Day 4: First Preliminary Alpha Test Etc. Screencast
Jon Pipitone and I have created a screencast showing the basics (and I really mean “the basics”) of pulling data out of a database using SQL. It took about 4.5 hours in total to produce 7 minutes of video:
| Writing script: | 1.5 hours |
| Doing 3 takes: | 1.0 hours |
| Editing: | 1.0 hours |
| Transcribing: | 0.5 hours |
| Screenshots: | 0.5 hours |
This was the first time either of us had used Camtasia; I expect that we’ll be able to cut the recording, editing, and screenshotting time in half with practice. Transcribing time is dictated by typing and editing speed, so it’s unlikely to come down, and scripting time will probably be the same or greater for future topics (since some of them will require diagrams and other time-consuming prep). We’re therefore looking at 3 hours of production for 5-7 minutes of screen time; if you figure a lecture is 50 minutes, and there are 25 of them in the course, that works out to… um… carry the six… 750 hours to get all of the existing material online. Pad by half to account for real-world effects, and that’s 35-40 weeks — tight, but doable.
What we need now from you, dear reader, is feedback. Is the pace right? Do you want larger chunks, or smaller ones? Is the transcript useful? (I know we have to re-do the screenshots to make them more readable — we’re thinking about exactly how to do that.) What else could/should we change, and why?
We look forward to your comments.
I think you should use:
<lvideo width="924" height="645" controls> <source src="http://dev.software-carpentry.org/db-01-a1/media/swc4db01a924.mp4" type='video/mp4; codecs="avc1.42E01E, mp4a.40.2"'> <source src="http://dev.software-carpentry.org/db-01-a1/media/swc4db01a924.ogv" type='video/ogg; codecs="theora, vorbis"'> </video>instead of/in addition to the flash version.
(using http://diveintohtml5.org/video.html#firefogg to generate the ogg version, natch.)
I do like the transcript, but wonder if you could add closed-captions to the video itself. (I saw a video about this recently, but can’t find the link. Hopefully Mr. Bolter can help you out.)
Later,
Blake.
I think it might have been useful to show what would happen if you didn’t add the parentheses, and explain why it did that, because that’s a mistake I could see people making, and getting confused by.
Later,
Blake.
Is the pace right? — The pace worked really well for me, but I know how to do all of this already so that could be an indication that it’s a little fast for folks to get everything the first time. That said, especially with small chunks like this I don’t think that it is necessary to get everything the first time through. Users can watch, try, and then either watch again or check out the transcripts if they can’t remember something important. So, I’d say it looks good (when you put up something a little more outside of my comfort zone I’ll let you know if this still holds).
Do you want larger chunks, or smaller ones? — I really like this chunk size. It often feels fairly easy to squeeze a handful of short things in during a busy day (it can actually feel pretty rewarding), whereas one longer thing can feel to daunting to start (or be usurped by some “more important” longer thing). Since the goal is to get already busy scientists to spend time taking this course I think that chunks of about this length will reduce the activation energy and will increase the number of people getting deeper in the course. In addition, one of the things that I’ve found difficult about dealing with 15+ minute screencasts is that I inevitably want to go back and see something again that is somewhere in the middle. This means that I have to go searching around in the video to find it if I don’t want to spend a long time waiting to get there. This size chunk is much more manageable in that regard.
Is the transcript useful? — Definitely! After watching a screencast I often want to remember how to do a particular thing that I remember seeing. This can be a hassle because first you have to find the right part of the video and then often you need to pause, do the first piece, play, pause, do the second piece, and so on. The text from the transcript is a much more useful format for this second round of information delivery.
I liked it. A few comments:
* I agree with Blake on the parenthesis issue –though I guess it’s a pain to re-record just for this bit, and it’s not that necessary.
* The transcript is good, but feels a bit clunky: big pieces of text just lying there. The text works in the screencast, but doesn’t make for such a good read.
* The pace may be a bit too fast for newbies
* To make the transcript better, perhaps you could annotate it with extra stuff (say, for instance, the parenthesis question, or the code snippets you use). In my monitor, I see relatively large images on the right, and text (and a lot of empty space) on the left –this empty space could be used, I think.
I think that:
* the screencast is definitely a good vehicle for showing how to manipulate a computer program.
* There is a lot of text and rapidly scrolling down it’s hard to randomly seek some key ideas when a lot of the screens look the same
* perhaps a list of titles/”key ideas” with timestamps where you start talking about them?
* really cool might be to click on the title and the screencast jumps to that point, removing the need for screenshots, leaving only the transcript sliced up by titles.
* I guess I’m biased against the scrolling down, so (returning again) it would be nice to have a table of content to tab through showing a screenshot on the left and section transcript on the right
but this is improvements on something that is good enough?
I’m a newbie to SQL, but I thought the pace was about right, it could even be sped up a little. I think it would be useful if viewers are given references to documentations on SQL.
It’s great. Keep it up. Good chunk size, good pace.