2004¦~11¤ë15¤é


Information theory and instructional design Information theory and instructional design

Submitted by david. on 2004-11-12 04:05 PM.

I’ve been reading this absolutely great book called "It Must Be Beautiful," which describes the history and meaning of the great equations of modern science. Something really struck me while reading the chapter on information theory by Igor Aleksander. I’m going to re-expose myself as a complete idiot here, but it seems to me there are some very useful things for instructional designers to think about...
(AVS should love this one... he hates my lame attempts at making interdisciplinary connections more than anything else I write.)

In discussing Shannon’s equation I = -p log2 p, Aleksander explains on p214 "the amount of information I depends on the surprise the message holds. This is because the mathematical way of expressing surprise is as a probability p; the less probable an event is, the more surprising it is and the more information it conveys."

This seems to be related to prior knowledge somehow. If a person has significant prior knowledge in an area, it is generally takes less instruction to help them learn a certain "thing." For example, if I already know algebra and trigonometry, it should take relatively little instruction to teach me the calculus. If I know nothing about algebra or trig, it should take relatively longer to teach me the calculus. I think there’s a relationship between this notion of "surprise" and what instructional designers call "prior knowledge," in the sense that calculus would be a huge "surprise" to me if I didn’t know algebra or trig. Of all the studies of individual differences, after all, the only one that seems to show stable interaction with instruction is prior knowledge.

You may be saying "Dave, we already know about prior knowledge and its relation to learning. You just made reference to some of the research! If this doesn’t tell us anything we didn’t know, then why bother being interested by it?" I like this notion because it might point toward a formal way of thinking about the importance of prior knowledge in learning without forcing us to talk about schema and other totally contrived descriptions of how the brain functions. I mean, this seems to provide an explanation of the phenomenon on its own terms, purely in terms of information instead of in terms of brains.

Is there someone applying Shannon's work to instructional design or learning sciences? Are these connections completely misguided, mildly interesting, or something in between?

5:31:15 PM    

New hardware video compression. New hardware video compression: Robert Cringely describes proven hardware today that can encode TV-quality signals at 384 kbps (internet-like) transfer rates. It's not mass-produced yet, but this seems inevitable. Compare this to what we see with puny webcams today, and... [JD on MX]
5:23:46 PM    

Phone UI complexity. Phone UI complexity: At Mobitopia, Frank Koehntopp rips a good rant on how mobile phones may have already reached the level of front-end complexity of a computer, which won't quite work for a universal device. What to do? I suspect... [JD on MX]
5:22:16 PM    

VA Software adding Wiki support to collaborative tool. VA Software in March 2005 plans to add Wiki integration to its SourceForge Enterprise Edition environment for collaborative application development, with the goal of providing more structure to Wikis. [InfoWorld: Top News]
5:15:31 PM    

A postmodern trial. The media became the evidence that helped find Scott Peterson guilty of murder. [Christian Science Monitor | Top Stories]
5:13:55 PM    

Is Google the new devil?. Looks like Microsoft may have been biding its time to get back at search giant Google. [Missing Links] [CNET News.com]
5:02:01 PM    

Can Microsoft buy RealNetworks' silence?. Companies mum on settlement speculation as the software giant's antitrust appeal proceeds in Europe. [CNET News.com]
5:01:20 PM    

Software2005: Open and Loosely Joined. Applying logic might suggest to us that there's zero-profit likely to be made in free software. On the surface, the open-source movement appears to defy basic economic principles. But, perhaps that flawed notion is where much of the confusion about this evolving commercial ecosystem might originate. As an example, using the source code might not require a purchase, but the creation and ongoing development certainly wasn't free of... [AlwaysOn Network]
1:26:04 PM    

Why Use Computer Games For Learning?. The debate has long been raging, with the traditionalists having always a highly skeptical judgment on such an educational approach, and only a few brave open-minded scholars taking the gaming matter with courage to the class. Alice Mitchell and Carol...... [Learning - Educational Technologies :: Robin Good's Latest News]
11:14:30 AM    

TechLearn.

This Tuesday, I'm giving a presentation at TechLearn. This is six years to the day since I described a model for eLearning at TechLearn 1998 at Disneyworld. I talked about a scenario planning exercise I'd just completed and invited people for a ride in the Internet Time Machine, which was modeled on the Back to the Future DeLorean. We were looking way out into the future...all the way to 2004.


These were brand-new ideas at the time. No vendor talked about eLearning until nearly a year later.


