Posted by mathieso on August 28, 2013 at 1:46pm
http://demo.cybercour.se shows (the beginnings of) a system for building courses. Course design and authoring tools only so far.
There's a list of short (< 3 min) videos about the system.
http://demo.cybercour.se shows (the beginnings of) a system for building courses. Course design and authoring tools only so far.
There's a list of short (< 3 min) videos about the system.
Comments
Some feedback
pseudent-pose -- what is this? I like the concept, seems like ability to make a token that generates content. Is this a custom module or did you pull some contrib together to accomplish this?
Might want to checkout the 4.x version of textbook module as it has some decent ckeditor plugins / additional usability defaults worth noting -- https://drupal.org/project/textbook
personally would never teach our faculty / IDs about wiki style mark up but that's an audience thing, some love it, some don't get it.
page restructuring could benefit from outline designer
From a purely structural standpoint it appears that we may have some parallel efforts overlapping w/ the https://drupal.org/project/mooc distro. Not saying you should use it, just that you might be able to pull some modules from that rig to see what guts are powering it.
The knowledge mapping tool, while I don't entirely understand what it nets you is very promising looking and seems to work well thus far from the little map I tried to build. Is this trying to illustrate what concepts need to be taught in order to gain skills / experiences?
Look forward to seeing a distribution / install profile of some kind to tinker with, especially as the knowledge map part is concerned. Functionality like that would be very cool to have in some of the more experimental interactive object systems I'm playing with.
Ex Uno Plures
http://elmsln.org/
http://btopro.com/
http://drupal.psu.edu/
Thnx for the feedback
This is based on learning research. In the literature, metacognition is what students think about learning. For example, "math is in your genes - you know it, or you don't", and "I'll never use this stuff in the real world." These beliefs affect effort, persistence, etc. The empirical evidence is strong.
In a university course, these issues should be addressed. That's one of the things pseudents do. Pseudent is short for pseudo-student. They're like students sitting next to you in a lecture, asking questions you hadn't realized were important, challenging the prof, complaining when things are too hard, and so on. You learn from their questions, and the prof's answers.
If you're interested, have a look at:
http://www.coredogs.com/lesson/tmi
At the top of the page, you'll see two pseudents - Renata and CC - complaining to a prof - Kieran - about being overwhelmed. Kieran acknowledges their complaint, and gives them heuristics that work most of the time.
Pseudents have other uses, too. For example, you can watch them as they work through practice problems. They make mistakes, e.g., in JS code. They find the bugs, and continue. Other times, they ask for help. In doing these things, they model what I want real students to do.
Pseudents illustrate what my project is about. I am NOT trying to help professors do what they do today. Instead, the goal is to give students content that implements principles from learning research, and make it easy for profs to teach using methods derived from that research.
Yes, yes, I know, it's hopeless, profs won't change, I'm a naive dreamer, etc, etc. I know the issues. My suggestion is to unbundle prof's jobs, have course design and content creation done by someone else (maybe IDs, probably not), and let profs do what they do best: be domain experts. There's a business model and other things to help achieve that.
Content type, views, custom module, custom CKEditor plugin.
What if you could show them that using markup was less work? What if you used GUI methods to make markup easier to write?
Imagine using CKEditor to type in markup. Yes, really. "But you use textareas to write markup!" Not if you're a heretic. You might create an editor that combines the precision of markup with GUI usability.
(Ooo, you could burn the heretic. Make smores. Actually, heretic jerky is quite tasty.)
There's a draft of a paper on the editor at:
https://docs.google.com/document/d/1lc155yuQFM8ppqaT4DP4SDPd6i1623qSjnH9...
I'll be talking about it at Michigan DrupalCamp on Saturday. Slides at https://github.com/coredogs/swim_slides.git.
BTW, I've switched from Markdown to reStructuredText, from the Python community.
My project is as much about course design, as it is about running courses. One approach to course design is called Understanding by Design. UbD is popular among education people. UbD is big on outcomes. Generally:
Set fairly precise goals as to what you want students to know.
Design experiences that help students learn those things. Use learning research to guide the design of those experiences.
Honor constraints, esp. on student time. Be realistic about how long deep learning takes (deep learning: learning how to DO things, not just remember facts for the exam).
That's what the learning map is designed to help with. It has two layers: knowledge, and experiences. Start by laying out the knowledge (facts and skills) that students should know. Start at the end of the course, and work backwards.
Of course, students don't have concepts poured into their heads (I've tried - IRB wasn't happy). Students construct new knowledge from the experiences they have. That's where the term constructivism comes from.
Hence the second layer of the learning map: experiences. Readings, demonstrations, exercises, videos, more exercises, field trips, even more exercises. Each experience is linked to the concepts it helps students learn.
Some experiences are required. Others are reinforcing - a particular student does them if s/he didn't learn the right stuff from the required experiences.
Each experience has time estimates attached. How much time would it take for good, average, and mediocre students to do it? That helps keep course time constraints in mind. Eventually, I plan to let authors run simulations, based on the time data.
Do all this, and you should have a kick ass course.
You're thinking: "This dude is fucking crazy! Profs won't do that!"
You're right. They won't. No way, no how. That's a negatory. NuhUH, girlfriend!
The course design tools are for course designers, not profs. Designers create courses that replace textbooks, and bake in all of this chocolately learning science goodness. Profs tell students to buy the courses. There's a sustainable business model behind all this, but you're tired of reading already.
There's much more to the project. Esp. a system for giving frequent formative feedback. Again, implements learning research guidelines, with a business model to make it practically feasible. Profs don't like massive grading, so have someone else do it. Not TAs, necessarily. Use the Internet thing (I'm agin it. It's a fad. Hey, there are kids on my lawn!) to distribute the work.
I'd appreciate any feedback about the project. I'll even buy you virtual coffee and cookies.
Kieran
Knight of the Woeful Countenance
Kieran Mathieson
kieran@dolfinity.com