Thursday, September 10, 2009

How Google Wave Could Improve Education: Group Work

Last night, at the Google Wave Sydney User Group, we brainstormed a lot of different ways that Google Wave could be used in a vast spectrum of fields. A recurrent theme was education, and we touched upon various ideas, and basically just agreed that Wave (and collaborative technology, generally) can have huge implications for education. One of those ideas we discussed was group projects, and I wanted to discuss that area in more detail here.

As a teacher, you often want to encourage group projects, since they can help students learn cooperation and teamwork, and since they can often have a synergistic effect and produce amazing results. But, there's always that same concern with group projects - someone will do all the work, someone will do none, and it's impossible to know who deserves the good grade. Well, Wave could potentially solve this, both in terms of knowing who to give credit to, and encouraging a better balance of work across group members.

Here's an example proposal based on my own experience:

You're a teacher running a game projects class, and you've divided the class into project teams for the semester. The first aspect of creating a game is writing a design doc for the game (describing gameplay, objective, characters, etc), and you make that the first assignment for each team. You create a Wave for each team that has the design doc template in it (headers), and you give them a week to work on it. You recommend that they use reply blips and inline blips to divy up the work, and tell them that you will be monitoring the project and will do a review of the first draft in a week, and grading of the final a week after that.

When you review, you reply in the Wave and comment on what parts need work or clarification, and which parts look good. When it comes time for grading, you use the playback functionality to see how everyone contributed. You see who took the lead and made decisions, who wrote up large sections, who revised sections for clarity, and who just sit back and watched. Now, hopefully, since every team member knows that you can see all the edits and attribute them to each of them, they will actually feel more motivation to contribute, and you won't have many students that just do nothing. But if you do, well, you can penalize them accordingly, and perhaps next time, they will learn that their actions (or lackthereof) are being watched.

You could even automate some aspects of this with robots. There could be a robot that counts the word count of each team member, and outputs that in a stats gadget at the end. This could feel a bit too strict for the students, but it can also serve as a nice "hey, you're not quite doing as much" reminder. That robot could also just monitor progress as a whole, making sure the team stays on progress. The gadget could list stats like "5 sections completed, 5 days left until final due". Then, those per-team-member stats just become another indicator of progress on the whole, and don't feel as accusatory.

One issue I see with this proposal is the tracking of "peer waving." Some students may sincerely find it more productive to sit at one computer and write up sections together - perhaps one giving the ideas, the other putting them into words. These words would only be attributed to one student, and the idea giver could be unfairly penalized. Possible solutions for this would be to annotate the text [Peer waved w/username@bla.com] or to recommend that students alternate between accounts when doing this.

Another issue, of course, is that not all group projects are writing up a blob of text, and in fact, many of them are not. In computer science, many of them are actually writing code. You can already ask students to use SVN and monitor commits from each team member for less fine-grained monitoring, but you could potentially even have students write their code in a Wave, and use a robot to compile it out, for real-time monitoring. In other fields, many group projects involve the the creation of a physical (non-digital) object. Perhaps you could have students write a design doc for that physical object, or even use a collaborative gadget to sketch it out.

So, yes, this solution isn't perfect, and for non-text projects, involves a bit more creativity. But, I still think it's pretty damn good. If you're a teacher, try it out and report back!

Tuesday, September 8, 2009

The Quest for Karaoke-able Slidesets

I'm a massive fan of karaoke, and I've often expressed my opinion that some songs are far more "karaoke-able" than others, and that someone/me/an entepreneur should create an app capitalizing on that. The app could let you rate the karaoke-ness of songs, upload video evidence, create karaoke playlists, and find lyrics for practicing. It'd be an awesome app to improve the karaoke experience, but somehow, people are surviving and managing to enjoy singing drunkenly without it.

Well, perhaps traditional karaoke doesn't need an app, but "Powerpoint Karaoke" definitely does. For the unfamiliar, Powerpoint Karaoke is yet another new style of presentation, and it basically involves having the speaker speak to a set of slides that he's never seen before, and that were originally delivered by someone else. It serves partially to make fun of the sometimes incomprehensible nature of slides, partially to just be damn funny as the surprised speaker deals with the surprise slides, and partially to improve the speaker's presentation skills. If you can present an arbitrary slideset and woo the audience, then you can present nearly anything.

