometimes conflicted with each other, as well as with mine. With that in mind, I’ll now define the four terms as I understand them.

Audience – Who the thing being designed is targeted towards. Could include information such as socioeconomic background, amount of previous education, languages spoken, technological literacy, etc.

Behavior – The specific behavior that you want to change. For example, you might want make it easier for students at a trade school to learn to rebuild an engine.

Conditions – The conditions of the environment that you are teaching in. For example, if one is developing a handheld learning device for the developing world, a limited access to electricity might be a condition.

Degree – To what extent you want to change behavior. This should contain specific, quantitative goals. For example, you might put down that you want to have 80% of students score above 2000 on their SAT.

I think part of my groups confusion was over things that could be potentially be classified in more than one category. Some things seem as though they could be either A or C. For example, assume a group of students is apathetic towards learning. One could write “apathetic” under audience or “apathetic students” under conditions.

At the end of the day, overlap such as this isn’t really that important I think. What is important about the model is that it allows a structured way for one to think about design. The overlap, while somewhat unfortunate, does not greatly detract from that end.

Bloom’s Taxonomy

Bloom’s Taxonomy is something that has been mentioned since the first meeting, but was not talked about in detail until last week. This was likely due to the fact that most people in the class already knew what it was. However, having no background in education, I did not know what it was until last week. According to Bloom, learning can be broken down into three different types.

Cognitive – Cognitive learning is the development of the mental, whether it be the learning of rote facts or how to perform some sort of abstract calculation.

Affective – Affective learning refers to the development of emotions or feelings. An example might be learning to empathize with a person or growing to like someone.

Psychomotor – Psychomotor learning is the development of physical or manual ability. For example, learning how to touch type or how to play a video game like Guitar Hero would be psychomotor learning.

In class, Professor Kim also alluded to a possible fourth type of learning he thought Bloom overlooked: social learning. However, I am somewhat unclear as to what social learning actually means. Does this refer to things that we learn about the people around us and how to interact with them? If so, it appears as though it would be a subset of affective learning. Does it refer to learning that happens as a result of discussing material with peers? If so, it seems as though social learning could fall into any of the three original categories depending on the skills being learned.

Anyway, see you next week.

Last week in in seminar we spent a lot of time going over ABCD and the difference between cognition and meta-cognition. This was a valuable discussion in that it provided me with a framework within which to design both for class and in other endeavors. However, what really got me thinking this last week was two of Professor Kim’s projects that he talked about in class.

First was the Pocket School, a handheld device for teaching the children of migrant workers in Mexico. Apparently one of the current stumbling blocks for the device is the ability to charge the batteries. Because there is no regularly accessible electricity where the children live, a hand crank generator was developed. However, it was rather inefficient, allowing only about four minutes of use for every two minutes of cranking if memory serves me right. Being an avid cyclist and a general proponent of the bicycle as a method of daily transportation, I am always on the lookout for cool bike projects. When Professor Kim mentioned that he was looking into the bicycle as a form of generation, it made me think of a start up my friend is currently working for that just last week publicly unveiled a high-efficiency bike powered generator.

The other project I’ve been thinking a lot about is the 1001 Stories project, a series of short story competitions for children living in undeserved areas. The goal of the project is to motivate kids to want to learn, as opposed to forcing learning on them. This idea really struck a chord with me, as motivation in students seems to be a problem not limited to the developing world. Irregardless of location or socioeconomic status, if a student has no will to learn, they simply won’t.

Inspired by 1001 Stories, I started to brainstorm various permutations of the idea. The most exciting that I came up with was a sort of innovation or engineering challenge in extremely impoverished communities. In the past few years, do-it-yourself culture has blossomed in the western world, leading to countless websites and print publications such as MAKE and Readymade. Many of the projects featured in these publications involve the creative reuse of found materials, and a few of the really impressive ones have come from the developing world. From my own travels, predominantly in rural China, I know that what makes it to the internet is only the tip of the iceberg in terms the ingenuity found amongst the underprivileged. In many cases the day to day survival of these populations depends on the ingenious reuse of common items. With that in mind, I think that fostering local engineering competitions would serve two purposes. Not only would it encourage people to create who might not otherwise, but it would also stimulate dialogue about and the sharing of projects within communities, empowering it’s members.

What actually excited me about this idea is the potential for it to be relatively self-sustaining. Like the 1001 Stories iPhone apps, iPhone apps detailing the construction of winning entries could be sold in a sort of “mini-magazine” format. Furthermore, content could be sold to websites and publications such as MAKE and Instructables.

Obviously, there are downsides to this idea. The most obvious to me being that the barrier to entering an engineering contest is much higher than a story one, and it would likely attract a smaller number of people. On the other hand, people might already have devices or objects built prior to the competition that they would want to submit. Also, the logistics of holding such a contest and actually documenting all entries is a bit more involved than hosting a story contest. However, I feel as though these obstacles are ones that are manageable.

As a final, unrelated note for any other students in the class who are actually reading this, I’m on the food committee, so feel free to ask for things to eat.

Hello! Welcome to my Education 391X blog. Over the course of the next quarter I’ll be blogging once a week about the class, web technology, learning, e-learning, and related topics. Our first meeting was relatively straightforward, and pretty much the entire time was spent giving introductions and going over the logistics of the course. I’m pretty excited about the wide variety of backgrounds in the class, and by the fact that I’m one of two undergraduates. I think I serve to learn a lot. Also, as someone planning to apply to co-term in the LDT program next year, it will be good to meet current students and learn about their experiences.

One of my big interests coming into this class is kids with learning differences. I have precisely no academic background in the area. Rather, my interest stems from the fact that my younger sister has LD, and has been schooled in a rather non-traditional way as a result. Watching her grow up, I often wondered how technology could assist her both academically and socially. Because she was placed in small special ed classes, she never interacted with most of the students at her schools, and her social life was rather limited. As a result, she hated school. However, she did not hate learning, and, when given the option to spend extra time studying outside of the classroom so that she could enroll in regular classes, she jumped at the opportunity. She was lucky in that my family had the resources to provide her with this extra help. I feel as though e-learning has the potential to provide such an opportunity to a broader ranger of kids with LD.

During our first meeting, I thought that the video about learning from data presented in different mediums was quite interesting. From being constantly around someone with LD basically my entire life, I’ve had it beaten into my head that multisensory learning is almost always a good thing. However, what the video seemed to stress was that, while the presence of multiple types of information (e.g. audio and visual) aids learning, the presence of more than once instance of the same type of information (e.g. written text and an image) can cause cognitive overload and hinder learning. For example, the video made the claim that one learns more readily from an image and spoken language than from an image, spoken language, as well as a written transcription of what was said. It seems as though keeping track of such insights would be important when designing technology to help kids with LD learn, as such learning often hinges on proper multisensory input.

Anyway, that’s about all I can think of for this week. I’m excited for the rest of the quarter and to see the work produced by everyone else in the class.