Update on WatchKnow Reader progress (April)

We are hip-deep in the initial building phase of the project.  We’ve got a demo going (nothing suitable for public consumption yet though) and made a lot of adjustments to the plan.  The whole plan of making an elaborate, complete demo turned out to be well justified; basically, it has turned out to be very important to make one “perfect” presentation before making the code that will allow me to replicate this forty times over.

So here’s the deal.  There will be at least eight slides per “cluster”:

1. The word, written large (in various fonts), sounded out very slowly, and displayed karaoke style (in various colors).
2. Sounded out faster.
3. Blended, but slowly.
4. Blended at normal speed.
5. A high-quality, illustrative photograph (with the blended word as voiceover).
6. A sentence including the word, which usually either defines the word or gives some basic fact about it.  Again, displayed karaoke style.
7. A brief video (with the sentence as voiceover).
8. Just the word, blended at normal speed again.

For each set of words and media, there will in fact be seven different presentations, based on selections of these slides: Sound It Out Slowly (which shows all the slides); Sound It Out Quickly (which shows the quick sound-out and then the word); Let Me Sound It Out; and then Audio Flashcards, Silent Flashcards (the words are not read to the user), Audio Sentences (illustrative sentences only, no words), and Silent Sentences.

I had something of a brainstorm a little while ago, which has meant a whole bunch of rethinking of stuff: if we’re going to have videos of someone reading the words, why not show the videos on the same slide that the word occurs?  And why not make the audio track of the video the voiceover for the word?  Well, we’ve tried it out, and it works great. I think it’s going to be especially great for babies and toddlers, who learn much more when they can see a face saying the word, with lips moving.

Another feature I’m adding is, for some words, multimedia dictionary entries: when you click on some words in the sentence slides, the sounding-out will be done on a pop-up, which will include the other content from a cluster (illustrative picture, sentence/definition, and maybe video).  It will be written so that, even if I have no visual media, I’ll still be able to produce short, kid-friendly definitions.  H. loves clicking on words in our ebook reader programs, to look them up in the dictionary; I think making a kid-friendly version of that feature, with pix, is just what is needed to teach vocabulary.  We’re also adding various handy options.

We’ve already done a lot of work, but we have a huge amount of work in front of us.  Once a single cluster is finally settled & debugged, we need to make it into a whole, perfect presentation.  And to do that, well, we have to write an authoring tool.  I have been working on requirements for that during the last couple of days.

Sometimes friends and family ask me why all this time and expense is necessary–why not use PowerPoint?  There are a couple reasons.  First, of course, not everyone has PowerPoint.  But the main reason is that we have to start from scratch if we want to have the main features, the “killer” features, that will make Reader potentially the best system  for teaching reading online: sounding out words while displaying letters (or groups of letters) individually, integrating that with video, including pronunciations and dictionary, and quickly and easily making new presentations that are all interlinked wiki-style.  In the end this will be basically an authoring platform for making a children’s multimedia encyclopedia–something I’ve wanted to make since I was working on Wikipedia.  We’re starting with a reading tutorial, but when we’re done with it, I’ll be able to jump pretty seamlessly into putting my old powerpoint presentations (like this one) on this platform, interlink them, spruce them up with dictionary definitions, add automatic quizzes…it should be pretty cool.