For our most recent set of internal lightning talks, I gave speakers the option of Powerpoint Karaoke instead of their own slideset, and one brave speaker (Jez) took on the challenge. And I, being the dumb-ass who suggested a format I have never done nor seen, was tasked with picking the slideset for him to present. Yes, I could have gone the route of randomly finding a slideset on the web as soon as Jez got on the stage, but considering that this was the first and only Powerpoint Karaoke that the office would witness that day, I wanted to make sure it went well. These were the approximate requirements I had in my head:

  • The slideset should fit well into our lightning talk 5 minute format, with about 15-30 slides. Ideally, I could have the slides auto-advance, to challenge the speaker's reflexes further.
  • The slideset should have a title and picture(s) on each slide, but not much else. If there was too much text, the speaker could just read that off and know the original intention of the slide. If there wasn't a title on any slide, then the speaker would have too much freedom in what he said.
  • The slideset shouldn't offend anyone in terms of politics/religions/beliefs - I didn't want to pick a slideset with the clear intention to make fun of some group. It might offend, and that's too easy, anyway.

So, using Slideshare's search and browse features, my search began. I looked at the top slides in various categories for all time, I did searches for random topics, I looked for slides from speakers that I knew did the kind of slides I was looking for. After an hour of searching, I had nothing. Most slidesets were the wrong length, the wrong amount of textual content, or just plain spam (slidesets with 1 image advertising a product). I then asked the office for suggestions of topics*, and used those random suggestions to seed my searches. After another half hour, I found 2 possibilities: "How to Defend Against Zombies" and "How I learned to love ballet". I picked the second one, as it had exactly 20 slides (Ignite format!), referenced the speaker's wife (Jez's wife works in the office too), and I thought it would be more humorous if Jez could turn ballet funny than if he could turn zombie invasions funny (since, well, they already are).

How did it go? Jez took that slideset and freaking ran with it - he did more with it than I could have imagined. His best line was when he got to the slide about "Barre" (a position) and explained that was the Japanese pronunciation for "Ballet".

But, I've learned now that it is much harder to find karaoke-able slidesets than karaoke-able songs, and it makes sense when you realize that the majority of songs are there are similar length & lyrics density, whereas the slidesets in the world are a highly diverse species. So, I would like an app for finding suitable slidesets, or even better, I would like Slideshare to add the following features to their search: restrict by # of slides (a range?), restrict by text density (avg words per slide?). Then, everyone in the world can get on the Powerpoint Karaoke bandwagon, and it can be the next great (geek) thing.


*And for your amusement pleasure, here are the list of topics that my colleagues proposed. I asked for these when I was contemplating actually creating the slideset myself with random images, but realized that isn't really the point of Karaoke, and used the ideas as keywords for slideshare searches. That could become an entirely new style of presenting though, especially if someone creates an app that can generate slidesets from just the titles - complete with images, graphs, and venn diagrams, of course. :)

  • "Lightning bugs"
  • "Japanese pop culture from the 1980's"
  • "Sesame Street and Alcoholism"
  • "100 Things to do with Dwarves and Leather"
  • "How to prepare for a zombie invasion"
  • "Analytic and algebraic topology of locally Euclidean parametrization of infinitely differentiable Riemannian manifolds"
  • "How Google could make billions through the power of hypnosis"
  • "Early 20th century ballet and swimming pool repair"
  • "Properly Disambiguated Java Generics"
  • "How to bluff and give an insightful talk when you have to make it up on the spot with no prior warning"
  • "Latin American Hedge Funds and You"
  • "The Social and Economic History of the Grapefruit"
  • "Care and Feeding of the Dugong"
  • "Unicorns"
  • "England Won the War of Independence - a counterfactual world history"
  • "Keelhauling, Walking the plank, and other piratical forms of corporal punishment"
  • "Why Robot Ninja Monkey Zombies are AWESOME"
  • "A complete etymology of My Little Pony"
  • "Zonkies: discuss"
  • "The renovations that Start City *should* be making"
  • "A 5-minute tutorial in interpretive dance"