On Tuesday, I plan to give an update. The Internet Time Machine has gotten a lot cooler. And we've come a long way in six years. In retrospect, our boundless enthusiasm back then seems naive. Now we've got real business to do. I'm still incredibly optimistic. We are going to change the world. But it's going to be a bit more complicated than we thought back in the last century.


I am concerned about the worker of the future. We have the potential to create wonderful, challenging, inspiring jobs. Or to take a small-minded, short-term, demeaning approach.

Imagine what this guy would do with the ability to monitor work right down to the keystroke. Networks work both ways. Surveillance of workflow creates the oppportunity for surveillance of workers. Join me if you'd like to discuss the Dark Side as well as the upside of the next phase of eLearning. It's a thorny issue, for the very nature of work is changing.



Of course, it's hard to imagine TechLearn in New York instead of at the Coronado in Disneyworld.

While others were paying steep rates at the Coronado ("Have a magical morning!"), I've booked a lot of time in $25/night motels outside the Disney gates in Kissimmee. I've seen gators in Gatorland leap out of the water to grab chickens, bought crap at the big swap meets, piloted an airboat through the swamps, rode a jet-ski, walked the streets of Centennial (Disney's Stepford community), shopped at ersatz factory outlets, and eaten supper in all-you-can-eat honky-tonks, surrounded by over-amped, Disneyfied children and parents whose exhaustion was palpable. What a weird place, Florida.


Moi, riding gator. (Dead gator).


The change of venue is nothing, however, compared to the impending change of maestro. Elliott's contract with Advanstar expires soon. Everyone expects him to arise again, taking the Consortium with him, and leaving Advanstar with little to show for the millions it spent for the franchise. ("Those who do not learn the lessons of history...".) It's as if they bought the barber shop but failed to see that without the barber, all you get is an empty room.

For me, TechLearn has always been Elliott and his entourage. Images pop into my head. Elliott and Cathy personally greeting everyone as they enter Tomorrowland or Main Street for the party. Stan rushing around behind the scenes making things work. Cathy's indefatigable family, the DeMicelis, AnneMarie and the dad and Matt... Elliott's mom, when she was still with us. Jen. Others I can see in my mind's eye but I'm blanking on their names. And the "regulars" like Beth Thomas, Diane Hessan & the SoundBytes, and Wayne Hodgins. These folks create community. In the early days, TechLearn was closer in spirit to Woodstock than to, say, an ASTD conference; we're all in this together, man.


The SoundBytes


The rolling schwag bag


From the stage during Lance Dublin's and my presentation in 2002.
We had an overflow crowd. The program mistakenly said EIliott would be speaking.


I was delighted to run into one of the
Raspini brothers in my local supermarket!


And you thought they were just buttons.
No, this is another way Elliott gets people
to talk with one another.

TechLearn has been a fun ride. I'm a marketing guy and designer at heart. Elliott is a marketing guy, too. He senses what people need and he delivers it. Elliott is the most savvy, effective, natural-born marketer I've even seen. Aside from that, he's a true mensch, with a big heart. I have learned so much at TechLearn I can't begin to describe it.

Elliott and I haven't talked in a year, so I'm unaware of his plans. I'm sure they'll be larger than life, and I hope I can take part. Elliott and Cathy, thanks for a great ride.
































[Internet Time Blog]
10:03:01 AM    

Chinese/Japanese dictionary with "native" MP3 pronunciations.

Although next to completely useless for me, Seiko Instruments has announced a new Chinese/Japanese dictionary that features actual recorded audio data for pronunciation purposes. Moreoever, the audio data has been compressed using MP3.

If you've ever used an electronic dictionary with some sort of audio feature, chances are, it used some sort of voice synthesis. Voice synthesis obviously isn't as accurate as a native speaker, therefore this is clearly better. I await a Japanese/English version of the same thing.

[TechJapan]
8:39:45 AM    

Gizmodo VR Blogging

rough_day.jpg imageA lot of you have asked to see the sort of working environment we have here at Gizmodo so I thought I'd share. That's me with the dank 'stache and my body-double on the right. I've got to head into the city later so I thought I'd grow a dupe while I stay here and brain up some more stories. (Thanks, Stx!)

- lev (joeljohnson@gmail.com) [Gizmodo]
8:36:29 AM    

Churchill Club, Top 10 Tech Trends

Last year this was a great event. They discussed 12 trends. By my count, they went 5 for 12, but had some direct hits that were not obvious last year. This year was less interesting, and very poorly produced.

But anyway, here are the trends with my thoughts as well. Keep in mind these are supposed to come true in the next 2-3 years.

1. Doerr – NextWeb. See my post on Web 2.0 conference below. General agreement among the panelists. Doerr singled our Blogging and RSS twice.
Of course I agree, I am betting my career here.

