Having learned a trick, old dogs want to do repeat performance. Back when I had a Fujitsu Tablet PC, I made a variety of screen capture movies of my writing a la delivering a microeconomics lecture, so my voice provides annotation for what appears on the screen. I posted some of those to Google Video. When that closed those videos were migrated to YouTube. So they are still available, but the date of creation has been lost. They probably date back to 2007 or 2008, such as this one on The Envelope Theorem. If I recall correctly, this is a capture of writing in Windows Journal. It was done in a 4:3 aspect ratio, which is why there are the black bars on each side. So it appears dated. The handwriting is, at best, so so. Yet the thing does what it is supposed to do. I'm surprised that it still gets some use. Evidently, there is an audience for this sort of thing.
I've wanted to do something similar for the class I'm teaching this fall, but I no longer have a Tablet PC. (This is one area where work beats retirement. When working I got all the equipment I wanted. Now I have to pay out of pocket for this stuff and ask whether the purchase is justified.) I'm writing this post on my home computer, an all in one Sony Vaio. It's three years old but works reasonably well, so no complaints there. I also have an original iPad. I wanted to produce these micro-lectures using one or both of those. What follows is a sequel to the post No Chalk Dust, No Smudges from last week. The lecture notes I produced there, using a cheap stylus from Amazon.com on the iPad and a free app that Leslie Hammersmith recommended called Note Anytime, were sufficiently good as to raise my hopes about producing a micro-lecture in much the same manner.
However, making a screen capture movie on a PC, say with SnagIt, is different from doing it on the iPad. On the PC there is one application running for the capture and some other application which has the content of what is being captured (Windows Journal in those older videos I made). On my original iPad you need the same app for both content and the capture. TechSmith, makers of SnagIt, have such an app called ScreenChomp. It sounds promising and is what I used. But there is also a different issue that I couldn't resolve well.
The Tablet PCs are "active matrix" meaning they pick up some signal from the stylus so know where to place the digital ink. Other objects on the screen, such as the heel of your hand that holds the stylus, have no impact on what is produced. The Fujitsu I had was well designed this way. In contrast, my iPad and my Vaio, both with touchscreens, are "passive matrix" and thus pick up input from anything that touches the screen. This makes writing with a stylus on the iPad something of a challenge, because if you're like me you want the heel of your hand resting on something solid while you write. You don't want it hovering above. (For those who can write well with their hands not touching the screen at all while holding the stylus, good for you. The rest of this post won't be interesting to you.) I should add that using the index finger instead of a stylus is even worse regarding the quality of what is produced.
Note Anytime has a nifty solution to this issue. You write in a little box at the bottom of the screen, with your hand on the table or edge of the iPad, not touching the screen. That little box is a recreation of some segment of the page, which is also shown. And you can move the segment around either via controls that are provided or by dragging it with your finger. So you can fill the page with writing but input only at the bottom of the screen. For note taking, that works reasonably well.
ScreenChomp doesn't have those features. It also doesn't allow you to scroll the page. It does allow you to erase the page and start again with a clean screen. That is what I did, borrowing the technique of writing on the bottom of the screen that I used in making the lecture notes. (There is no voice over in this video, only writing. I found it difficult to produce the writing so if I were going to use this I'd make another capture of the writing already done, inserting my voice at that time.) But I didn't like the result. No context is built this way, showing only one line at a time with nothing previous shown. Further, because the pen tools in ScreenChomp are a little blunt, I felt I had less control of the stylus than I did with Note Anytime. The upshot is that I wasn't satisfied with the result and opted to try something else.
It occurred to me that I could animate a PowerPoint presentation on a character by character basis and thereby simulate handwriting in the process (really, it would simulate typing on the screen). This would make all characters well formed and easy to read, a distinct plus. You can judge for yourself here (again, there is no audio). But there are a variety of issues that arose doing this.
The lecture notes transferred into PowerPoint took six slides. The average number of characters per slide was close to 190. I made the animation by making multiple copies of the same slide and then having the subsequent slide have the next character show. (Actually, what I did was delete a character in the previous slide, as that proved easier to do.) So there ended up being 190 copies of the same slide, but with different characters being visible. I then put in the timings using the record Narration. My aim was for about one minute on the entire page, so that meant a little more than three characters per second. I practiced pushing the space bar while using a stop watch to get the desired speed. I then discovered that if you already have a timing for a slide and you make a duplicate of the slide then the copy inherits the same timing.
Not having done this before, I had no sense for how long it would take to do. It took quite a while, which was the main problem. It was mind numbing work, and sometimes I would get the slides out of order. I did find that by saying my procedure aloud (duplicate the slide, go back one slide, delete the particular character) I made fewer mistakes and it went faster. But it was still laborious. In the middle of the work I realized I made the slides 4:3 instead of 16:9. But changing the aspect ratio ruined the slides. I'd have to start from scratch. Then I found that I had made an error in the math on the last slide. But with all the copies I made the same error propagated to perhaps 100 slides or so. To correct it I had to make changes on each of them. My point with all this detail is to show that it is not a very robust way of doing things.
After I had finished this I recorded audio in Audacity for the voice over. Then I moved to combine the two. My first thought was to try Slideshare.net and make a slidecast. But that didn't work well at all. Instead of one slide following another it had slide, then black screen, then slide, then black screen, etc. I can only guess why. Having multiple slides per second is not the norm and it didn't like it.
So I thought to put the audio directly into PowerPoint with the intent of then recording the slideshow. In the process of doing that, I learned that PowerPoint has an upper limit on the number of slides moving forward from a given slide that the audio would play. That limit is 999. I had over 1100 slides. So if I were to do this approach, I'd need to split the audio into two pieces and insert the second piece somewhere in the middle of the PowerPoint presentation. That was unattractive to me.
The straw that broke the camel's back was this and I really could have anticipated it ahead of time, but I didn't. In the ideal, since it is the verbal explanation that adds value over what is already in the lecture notes, the pace of what appears on the screen should follow the verbal explanation, not vice versa. On some things that are hard you want to linger a bit with the verbal explanation. On other things that are transparent you can zip right through. With the slide transitions already put in advance you can't do that. And you feel under pressure providing the narration, either to keep up with what is going too fast on the screen or to not get too far ahead. It's that which I was referring to in the post title.
So what I ended up with, which is definitely more sensible as an approach but abandons the goal of simulating the handwriting, is to animate on a line by line basis. The same text stays on the screen for longer that way and it is not that hard to put in the voice over and advance the slides when appropriate. Whether this is better or worse than the type of movies I was making on the Tablet PC, I can't say. That is for others to judge. But I do know that I can produce these things in reasonable time and get something that is usable.
Let me close with one other point. I don't think it is a good idea to do ordinary slideshow mode in PowerPoint for doing the screen capture. The SnagIt controls end up in the system tray that way and it is somewhat difficult to find them afterward. Further, you get no feedback on how long you've been talking during the presentation. In what I recorded above I had PowerPoint in the normal editing view and advanced the slides using the scroll wheel on my mouse. That worked reasonably well, though the arrow turns into a cursor when it hovers over text and once or twice I wheeled through several slides at one time and had to go back. I did discover afterward that you can be in slideshow mode in a window, which might be the best way to do this. Go to the Slide Show menu in the Ribbon, select Set Up Slide Show and then there is a button on the left top for Browsed By and Individual (window). Select that, click the OK button and you're done.