2. Dyson - Personal Electronic Health Records emerge and create opportunities for data sharing, protection and search. Panel was skeptical.
This happening in 2-3 years is laughable. When the boomers retire and taxpayers are hit with the triple whammy (shrinking number of taxpayers, increased number of retires, and escalating healthcare costs) maybe there will be enough force to break through the inertia. We are still 15-30 years away from this happening.

3. McNamee – There will be no major waves of enterprise technology spending equivalent to Windows, Y2K or ERP for 5 years. He elaborated to say that the next big trend will be Web Services, but today we do not even understand the processes we need to automate. Thus we are in a DIY period for IT. Investors in public software companies will suffer. Panel disagreed. Doerr – web services are happing today. Schoendorf – this is a bullish sign, next 2-3 years will be a golden age.
Disagreeing with McNamee is perilous, but I will. We are through the dry spell. Companies are looking to IT to drive competitive advantage and IT is looking to vendors, particularly start-ups. In our corner of the world, we see revenue being generated and processes being automated with RSS everyday.

4. Schoendorf – China is the next global innovator. China is the first big threat to the quality of life of Silicon Valley. The panel disagreed mostly due to timing reasons.
I agree strongly. I particularly like the second sentence. Currently the Valley leads the world in entrepreneurship. As a result, many people of modest talent and work ethic enjoy a fabulous quality of life. As China asserts its scale and people advantages, this will change. To stay on top, leading companies must work with Chinese nationals, not just to leverage low cost labor, but to do R&D, and to open a new market. Else our children will inherit a world with the majority of the global 2000 companies being Chinese companies.

5. Perkins – Mainstream media and entertainment will relent to the “Open Source Media Revolution” leading to a mini-boom for content creators, blogging and social networking tools and application developers. He elaborates to say that any media company that does not open up to sharing, collaboration, and transparency will not have consumers trust and will parish. Panel felt this happened last year.
I agree, but the boom will be in attention paid, not in revenue or profits.

6. Doerr – Stem Cells will be a medical revolution and California will lead it. Panel agreed, but felt it will take more than 2-3 years.
I don’t follow this market, but I hope he is right.

7. Dyson – Cell Phone text messaging will boom in America and be used for many new applications including personalized marketing and drug compliance. Panel agreed.
I think SMS text messaging is still too hard. When Treo 600 type functionality is bundled with service, mobile applications will explode.

8. McNamee – Consumer Technology and Content that targets people over 30 will be more successful than products that target younger people. He sites demographic trends, average iTunes users being over 30, Finding Nemo being the number one selling DVD and the fact that those under 30 do not buy but share/steal. Panel disagreed. Doerr did a scholarly job of refuting and Schoendorf noted that globally this is ridiculous.
I disagree that it will be more successful, but it is worth noting that there are lots of tech savvy folks over 30 for the first time in human history.

9. Schoendorf – Digital Living – Throw out your TV, DVDs, CDs and Stereo because it is all obsolete. From here is will all be stored in one place and transmitted wirelessly. Panel was mixed. Doerr thinks no one would trust this job to Windows. Dyson felt sharing will drive this trend.
I agree, but sadly, my wife doesn’t.

10. Perkins – A cultural move to IT as a utility in computing will keep the IT business growing. Panel agreed.
Obviously the panel has never sold a service to IT. This is a complex and long term trend that has more to do with the IT department’s employee incentive system, than to do with technology.

Last year they had audience participation at the end. I wrote down “RSS revolutionizes communication,” They did not read it. But it’s still my top pick. As Dave Winer says, if you don’t have some apprehension about pushing the publish button, you are not doing a good job of blogging. Time will tell who is right.


8:35:22 AM    

Knowledge Transportation.
A very good metaphor for KM. Taking the metaphor another step, some people view the transportation infrastructure as a mean of transporting "knowledge packages" and other as allowing a learner to travel from one point to another. After all, human can learn from first person AND third person experience. In some cases, simple instructions can get work done just-in-time. Get the "knowledge
By noemail@noemail.org (Albert Ip). [Random Walk in E-Learning]
8:26:21 AM    

e-Learning: In Search of a Better Definition.
From Big Dog, Little Dog by Donald Clark there are two parts to learning, just as there are in performance. Gilbert said that performance has two aspects: behavior being the means and its consequence being the end (1998). Learning is similar in that it also has two aspects: training or teaching being the means and its consequence being the end. Now the training or teaching may be self-taught,
By noemail@noemail.org (Albert Ip). [Random Walk in E-Learning]
8:24:50 AM